GLIOBLASTOM - Anni Hofmann Stiftung

The dual role of mesenchymal stem cells in gliomas: new therapies and therapeutic target.

Prof. Dr. Rainer Glaß, Leiter der Neurochirurgischen Forschung, Klinik für Neurochirurgie, Klinikum der Universität München, Marchioninistr. 15 81377 München. Email:  
Therapeutic resistance of malignant brain tumours (gliomas) is in large part mediated by the tumour-parenchyma. We and others showed that mesenchymal stem cells (MSCs) are a pathologically important parenchymal component of gliomas. How MSCs influence glioma is currently unclear, but our pilot-studies indicated that MSCs significantly modulate the course of pathology by altering tumour cell-death levels when gliomas are challenged by a stressful environment or propagated under steady-state conditions.  In particular, we observed that MSCs release anti-tumorigenic factors and tumour supporting molecules - depending on the environmental conditions. It is a central aim of our current study to uncover the molecular factors responsible for the pro- and anti-tumourigenic effects of MSCs in gliomas - as we can adapt the endogenous paracrine tumour suppressors for new glioma therapies and use the MSC-derived pro-tumourigenic factors as a target for new adjuvans therapies.

Currently we have gathered a large set of physiological data for our work-packages-1 and -2, which we aimed to address in the first year of our research schedule. As outlined in our funding application we have performed sensitive and reliable assays to quantify tumour-cell viability levels after stimulation of gliomas in vitro with MSC-conditioned media (MSC-CM) in a stressful environmental context. We used 5 primary human GBM cultures in 4 to 6 independent experiments (for each primary culture) and exposed these to our MSC-CM under the challenging environmental paradigm. As predicted by our pilot study (see aplication) we consistently found that MSCs promote glioma cell viability and that the MSC-derived tumour-supporting mechanism is an important therapeutic target for GBM. Furthermore, we could recapitulate these experiments using glioma cultures derived from transgenic mouse models (two lines in 3 to 4 experiments), which provide additional information on the physiological role of tumour-associated MSCs in GBM of distinct genetic subtypes. Strikingly, we found that the tumour promoting activity of MSCs is associated with individual GBM subtypes and is not a general event in all genetic classes of GBM. This is important information for the stratification of patients that can profit from MSC-directed adjuvans treatments. We have completed time-course experiments determining the extent of MSC-mediated glioma modulation over several days and simultaneously collected cell-material from these experiments for genome-wide expression analysis by microarray (n = 4). The cellular material has now passed all quality controls for genetic studies and will be hybridised with our array chips. In order to filter the background data that are normally obtained in large quantity in genome-wide arrays, we will compare the bioinformatic data with datasets gathered from GBM-lines that are non-responsive to MSCs (n = 2). Additionally, we have obtained a set of data showing that MSC-derived pro-tumourigenic factors are functional between species, thereby supporting the robustness of our observation (n = 4). In the next steps we will analyse our cell samples (used for the array-study) by additional techniques (e.g. real-time PCR) for gene expression changes for a range of candidate genes. Also, we will use chromatography to separate MSC conditioned media and to get a further insight into the molecular nature of the tumour-supporting factor(s) released from MSCs in gliomas. In our cooperative project with Prof. Christel Herold-Mende and Prof. Katrin Lamszus we have recently received valuable cell-material (relapsing GBM from CHM) to explore the therapeutic impact of MSC-directed adjuvans treatments for these patients. We will make our data (form stressed / non-stressed GBM) accessible to our cooperation partners to identify immune markers and metabolic pathways that are altered under this condition. Overall, we have performed all our experiments according to the time-line given in our application, are fully on schedule with the new experiments in our research-plan and have uncovered new, unforeseen and clinically relevant roles of MSCs in gliomas. In our view, the good progress of the projects fully merits the valuable support by the Anni-Hofmann trust.
Progress report for the project "Relevance of tumor metabolism for glioblastoma-host interaction"

The project aims to elucidate the functional relevance of the pentose phosphate pathway (PPP) and of glycolysis for the growth and invasion of glioblastoma cells, with the ultimate goal of identifying potential therapeutic targets. The project builds on previous work, in which we discovered that hypoxia induces a metabolic switch in glioblastoma stem-like cells with diminished activity of the PPP and decreased tumor cell proliferation, concomitant with increased glycolysis and enhanced cell migration. The work plan is divided into three parts.

