We have isolated the cDNA encoding a novel human myogenic factor, Myf‐5, by weak cross‐hydridization to the mouse MyoD1 probe. Nucleotide sequence analysis and the identification of the corresponding gene indicate that human Myf‐5 is a member of a small gene family which also contains the human homologue to MyoD1. Although structurally related to the mouse factor, the human Myf‐5 constitutes a different protein which nevertheless is capable of inducing the myogenic phenotype in embryonic C3H mouse 10T1/2 ‘fibroblasts’. The existence of more than one MyoD1‐like protein in human skeletal muscle is further suggested by the detection of several similar but distinct cDNA clones. The phenotypic conversion of 10T1/2 cells by the human factor is recognized by the capacity of the cells to form multinucleated syncytia and synthesize sarcomeric myosin heavy chains. Moreover, transient expression of Myf‐5 in 10T1/2 cells leads to the activation of a co‐transfected muscle‐specific CAT reporter gene which by itself is transcriptionally silent in the non‐muscle cell background. The deduced amino acid sequence of clone Myf‐5 reveals a region which is highly similar to myc proteins and the developmental factors from Drosophila encoded by the achaete scute locus and the twist gene. The myc homology region and a preceding cluster of basic amino acids are located in a larger sequence domain with strong similarity to the mouse myogenic factor MyoD1. Two additional short segments with high serine and threonine content are conserved between the two proteins.(ABSTRACT TRUNCATED AT 250 WORDS)

skeletal muscle and skeletal muscle precursors, and on cell culture experiments, all of which pointed to func-and B. J. Wold § *Hamon Center for Basic Cancer Research tions in specification of myoblasts and execution of muscle differentiation. The major muscle phenotypes The University of Texas Southwestern Medical Center at Dallas described thus far are the failure of muscle differentiation in mice homozygous for myogenin null mutations Dallas, Texas 75235-9148 † Technische Universitat Braunschweig (Hasty et al., 1993; Nabeshima et al., 1993) and failure to produce the myogenic precursor cell population in Institut fur Biochemie und Biotechnologie 3300 Braunschweig mice deficient in both MyoD and Myf5 (Rudnicki et al., 1993). Unlike muscle deficits, the skeletal phenotypes Federal Republic of Germany ‡ Division of Eukaryotic Molecular Genetics were not expected, although they can account for the lethality of several MRF mutations. MRC National Institute for Medical Research London NW7 1AA MRF4 (Rhodes and Konieczny, 1989) is the final myogenic bHLH gene to be targeted in mice. It is located United Kingdom § Division of Biology 156-29 approximately 8 kb 5Ј of Myf5 on mouse chromosome 10. The proximity of MRF4 and Myf5 to each other raises

The Myf-6 gene, a novel member of the human gene family of muscle determination factors has been detected by its highly conserved sequence coding for a putative helix-loop-helix domain.

No abstract

A major challenge of the postgenomic era is the functional characterization of every single gene within the mammalian genome. In an effort to address this challenge, we assembled a collection of mutations in mouse embryonic stem (ES) cells, which is the largest publicly accessible collection of such mutations to date. Using four different gene-trap vectors, we generated 5,142 sequences adjacent to the gene-trap integration sites (gene-trap sequence tags; http:͞͞genetrap.de) from >11,000 ES cell clones. Although most of the gene-trap vector insertions occurred randomly throughout the genome, we found both vector-independent and vector-specific integration ''hot spots.'' Because >50% of the hot spots were vector-specific, we conclude that the most effective way to saturate the mouse genome with gene-trap insertions is by using a combination of gene-trap vectors. When a random sample of gene-trap integrations was passaged to the germ line, 59% (17 of 29) produced an observable phenotype in transgenic mice, a frequency similar to that achieved by conventional gene targeting. Thus, gene trapping allows a large-scale and cost-effective production of ES cell clones with mutations distributed throughout the genome, a resource likely to accelerate genome annotation and the in vivo modeling of human disease.