Tom Mikkelsen

SPARC: A Signal of Astrocytic Neoplastic Transformation and Reactive Response in Human Primary and Xenograft Gliomas

In an attempt to identify genetic alterations occurring early in astrocytoma progression, we performed subtractive hybridization between astrocytoma and glioblastoma cDNA libraries. We identified secreted protein acidic and rich in cysteine (SPARC), a protein implicated in cell-matrix interactions, as a gene overexpressed early in progression. Northern blot and immunohistochemical analyses indicated that transcript and protein were both elevated in all tumor specimens (grades II-IV) examined when compared with levels in normal brain. The level of SPARC expression was found to be tumor-dependent rather than grade-related. Immunohistochemically, SPARC protein was found to be overexpressed in 1) cells in the less cellularly dense regions within the tumor mass, 2) histomorphologically neoplastic-looking cells in adjacent normal brain at the tumor/brain interface, 3) neovessel endothelial cells in both the tumor and adjacent normal brain, and 4) reactive astrocytes in normal brain adjacent to tumor. Using a combination of DNA in situ hybridization and protein immunohistochemical analyses of the human/rat xenograft, SPARC expression was observed in the human glioma cells within the tumor mass, and in cells that invaded along vascular basement membranes and individually into the rat brain parenchyma, suggesting it may be an invasion-related gene. While it remains to be determined whether SPARC functionally contributes to tumor cell invasion, these data suggest that the early onset of increased SPARC expression, though complex, may serve as a signal indicative of neoplastic astrocytic transformation and reactive response to tumor-induced stress.

INCREASED SPARC EXPRESSION PROMOTES U87 GLIOBLASTOMA INVASION IN VITRO

Our recent studies have focused on identifying invasion‐promoting genes that are expressed early in brain tumor progression. We identified and characterized SPARC (secreted protein acidic and rich in cysteine) as a potential candidate. To determine whether increased SPARC expression functionally promotes brain tumor invasion, SPARC was transfected into U87MG glioblastoma cells using the tetracycline‐off gene expression system. The parental cell line (U87MG), the parental transactivator‐transfected clone (U87T2) and three selected U87T2–SPARC‐transfected clones (A2bi, A2b2 and C2a4) were characterized for endogenous and transfected SPARC expression. In comparison to the parental or U87T2 cell lines, the SPARC‐transfected clones demonstrated: (1) morphological changes, (2) increased SPARC transcript and protein abundances that were down‐regulated by the tetracycline analog doxycycline, (3) perinuclear localization of the transfected SPARC (consistent with reported localization of SPARC in normal cells in culture) and (4) altered adhesion and increased invasion as assessed by the spheroid confrontation assay. These data indicate that increased SPARC expression contributes to U87 glioblastoma tumor invasion in vitro and that these cell lines will serve as useful reagents to investigate the mechanism(s) by which SPARC promotes this phenotype in vitro and in vivo.

An Integrative Approach for In Silico Glioma Research

The integration of imaging and genomic data is critical to forming a better understanding of disease. Large public datasets, such as The Cancer Genome Atlas, present a unique opportunity to integrate these complementary data types for in silico scientific research. In this letter, we focus on the aspect of pathology image analysis and illustrate the challenges associated with analyzing and integrating large-scale image datasets with molecular characterizations. We present an example study of diffuse glioma brain tumors, where the morphometric analysis of 81 million nuclei is integrated with clinically relevant transcriptomic and genomic characterizations of glioblastoma tumors. The preliminary results demonstrate the potential of combining morphometric and molecular characterizations for in silico research.