In the first part we investigate whether the assocation between the PPP and proliferation as well as the association between glycolysis and cell migration also exist independently of changes in oxygen levels. We first developed and tested different techniques for enriching either highly proliferative or highly migratory glioblastoma cell subpopulations. Eventually we optimized a method that utilizes a fluorescent dye, which incorporates into cell membranes and dilutes with every cell division, to select for highly proliferative cells. A transwell assay was appropriated to select for highly motile cells. Expression analyses on populations enriched by these methods showed that highly proliferative cells exhibit decreased expression of glycolysis enzymes but increased expression of several PPP enzymes, whereas an inverse expression pattern is present in highly migratory cells, providing preliminary evidence for our hypothesis of an oxygen level-independet association between metabolic pathways and cellular function.

In the second part we investigate whether glycolysis and the PPP are directly causally relevant for glioblastoma cell migration or proliferation, respectively. We downregulated the expression of aldolase C (the most strongly hypoxia-induced enzyme of glycolysis) and of glucose 6-phosphate dehydrogenase (G6PD, the first and key regulatory enzyme of the PPP) using shRNA in a glioblastoma cell line. Our first analyses showed that downregulation of aldolase C causes reduces migration and increases proliferation, whereas downregulation of G6PD entails opposite effects. Additional analyses using chemical enzyme inhibitors (e.g. 6-aminonicotinamide) confirmed that inhibition of G6PD results in reduced proliferation but increased migration.
In the third part we assess whether the oxygen concentration-dependent switch between the PPP and glycolysis is a general phenomenon. As a first step, we extended our switch analyses to additional glioblastoma cell lines, which were not cultured under neural stem cell conditions (unlike the cell lines for which we had originally discovered the switch). Our first analyses show that the hypoxia-induced downregulation of PPP enyzmes and upregulation of glycolysis enzymes is also present in other glioblastoma cell lines.

In future work, the analyses need to be extended to other types of tumor cells (as well as to normal cell types in part 3). In addition, in vivo experiments are a next important step in part 2 of the project.

Immunological changes and adaptation of the tumor microenvironment during glioblastoma progression - state may 2014 - 

Our research project consists of three interacting parts. Aim of the first part is the characterization of the tumor microenvironment and cellular composition of primary and corresponding recurrent tumor tissues. To date, 34 pairs of primary and recurrent GBM (pGBM, rGBM) were analyzed regarding T cell infiltration by immunofluorescent staining. In some of the samples, there was a distinct increase of T cell infiltration in the recurrent tumor, especially of regulatory T cells. To complete our study sample, 17 additional pGBM/rGBM-pairs were prepared for the same analyses. Moreover, we aim to identify immune relevant pathways and genes by microarray analyses in the very same samples. The analyses of 12 pairs have been finished showing an altered gene expression in the recurrent tumor as compared to the primary tumor of the same patients. RNA samples of further 38 pairs were isolated to perform this type of analysis. Further, proteins of 32 pGBM/rGBM-pairs were isolated (of 40 intended) to analyze the chemokine and cytokine levels as compared to the observed infiltration of immune cells.

In the second part of the project, tumor cells and immune cells from eight pGBM and rGBM were isolated. These cell isolates were collected to finally investigate the effect of selected genes on the infiltration of immune cells by functional assays based on the first part of the project.

In the third part of the project we focus on the identification of changes in the repertoire of immunogenic antigens, where we use a combination of protein fractionation (PF2D) and T cell activation assays (IFN-γ ELISpot assay). To date, two primary and one recurrent GBM tissues were analyzed in detail. So far, we observed an increase of immunogenic tumor protein fractions especially in the recurrent tumor. In total, more than 1500 proteins of all single fractions were identified by mass spectrometry. An intensive filter process resulted in the selection of a couple of potential immunogenic targets in primary GBM as well as in recurrent GBM which are currently further validated. In addition, tumor tissues as well as sufficient numbers of blood cells corresponding to three further pGBM/rGBM-pairs were collected to complete these sophisticated analyses in a total of five pairs.