Abstract 3300: Cancer stem cell heterogeneity and cell-of-origin in medulloblastomas

Abstract Understanding the origin of cancer stem cell (CSC) heterogeneity within a tumor type is one of the major challenges in the CSC field currently. In this study, we report that Ptch+/− mice develop three distinguishable subtypes of medulloblastomas, and that CSCs in each subtype depend on diff...

Full description

Saved in:
Bibliographic Details
Published inCancer research (Chicago, Ill.) Vol. 72; no. 8_Supplement; p. 3300
Main Authors Harris, Molly, Shin, Dong-mi, Choi, Seungbum, Low, Benjamin, Miller, Emily, Rybinski, Brad, Bronson, Roderick, Yun, Kyuson
Format Journal Article
LanguageEnglish
Published 15.04.2012
Online AccessGet full text

Cover

Loading…
More Information
Summary:Abstract Understanding the origin of cancer stem cell (CSC) heterogeneity within a tumor type is one of the major challenges in the CSC field currently. In this study, we report that Ptch+/− mice develop three distinguishable subtypes of medulloblastomas, and that CSCs in each subtype depend on different signaling pathways for self-renewal and survival. In all three tumor subtypes, the tumor initiating cell frequency is low (< 1/300 cells) and 100% of the tumors can be serially passaged in vivo for at least 5 generations, indicating that each tumor subtype contains rare, long-term self-renewing cells with tumor-initiating ability. However, CSCs in Type I tumors require in vivo microenvironment for survival/proliferation; hence, they do not grow in stem cell cultures in vitro. CSCs in Type II and Type III tumors form tumorspheres but require different growth factor signaling pathways for self-renewal and proliferation. Remarkably, except in rare cases where spontaneous tumor progression occurs, the CSC phenotype remains identical to the original tumor through multiple in vivo serial passages, indicating that molecular and cellular phenotypes of CSCs in each tumor are stable, cell-intrinsic characteristics of CSCs. Based on their in vitro culture requirements and their gene expression patterns, we hypothesized that different CSC subtypes arise from different cell-of-origin. We tested this hypothesis by activating the SHH pathway in cell type-specific manner using genetically engineered mice. We provide strong evidence that Type I tumors arise from EGL progenitors and that Type II tumors arise from embryonic neural stem cells. We also show that Type I and Type II tumors can spontaneously progress onto Type III tumors both in vivo and in vitro, and that this progression is characterized by gain of chromosome 6. Importantly, the three tumor subtypes can be distinguished at the bulk tumor level by unique gene expression patterns. Analyses of subtype-specific signature genes against human medulloblastoma gene expression data set published by Kool et al., show that different subtypes of Ptch+/− tumors correspond to different human medulloblastoma subtypes. Importantly, they do not all correspond to the SHH-subtype. In summary, we propose that the same initiating oncogenic event can lead to generation of CSCs with distinct molecular and cellular phenotypes depending on the cell of origin. Our observation provides a partial explanation for the failure of SHH-inhibitors to “cure SHH-tumors” in preclinical and clinical trials. Our findings suggest that identification and analyses of CSCs in each tumor may be necessary for efficient stratification and targeting of CSCs in histologically indistinguishable tumors. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3300. doi:1538-7445.AM2012-3300
ISSN:0008-5472
1538-7445
DOI:10.1158/1538-7445.AM2012-3300