Mouse models of radiation-induced glioblastoma

Glioblastomas (GBM) are lethal brain tumors that can be triggered by exposure to ionizing radiation (IR), even at low doses from CT scans [1]. High doses of IR are also used to treat GBM, but the irradiated tumors inevitably recur. This raises the possibility that genomic changes induced by radiation may contribute not only to glioma initiation, but also to tumor recurrence. Thus, there is a compelling need for experimental model systems that recapitulate the process of radiation-induced gliomagenesis. Such models could not only help predict GBM-development risks from radiation exposure, but also help identify genetic alterations defining radiation-induced GBM, thereby facilitating the development of rational therapies for treating these recalcitrant tumors. Our study published in the journal Oncogene employed a systematic approach to develop sensitive mouse models that can be used to study radiation-induced gliomagenesis [2]. Ink4a, Ink4b and Arf are key tumor suppressor genes that are deleted in a majority of GBMs [3]. We utilized transgenic mice with brain-restricted deletions of these tumor suppressors, individually and in combination, and examined their susceptibility to IR-induced GBM development. The most deleterious lesion inflicted by IR is the DNA double-strand break (DSB). We have shown previously that accelerated ions (particle radiation) induce complex DSBs that are refractory to repair unlike the simple breaks induced by X-rays (electromagnetic radiation) which are repaired to completion [4]. Therefore, we intra-cranially irradiated these transgenic mice with either X-rays or accelerated Fe ions to understand the process of radiation-induced gliomagenesis, and how this may be influenced by DNA damage complexity. We found that these mice did not develop gliomas spontaneously, but were prone to GBM development after exposure to a single, moderate dose of radiation. Remarkably, we found that Fe ions were at least four-fold more effective than X-rays in inducing these tumors, thereby confirming that complex DSBs triggered by accelerated ions are more harmful than simpler breaks induced by X-rays. This finding has important implications as the use of particle radiation (such as protons and carbon ions) for cancer therapy is steadily increasing. Our work indicates that particle radiation could indeed turn out to be more effective than X-rays for tumor control, but this also raises the specter of increased likelihood of secondary cancers triggered by such radiation. Interestingly, while wild type mice did not develop gliomas upon radiation exposure, loss of Ink4a and Arf was sufficient to render these mice susceptible to IR-induced gliomas; …