Glioblastoma (GB) is the most common and most aggressive malignant brain tumor, characterized by extensive heterogeneity, heightened cell proliferation, cellular invasiveness, and angiogenesis. GBs are composed of many different cell populations, all of which interact in very complex ways with one another and with their tumor microenvironments. For this reason, among others, these tumors remain incurable and evade current standards of care, including maximal surgical resection, radiation, and chemotherapy. Our lab studies primarily three aspects of brain tumors:

• Intratumoral diversity and its impact on brain tumor therapy
• Molecular and cellular pathways of glioma cells
• Tumor immunotherapy

From Bohman et al., 2010.
FIGURE 3. Glioma growth model simulations with varied points of origin. Images at 4 time points each for 3 simulated lesions provided in the sagittal, coronal, and axial planes for lesion start points at the anterior dorsolateral subventricular zone, anterior deep white matter, and anterior superficial white matter. Green area reflects estimated T2-weighted image abnormality on magnetic resonance; red area reflects estimated T1-weighted image postgadolinium abnormality.

      Intratumoral diversity poses great challenges to brain tumor therapy. Malignant brain tumors, which are composed of myriad different cell populations, cannot be treated uniformly. It is thus essential for researchers to understand differences and similarities across cell populations in order to develop versatile therapies that can be used on a case-by-case basis, depending on the characteristics of a specific patient’s tumor.
Using The Cancer Genome Atlas (TCGA), which catalogues recurrent genomic abnormalities in GB, Verhaak et al. established four distinct subtypes of GB defined by genomic and clinical characteristics: Proneural, Neural, Classical, and Mesenchymal (Verhaak et al., 2010). These subtypes, however, only convey part of the story. They describe trends across tumors, rather than inherent attributes of tumors. For example, while Classical GBs are characterized by abnormally high levels of epidermal growth factor receptor (EGFR), a tumor cell population that has high levels of EGFR does not necessarily conform to the profile of the typical Classical GB. Additionally, what defines a tumor cell is not constrained to the profile of a single cell type: tumors are composed of earlier generations of malignant cells and later generations, or recruited cells, which have become malignant due to their interactions with other malignant cells (cite). Diversity within tumor subtypes and within the tumors themselves limits the success of treatments based on trends. Our lab strives to characterize diverse tumor populations and subpopulations to the degree that we can develop effective patient-specific therapies. In particular, the Bruce lab utilizes convection enhanced delivery (CED) to deliver chemotherapeutic agents directly to the site of brain tumors.
From Lopez et al., 2011. CED of topotecan causes loss of both tumor-initiating cells and recruited glial progenitors
A. Immunohistochemistry for GFP (green) and PDGFR-α (red) shows an increase in both the initially infected PDGF-expressing cells (GFP+) as well as the recruited progenitors (GFP-/PDGFR-α+) over 1, 4, and 7 days when animals are given CED of PBS (upper row). CED of topotecan causes a similarly time-dependent decrease in cells from both populations (lower row). Hoechst nuclear counterstain in blue. B. Increase in GFP+ and GFP-/PDGFR-α+ cells per HPF over time in PBS-treated tumors was statistically significant on one-way ANOVA and also positive for test of linear trend (p=0.0002 and p=0.0002, respectively). C. Decrease in GFP+ and GFP-/PDGFR-α+ cells per HPF over time in topotecan-treated tumors was statistically significant on one-way ANOVA and also positive for test of linear trend (p=0.0103 and p=0.0212, respectively). Fields were taken at 400× magnification (scale bar= 50 μm).
GB diversity is observed in all aspects of the tumor, at the molecular and cellular levels. While there are certain molecular and genetic mutations that recur within and across tumors, they are often contained to distinct subpopulations of cells and rarely in the entire tumor cell population. Our lab and other labs have identified expression patterns that are consistent across GB tumor populations, including overexpression of oncogenes epidermal growth factor receptor (EGFR) and platelet-derived growth factor receptor (PDGFR), highly deregulated signaling pathways, and mutations and deletions of tumor suppressor genes p53 and PTEN (Nakada et al., 2011). Knowing these patterns of expression, we have been able to research GB through identifying differences within and across tumor cell populations.
Figure 1. Retrovirus induced murine tumors show histological features of human GB. Histology of end stage tumors with different genetic alterations: Pten- mice (B and C) and Pten-/p53- mice (D and E). All tumors show salient features of GB, including areas of pseudopalisading necrosis.
Figure 2. Tumor cells express OPC genes. Triple immunofluorescence shows the expression of different markers: (G) ki67 (red), YFP+ (green) and PDGFRα+ (blue). (H) Olig2 (red), YFP (green) and GFAP (blue). (I) Olig2 (red), YFP (green) and NeuN (blue).
Our lab uses a PDGF-expressing retrovirus that produces brain tumors in mouse and rat models that closely resemble human GB, in particular the proneural subtype. The retrovirus, injected intracranially into the subcortical white matter, selectively infects resident populations of cycling glial progenitors and causes the rapid formation of de novo GB-like tumors. Unlike many animal models, ours does not rely on the use of xenografts, which do not evolve from resident cell populations. Our animal model recapitulates the prominent features of GB reliability and rapidly, allowing us to better develop treatments that are translatable to human GB.
In conjunction with our animal model, our lab studies GB through the use of human tissue samples. The Bartoli Brain Tumor Laboratory maintains an extensive repository of human GB specimens that we use to study aspects of human GB such as intra- and intertumoral heterogeneity, tumor population subtypes, and genetic backgrounds. Currently, studies investigate differences in cell populations within the tumor and tissue in the peritumoral space, to determine to what extent cells in different regions have different characteristics. Understanding differences and similarities across cell populations within the same patient will allow clinicians to modify their treatment regimens based on the patient’s unique situation. The Bartoli Brain Tumor Laboratory Tumor Bank serves not only as an invaluable resource in our development of patient-specific therapies, but also as a tool for our collaborators in various short- and long-term studies.
Figure 3. Gene expression in mouse tumors most resemble human Proneural GB. Classification heat map of RNA expression levels of mouse and human GB samples. Red to blue color scale indicates the range from the highest positive to highest negative correlation. The correlation was computed between the mouse samples and the TCGA classes of human GB.

