Keeping a good pathway down: transcriptional repression of Notch pathway target genes by CSL proteins

Abstract

CSL [CBF-1, Su(H), Lag-1]-type transcription factors are the primary effectors of the Notch pathway, a signal transduction cascade that is essential for the development of all metazoan organisms. Interestingly, CSL proteins were originally classified as transcriptional repressors in vertebrates, but as transcriptional activators in model invertebrate organisms. Resolution of this paradox came with the realization that repression and activation by CSL proteins occurs in both systems and that the switch involves recruitment of distinct co-repressor and co-activator complexes. Although CSL proteins appear to utilize a common co-activator complex of largely similar constitution, recent studies have demonstrated that vertebrate and Drosophila CSL interact with a variety of distinct co-repressor complexes. This review highlights differences in composition and similarities in function of different CSL co-repressor complexes, which actively repress Notch pathway target genes in the absence of Notch pathway activity.

Fig. 1. CSL proteins associate with co-repressor complexes containing HDAC activity and a co-activator complex containing HAT activity. CSL proteins (blue) bind with high affinity to the consensus site YRTGDGAD (certain other sites included by the broader consensus RTGRGAR are bound with lower affinity). The vertebrate ortholog CBF-1 interacts with the CIR+SMRT/N-CoR co-repressor complex (A), while Drosophila Su(H) uses Hairless (H) to recruit both dCtBP (B) and Gro (C) co-repressor complexes. Drosophila SMRTER associates with Su(H), but has not yet been shown to mediate repression by Su(H) in vivo. All of these co-repressors have been functionally linked to HDACs (red), whose activity leads to a transcriptionally repressed chromatin state. Note that the composition of these repression complexes has been extrapolated from multiple studies and is tentative. Other components of these co-repressor complexes may exist, particularly for the Gro and CtBP complexes, which have not been as extensively characterized as the SMRT/Sin3A complex. In Drosophila, Gro represses transcription over greater distances than does dCtBP (as depicted by a greater extent of histone deacetylation), but it remains to be determined if these co-repressors mediate separable modes of ‘short-range’ and ‘long-range’ repression by Su(H). HDAC-independent modes of repression might also exist for these co-repressors or for CBF-1 (data not shown, see text for details). (D) Vertebrate and Drosophila CSL proteins interact with a co-activator complex that includes N^IC, Mam and HATs (green).

Eric C. Lai