The formation of mature tissues and cell types in eukaryotic organisms requires that cells undergo a regulated series of steps leading to terminal cell differentiation, which includes a permanent withdrawal from the cell cycle and the eventual cessation of all cell proliferation. Concomitant with this process is the stable repression of many genes involved in normal cell growth and cell cycle control, accompanied by profound changes in nuclear chromatin structure that arise as previously active genes gradually become silenced and are modified epigenetically to form facultative heterochromatin (Grigoryev et al. 2006). At the same time, newly activated signaling pathways must induce the de novo expression, function, or nuclear localization of the various tissue-specific enhancer-binding proteins needed to regulate the genes responsible for creating and maintaining the differentiated cell phenotype. A powerful system used to identify the fundamental principles of cell fate specification is the process of skeletal muscle differentation, or myogenesis, in which multipotential mesodermal precursor cells commit and differentiate to a muscle cell fate (Sartorelli and Caretti 2005). This process can be studied in defined sequential stages, wherein the precursor cells first commit to form undifferentiated myoblasts, then differentiate and fuse to form multinucleated myotubes, and subsequently mature into functional myofibers. Key upstream activators in this process include members of the MyoD family of basic helix–loop–helix (bHLH) regulators (MyoD, myogenin, MRF4, and Myf5), as well as the MEF2 family of MADS-box factors (MEF2-A, MEF2-B, MEF2-C, MEF2D). In this system, the myogenic bHLH factors form heterodimers with E-box factors (E12, E47), and function cooperatively with the MEF2 proteins to activate skeletal muscle-specific genes, while also creating strong autoregulatory feedback loops that ensure their own sustained expression (Bassel-Duby and Olson 2006; Baugh and Hunter 2006). MEF2 can also act independently of MyoD through its ability to associate with myocardin and MASTR/SAP domain protein in cardiac and skeletal muscle cells, respectively (Creemers et al. 2006). Moreover, MEF2C also interacts directly with the Notch coactivator, Mastermind (MamL1), the targeted deletion of which causes a severe muscular dystrophy in mice (Shen et al. 2006). These factors assemble large muscle-specific enhancer complexes and cooperate with other DNAbound activators, such as the serum response factor (SRF), to displace repressors and recruit the coactivators needed to up-regulate transcription (Pipes et al. 2006). In this issue of Genes & Development, Deato and Tjian (2007) describe a remarkable new mode of transcriptional regulation in differentiated cells that is centered on the core transcriptional machinery. Detailed analysis of core promoter factor expression in these cells reveals that myotube differentiation is accompanied by a dramatic loss of the TATA-binding protein (TBP) as well as many of the RNA polymerase II (RNAPII) TATA-associated factor (TAF) subunits (TAF1, TAF4, TAF9, TAF10), suggestive of a wholesale extinction of the TFIID complex. The loss of TFIID components, which was detected at both the RNA and protein level, may provide an effective mechanism for large-scale silencing of genes whose functions are needed only in undifferentiated cells, including those that control cell proliferation and progression through the cell cycle. These findings confirm and extend an earlier report from Perletti et al. (2001) that both TBP and TAF4 are down-regulated by targeted proteolysis in C2C12-derived myotubes and differentiated F9 cells. Deato and Tjian (2007) further discovered that the loss of core factors in differentiated myotubes is highly selective and that the expression of certain other core factors, namely the TATA-related factor TRF3 and the TAF3 protein, persists in differentiated cells. Interestingly, TRF3 and TAF3 were found to interact directly to form a minimal core promoter recognition complex that binds to the Myogenin core promoter and cooperates with the myogenic upstream activators to activate transcription in differentiated myotubes. Further analyses in stable small interfering RNA (siRNA)-expressing cell lines established that TRF3 and TAF3 are both necessary for myogenesis, and have no effect on TFIID function or stability in dividing myoblasts. Taken together, these findings support a model in which core promoter switching mediates the global repression of TFIID-dependent genes and simultaneously allows for selective activation of myogenic genes that deCorrespondence. E-MAIL jones@salk.edu; FAX (858) 535-8194. Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1598007.