Molecular and Cellular Biology, November 2001, p. 7587-7600, Vol. 21, No. 22
0270-7306/01/$04.00+0 DOI: 10.1128/MCB.21.22.7587-7600.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.


Department of Biochemistry and Molecular Biology,1 Department of Urology,2 and Department of Surgery,3 Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, California 90089-9176
Received 7 June 2001/Returned for modification 30 July 2001/Accepted 16 August 2001
Tumor suppressor gene inactivation is a crucial event in oncogenesis. Gene inactivation mechanisms include events resulting in loss of heterozygosity (LOH), gene mutation, and transcriptional silencing. The contribution of each of these different pathways varies among tumor suppressor genes and by cancer type. The factors that influence the relative utilization of gene inactivation pathways are poorly understood. In this study, we describe a detailed quantitative analysis of the three major gene inactivation mechanisms for a model gene at two different genomic integration sites in mouse embryonic stem (ES) cells. In addition, we targeted the major DNA methyltransferase gene, Dnmt1, to investigate the relative contribution of DNA methylation to these various competing gene inactivation pathways. Our data show that gene loss is the predominant mode of inactivation of a herpes simplex virus thymidine kinase neomycin phosphotransferase reporter gene (HSV-TKNeo) at the two integration sites tested and that this event is significantly reduced in Dnmt1-deficient cells. Gene silencing by promoter methylation requires Dnmt1, suggesting that the expression of Dnmt3a and Dnmt3b alone in ES cells is insufficient to achieve effective gene silencing. We used a novel assay to show that missense mutation rates are also substantially reduced in Dnmt1-deficient cells. This is the first direct demonstration that DNA methylation affects point mutation rates in mammalian cells. Surprisingly, the fraction of CpG transition mutations was not reduced in Dnmt1-deficient cells. Finally, we show that methyl group-deficient growth conditions do not cause an increase in missense mutation rates in Dnmt1-proficient cells, as predicted by methyltransferase-mediated mutagenesis models. We conclude that Dnmt1 deficiency and the accompanying genomic DNA hypomethylation result in a reduction of three major pathways of gene inactivation in our model system.
Present address: Division of Molecular Genetics and Center of
Biomedical Genetics, The Netherlands Cancer Institute, 1066 CX
Amsterdam, The Netherlands.
Present address: Life Sciences Division, Lawrence Berkeley
National Laboratory, Berkeley, CA 94720.
§
Present address: Hubrecht Laboratory/Netherlands Institute for
Developmental Biology, 3584 CT Utrecht, The Netherlands.
This article has been cited by other articles:
| J. Bacteriol. | J. Virol. | Eukaryot. Cell |
|---|
| Microbiol. Mol. Biol. Rev. | Clin. Vaccine Immunol. | All ASM Journals |
|---|