1. Division of Molecular Carcinogenesis, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands
2. Division of Molecular Genetics, The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands
3. Dana Farber Cancer Institute, Harvard Medical School, 44 Binney Street (M463), Boston , Massachusetts 02115, USA

Correspondence to: Maarten van Lohuizen¹ Correspondence and requests for materials should be addressed to M.v.L. (e-mail: Email: lohuizen@nki.nl).

Abstract

The bmi-1 gene was first isolated as an oncogene that cooperates with c-myc in the generation of mouse lymphomas¹,². We subsequently identified Bmi-1 as a transcriptional repressor belonging to the mouse Polycomb group^{3, 4, 5, 6}. The Polycomb group comprises an important, conserved set of proteins that are required to maintain stable repression of specific target genes, such as homeobox-cluster genes, during development^{7, 8, 9}. In mice, the absence of bmi-1 expression results in neurological defects and severe proliferative defects in lymphoid cells, whereas bmi-1 overexpression induces lymphomas⁴,¹⁰. Here we show that bmi-1-deficient primary mouse embryonic fibroblasts are impaired in progression into the S phase of the cell cycle and undergo premature senescence. In these fibroblasts and in bmi-1-deficient lymphocytes, the expression of the tumour suppressors p16 and p19^Arf, which are encoded by ink4a, is raised markedly. Conversely, overexpression of bmi-1 allows fibroblast immortalization, downregulates expression of p16 and p19^Arf and, in combination with H-ras, leads to neoplastic transformation. Removal of ink4a dramatically reduces the lymphoid and neurological defects seen in bmi-1-deficient mice, indicating that ink4a is a critical in vivo target for Bmi-1. Our results connect transcriptional repression by Polycomb-group proteins with cell-cycle control and senescence.