Addition of MR imaging features and genetic biomarkers strengthens glioblastoma survival prediction in TCGA patients

PURPOSEnThe purpose of our study was to assess whether a model combining clinical factors, MR imaging features, and genomics would better predict overall survival of patients with glioblastoma (GBM) than either individual data type.nnnMETHODSnThe study was conducted leveraging The Cancer Genome Atlas (TCGA) effort supported by the National Institutes of Health. Six neuroradiologists reviewed MRI images from The Cancer Imaging Archive (http://cancerimagingarchive.net) of 102 GBM patients using the VASARI scoring system. The patients clinical and genetic data were obtained from the TCGA website (http://www.cancergenome.nih.gov/). Patient outcome was measured in terms of overall survival time. The association between different categories of biomarkers and survival was evaluated using Cox analysis.nnnRESULTSnThe features that were significantly associated with survival were: (1) clinical factors: chemotherapy; (2) imaging: proportion of tumor contrast enhancement on MRI; and (3) genomics: HRAS copy number variation. The combination of these three biomarkers resulted in an incremental increase in the strength of prediction of survival, with the model that included clinical, imaging, and genetic variables having the highest predictive accuracy (area under the curve 0.679±0.068, Akaikes information criterion 566.7, P<0.001).nnnCONCLUSIONnA combination of clinical factors, imaging features, and HRAS copy number variation best predicts survival of patients with GBM.

Temozolomide single-agent chemotherapy for newly diagnosed anaplastic oligodendroglioma

The treatment of patients with anaplastic oligodendroglioma (AO) has been significantly impacted by the molecular detection of loss of sequences on chromosomes 1p and 19q. We performed a clinical trial to prospectively evaluate the safety of treating patients with AO with temozolomide (TMZ) alone in patients with chromosome 1p/19q loss and with chemo-radiation in patients not harboring this loss. Forty-eight patients were enrolled, 36/48 (75%) with evidence of chromosome 1p/19q loss treated with TMZ alone and 12/18 (25%) without such losses, treated with pre-radiation TMZ followed by chemo-radiation. Despite more aggressive treatment, patients without 1p/19q loss had a shorter progression-free survival (PFS) of 13.5xa0months. With a median follow-up time of 32xa0months, patients with 1p/19q LOH had a median TTP of 28.7xa0months. Patients with AO with 1p/19q LOH can be safely treated with single-agent TMZ and do not appear to experience earlier or more frequent tumor progression. This treatment regimen should be studied as part of a formal randomized clinical trial.

Increased chemotactic migration and growth in heparanase-overexpressing human U251n glioma cells

Heparanase is an endoglycosidase that degrades heparan sulfate, the main polysaccharide constituent of the extracellular matrix (ECM) and basement membrane. Expression of the heparanase gene is associated with the invasion and metastatic potential of a variety of tumor-derived cell types. However, the roles of heparanase in the regulation of gene expression and the subsequent cell function changes other than invasion are not clear. In the current study, we overexpressed the human heparanase gene in a human U251n glioma cell line. We found that heparanase-overexpression significantly increased cell invasion, proliferation, anchorage-independent colony formation and chemotactic migration towards fetal bovine serum (FBS)-supplied medium and stromal cell-derived factor-1 (SDF-1). These phenotypic appearances were accompanied by enhanced protein kinase B (AKT) phosphorylation. Focal adhesion kinase (FAK) and extracellular signal-regulated kinase 1 (ERK1) signaling were not altered by heparanase-overexpression. These results indicate that heparanase has pleiotropic effects on tumor cells.

Development of a non-selecting, non-perturbing method to study human brain tumor cell invasion in murine brain

SummaryThe infiltrative nature of glial and some meningeal neoplasms is responsible for the failure of surgical removal and high recurrence rate of these tumors. Modeling of this processin vitro andin vivo will lead to a better understanding of the pathophysiology of this process and identify targets for novel therapy directed towards this phenotype. We present the results of the development and refinement of two model systems of tumor invasion: onein vitro barrier assay using the basement membrane extract Matrigel, and onein vivo where molecular detection of tumor cells allows single cell discrimination byin situ hybridization histochemistry. These techniques have strong correlations which validate their utility as measures of nervous system tumor invasion.

Tamoxifen increases photodynamic therapeutic response of U87 and U25ln human glioma cells.

