Abstract
Transgenic mouse models of T-cell oncogenes such as TAL1 and LMO1, have confirmed their critical role in the development of T-lymphoblastic lymphoma (T-LL). In addition, retroviral insertional mutagenesis (RIM) in mice is a powerful system for the identification of genes involved in T-cells leukemogenesis. However, incomplete knowledge of the pathogenesis of T-ALL limits the ability to stratify patients and to deliver tailored therapy accordingly. Thus, a major goal is to identify common oncogenic pathways downstream of T-cell oncogenes that can serve as therapeutic targets for the treatment of T-ALL. We hypothesize that T-cell oncogenes operate through a limited number of oncogenic pathways with distinct gene expression profiles. Furthermore, we propose that gene expression profiling of murine models of T-cell lymphomas, which harbor specific genetic lesions, will serve to identify such oncogenic pathways and to establish a molecular classification of T-ALL. To test this hypothesis we have analyzed normal thymus and mouse T-cell lymphomas originated from retroviral insertional mutagenesis (n=11) and transgenic and knock-outs models (n=30) using Affymetrix 430A2.0 microarrays. Unsupervised analysis clearly distinguishes tumor samples from normal thymus and clusters the tumors into three major groups. TAL1, LMO2 and E2 proteins are known to be part of a transcriptional complex and to cooperate in T-cell leukemogenesis by suppressing E2A function. Accordingly, mouse tumors originating in TAL1, TAL1/LMO2 transgenic and E2A knock-out mice share a common gene expression signature and cluster together. Supervised analysis of tumors generated by retroviral mutagenesis showed increased expression of the proviral tagged genes and identified corresponding known downstream targets. Nearest neighbor analysis identified high levels of Notch1 expression in tumors with proviral insertion in the Notch1 locus and in tumors generated in the TAL1/, TAL1/LMO2, OLIG2/LMO1, ThPOK and Ikaros mouse models, which harbored activating mutations in NOTCH1. Our results demonstrate that gene expression profiling identifies common oncogenic pathways in T-cell tumors generated in mice, establishes common mechanisms of transformation for several T-ALL oncogenes and allows coupling of poorly characterized genes identified in proviral insertional sites with well characterized oncogenes and downstream molecular pathways. The identification of mechanisms of T-cell transformation common to tumors of different origin lays the ground for the identification of new therapeutic targets for the treatment of T-ALL.
Disclosure: No relevant conflicts of interest to declare.
Author notes
Corresponding author