Gene Expression

Embryonic stem cell-specific microRNAs. Houbaviy, H. B. et al. Dev. Cell 5, 351–358 (2003)

The authors identify a set of microRNAs in undifferentiated and differentiated mouse embryonic stem (ES) cells. Their sequences are similar and are encoded by loci that cluster within 2.2 kb of each other. Because their expression is repressed as the ES cells differentiate, the authors suggest that they might have a role in the maintenance of the pluripotent character of these cells and in the regulation of early mammalian development.

Evolution

Molecular phylogenies link rates of evolution and speciation. Webster, A. J. et al. Science 301, 478

There are good theoretical reasons to think that the rates of genetic evolution and speciation should be linked, but empirical confirmation of a general relationship between the two has been lacking. In their careful analysis, Webster et al. show that these rates are correlated in 30–50% of 56 previously published DNA-based phylogenies. The authors suggest that a punctuated molecular clock, in which rapid genetic evolution is often linked to speciation, best explains this general pattern.

Functional Genomics

A large-scale, gene-driven mutagenesis approach for functional analysis of the mouse genome. Hansen, J. et al. Proc. Natl Acad. Sci. USA August 2003 (10.1073/pnas.1633296100)

Gene traps — vectors that insert reporter genes at many sites throughout a genome — can be used to establish mouse embryonic stem (ES) cell libraries with mutations in most genes. Hansen et al. show that different gene traps have different insertional 'hot spots' in the mouse genome and so, when used in combination, complement each other. They also show that gene trapping can mutagenize all functional classes of genes and is as efficient as conventional gene targeting.

Cancer Genetics

A mechanism of cyclin D1 action encoded in the patterns of gene expression in human cancer. Lamb, J. et al. Cell 114, 323–334 (2003)

Cyclin D1 (CD1) is overexpressed in many types of tumour, but exactly how it contributes to the development of cancer was previously unclear. Lamb et al. examined the microarray-expression profiles already available for a wide range of tumour types, for genes that are frequently co-expressed with CD1. Their sophisticated data-mining approach allowed them to identify a gene not previously associated with CD1 as a key effector of its influence on transcription. Their approach could be used to analyse many genes that are involved in cancer development.