We tested the hypothesis that Tamoxifen (TMX), an inhibitor of protein kinase C (PKC), augments the cytotoxicity of photodynamic therapy (PDT) treatment of human (U87) and (U25ln) glioma cells. U87 and U251n glioma cells were plated and treated with PDT using Photofrin as the sensitizer. Cells were treated with Photofrin at various doses and with various optical (632u2009nm) irradiation intensities 24u2009h later. Cells were also treated with Photofrin at a fixed dose alone and with various doses of Tamoxifen and subjected to laser treatment 24u2009h later. Tumor response was tested using the (3-94,5-dimethyl-2-yl)-2,5-diphenyl-tetrazolium (MTT) method. Total toxicity of U87 cells was achieved with PDT at all doses of Photofrin (1, 2.5, 5, 10u2009μg/ml) with irradiation densities equal to or greater than 200u2009mJ/cm2. Using an irradiation intensity of 100u2009mJ/cm2, U87 and U251n cells were killed in a Photofrin dose-dependent manner. Significant cytotoxicity was detected with Photofrin doses of 5u2009μg/ml (p < 0.05) and 10u2009μg/ml (p < 0.001). Tamoxifen at a dose of 500u2009μg/ml and higher, significantly increased the toxicity of the PDT response with 5u2009μg/ml Photofrin and 100u2009mJ/cm2 (p < 0.05). In summary, our data demonstrate that Tamoxifen significantly enhances the Photofrin PDT activity of U87 and U251n human glioma cells.

A quantitative model of tumor-induced angiogenesis in the nude mouse.

OBJECTIVE:Novel animal models allowing for the quantification of tumor-induced angiogenesis and cell migration may offer significant insight into the characterization and multidisciplinary treatment of brain tumors. In this study, we seek to establish such a model in tumor-bearing brain, allowing for a clear demarcation of primary and satellite tumor tissue in conjunction with precise quantification of cerebral microvasculature. METHODS:We used green fluorescent protein–transfected 9L-gliosarcoma cells stereotactically injected into the brain parenchyma of nude mice perfused with tetramethylrhodamine-dextran immediately before they were killed. New three-dimensional analytical software developed in our laboratory provided a quantitative analysis of laser-scanning confocal microscopy images of dextran-labeled cerebral microvessels. RESULTS:Our data confirm significant angiogenesis in tumor and brain adjacent to tumor. CONCLUSION:Because these highly infiltrative malignant brain tumors interdigitate with normal brain parenchyma through finger-like projections at the periphery of the solid tumor boundary, therapeutic options targeting tumor blood flow—combined with novel three-dimensional imaging to localize and track such interventions—may offer new hope for glioma management. To our knowledge, this system represents the first animal brain tumor model allowing for the precise colocalization and quantification of angiogenesis and tumor cell invasion, which may play an important role in the development of future therapy for brain tumors.

CHAPTER 14 – Tumor Invasiveness and Anti-invasion Strategies

Considerable progress has been made in identifying the molecules involved in promoting glioma migration and invasion. Molecules active in the process of glioma cell motility and invasion, therefore, may be viable targets for drug development. Molecular targets for the treatment of glioma invasion may consist of the stromal- or tumor-derived ECM molecules, adhesion molecules, matricellularproteins, proteases, and growth factors and their receptors. The ECM molecules, consisting of proteoglycans, glycosaminoglycans (GAG), and structural glycoproteins, can interact independently with the tumor cell-associated adhesion molecules, secreted matricellular proteins, and proteases. There are several classes of adhesion molecules that govern cell–cell and cell–ECM interactions, including the cadherins, the selectins, receptor protein tyrosine phosphatases (RPTPs), integrins, and galectin. The matricellular proteins include thrombospondins 1 and 2, Secreted Protein Acidic and Rich in Cysteine (SPARC/osteonectin/BM40), likely hevin/SC1 (a member of the SPARC family of proteins), Tenascins-C and -X, osteopontin, and the CCN1-6 family of proteins. Many antiproliferative and proapoptotic therapeutic strategies have been employed. Molecular and biological characterization of the gliomas has identified a number of molecules, or classes of molecules, that could be evaluated as therapeutic targets. Only a few have been examined to date, such as integrin antagonists, matricellular protein tenascin-C, marimastat, and cathepsin B.