We have generated a transgenic mouse model for astrocytoma by expressing the v-src kinase under control of the glial ®brillary acidic protein (GFAP) gene regulatory elements in astrocytes. Abnormal astrogliosis was observed in all transgenic animals already at 2 weeks postnatally, frequently followed by the development of dysplastic changes. Later, small proliferative foci arose, and overt astrocytoma developed in the brain and spinal cord in 14.4% of mice after a follow up time of 65 weeks. While early lesions were histologically consistent with low-grade astrocytoma, at later stages most tumors were highly mitotic and frankly malignant. Vascular endothelial growth factor (VEGF) was expressed by tumor cells already at early stages, suggesting induction by v-src, and it was most pronounced in pseudopalisading cells surrounding necrotic areas, implying additional upregulation by hypoxia. In larger lesions, mitotic activity and expression of¯k-1, the cognate receptor of VEGF were induced in endothelial cells. Therefore, end-stage tumors mimicked the morphological and molecular characteristics of human glioblastoma multiforme. Time course and stochastic nature of the process indicate that v-src did not su ce for malignant transformation, and that astrocytomas were the result of a multistep process necessitating co-operation of additional genetic events.

No abstract

No abstract

Transgenic animal models for neurocarcinogenesis have provided signi®cant insights into the molecular mechanisms underlying carcinogenic processes, including those which affect the nervous system. In view of the very rapid pace of acquisition of knowledge, it is not possible to cover all transgenic mouse models for neural tumors. Instead, this article discusses some of the most important technical innovations for manipulation of the mammalian genome (notably the various methods for targeted genome modi®cations, as well as the technology for introducing large DNA fragments into the germ line of mice), and presents a selection of the transgenic mouse models which are proving most promising for furthering our understanding of the pathogenetic basis of cancer in the nervous system.

No abstract