The immune system exists to protect the body from illness due to foreign invaders (e.g. viruses, bacteria, fungi, parasites) as well as wayward cells of one’s own (e.g., cancer). Humans utilize a system of checks and balances to ensure that true threats are identified and an effective immune response is mounted, while “self” cells and non-threatening foreign objects do not generate a chronic, misguided response (e.g., allergies, rheumatoid arthritis, lupus, Crohn’s disease). The ability to tip this balance toward immunosuppression is a major factor in both chronic infectious diseases and cancer, by thwarting our natural anti-pathogen and anti-tumor responsiveness, respectively.

From Waziri et al., 2008. Flow cytometry of PBMC (upper panels) and tumor-infiltrating lymphocytes (lower panels) from patients with GB (A, D), meningioma (B, E). Lymphocytes were gated upon CD3 (x-axis) and CD56 (y-axis) CD3+CD56+ represent NKT cells, which are unusually abundant in GB tissue and in the periphery, compared to meningioma.

During the progression of GB, immunosuppression occurs on a number of levels which compromise the ability to mount an anti-tumor response — firstly, the tissues of the brain is an anti-inflammatory environment, which normally helps prevent autoimmunity against the nervous system; second, macrophages and microglia, cells which interact with T cells and other immune effectors to “instruct” activated or suppressive behavior, are associated with GB immunosuppression (Kennedy et al., 2009); finally, GB tissue becomes infiltrated with regulatory T cells (Treg), which inhibit the anti-tumor activity of activated T cells, and instead are populated by skewed populations of CD4+ and CD8+ T cells and normally scarce populations (e.g. NKT cells), which can be observed systemically as well (Waziri et al., 2008). Our current research focuses on how the amplification of immunosuppression, which allows tumors to persist, are linked to the development and complexity of the tumor cell population in our murine glioma model as well as in patient samples, and how effective anti-tumor responses might be restored through vaccination and targeted immunotherapy.
Glial Transformation and Gliomagenesis

Glioma Heterogeneity Project

Glioma Associate Seizures

Immunology of Brain Tumors

Experimental therapies (GB)

Glioblastoma (GB)