University of Southern California
Ray R. Irani Hall
Molecular and Computational Biology
Seminar Series
Hui Zong
University of Oregon
“Analyze Gliomagenesis as a Developmental Lineage Problem with MADM, a Mouse Genetic Mosaic System”
Abstract:
Gliomas are incurable due to the diffuse nature of tumor cells and their resistance to radio- and chemo-therapy. Designing of effective therapies will depend on the full understanding of properties of tumor propagating cells. Since end stage tumor cells aberrantly express many genes, it would be ideal to study glioma-forming cells from the beginning. However, pre-malignant lesions, even in animal models, are difficult to identify due to the lacking of well-defined pathologic features. Here we used a novel mouse model termed MADM (Mosaic Analysis with Double Markers, Zong 2005 Cell; Muzumdar 2007 PNAS) to circumvent the problems by studying the entire process of gliomagenesis.
p53 and NF1, two of the most mutated genes in human gliomas, were mutated in a very small number of neural stem cells (NSCs) and concurrently marked with GFP by MADM in an otherwise normal colorless mouse. Sparse and unequivocal labeling of mutant cells enabled us to analyze their cellular identity at single-cell resolution throughout glioma progression. Lineage analysis of GFP+ mutant cells during pre-malignant stages indicated that the only expanded lineage at this stage was oligodendrocyte precursor cells (OPCs), a type of glial progenitor. Pre-transformed OPCs were BrdU positive after the long-term pulse chase, suggesting that they divide slowly as stem cells would. By the age of six months, we found that brain tumors bearing salient human glioma features formed at almost full penetrance in the MADM-based mouse model. All tumor cells were GFP+ and maintained the expression of a panel of OPC-specific markers. Surprisingly, these OPC-like tumor cells purified from fully developed gliomas acquired many stem cell properties. They can grow into renewable “tumor spheres” in neural stem cell culture medium (EGF+FGF); upon serum induction, differentiate into all three CNS cell lineages; and form secondary and tertiary tumors when injected into NOD-SCID mice. These data suggest that tumor propagating cells in our glioma model have an amalgamated identity between OPCs and NSCs. Currently we are investigating the molecular mechanisms that lead to such aberrant identity with genome and transcriptome analysis. The identification of causal mechanisms should provide important guidance for effective therapeutic strategies toward this devastating disease.
WEDNESDAY, April 21, 2010
12:00 noon
RRI 101
Hosts: Norm Arnheim & Michelle Arbeitman
