A Notch Affair

(Fortini and Artavanis-Tsakonas, 1994), or that activation Poulson first described Drosophila embryos that lacked brings about proteolytic cleavage of Notch, so that an Notch function in 1937 (Poulson, 1937), referring to them intracellular portion of Notch itself moves to the nucleus as “a kind of hopeless monster.” Half a century later it and collaborates with CSL to activate transcription (Kois becoming clear how universally important Notch-like pan et al., 1996). Four talks presented evidence in supmolecules are in development, and the discovery that port of the latter model. Raphael Kopan (Washington a late onset disease, CADASIL, is caused by mutations University, St. Louis), who has long been a proponent in Notch 3 suggests these proteins may also have roles of Notch processing, presented data identifying the inin the adult (Joutel et al., 1996). The Drosophila Notch tracellular site of cleavage in mNotch1. Using transgene encodes a 300 kDa receptor, and in spite of the activation of the HES-1 target-gene promoter as an fact that Notch mutations have been around for deassay in cell transfection experiments, he demonstrated cades, there are still many aspects of the structure and that the activation was curtailed when the intracellular function of this receptor and its transmembrane ligands cleavage site was mutated in a transmembrane derivathat remain enigmatic. For example, the extracellular tive of mNotch1. Similar mutations in an mNotch1 derivdomain of Notch contains 36 EGF repeats, but only ative that lacked the transmembrane domain had no two of these (11 and 12) have been assigned a specific effect. Alain Israel (Institut Pasteur, Paris) added further function in ligand binding (Rebay et al., 1991). Furtherdimensions by proposing that there are three sites for more, although transmission of theNotch signal involves processing within the murine Notch1 molecule (Figures a DNA-binding protein that interacts with the intracellu1B and 1C). The first processing event is due to a constilar domain of Notch (Fortini and Artavanis-Tsakonas, tutive protease pathway that cleaves the protein in the 1994), the mechanisms of activation and transduction extracellular domain, resulting in two fragments that remain controversial. Finally, we are still trying to underhave been observed in previous immunoprecipitation stand what information Notch signaling conveys to the studies (Blaumueller et al., 1997; Pan and Rubin, 1997) cell. These issues are all relevant to the Notch homologs and which remain associated. The enzyme involved is in vertebrates, whose fundamental structure is well conapparently not the ADAM protease Kuzbanian as origiserved (Artavanis-Tsakonas et al., 1995), and to the relanally proposed (Pan and Rubin, 1997). This instead pertives in C. elegans that have smaller extracellular doforms a second, previously undescribed, processing mains (Kimble and Simpson, 1997). In addition, the step that again cuts Notch in the extracellular domain. subtly different structures and diverse expression patIt was proposed that cleavage by Kuzbanian occurs as terns of the individual vertebrate Notch proteins and a consequence of DSL ligand binding and that it rapidly ligands opens up the potential for further complexity. So, triggers a further processing event inside the cell. This were the mysteries of Notch unraveled at the stimulating final intracellular cleavage can be blocked by a proteaJuan March meeting held March 9–11, 1998, and orgasome inhibitor, and it equates with that described by nized by Alfonso Martinez-Arias, Sonsoles Campuzano, Kopan. and Juan Modolell that brought together people from Two labs had investigated the mechanisms of signal the Drosophila, C. elegans, and vertebrate worlds of transduction by developing functional assays to detect Notch/LIN-12 signaling? That would have been a tall Notch intracellular domain (NICD) in the nucleus. David order but we were shown clues to at least some parts Ish-Horowicz (ICRF, London) described the effects of of the riddle. Below I have drawn together the main fusing transcriptional activation and repression domains threads which ran through the meeting, and I regret that to the intracellular portion of a transmembrane-tethered it has not been possible to weave in everything. The derivative of XNotch. When assayed in Xenopus embryos, now-accepted core for what has become known as the the former led to a decrease in neurons whereas the “Notch signaling pathway,” around which discussions latter caused an increase in neurons, consistent with in Madrid revolved, includes the Delta and Serrate famithe predicted outcome if a Notch fragment containing lies of ligands and the CSL (CBF, Suppressor of Hairless, the heterologous domains reached the nucleus. An eleLAG-1) transcription factors (Figure 1A; Artavanis-Tsagant assay described by Francois Schweisguth (Ecole konas et al., 1995; Kimble and Simpson, 1997; Robey, Normale Superieure, Paris) involved making transgenic 1997). The questions addressed can be broadly divided flies expressing a fusion protein that incorporated the into three areas: the mechanisms underlying the transGAL4 DNA-binding domain into the intracellular domain duction of the signal from Notch, the ways that Notch of a Notch full-length protein. If Notch is proteolysed, activitymay be regulated, and thedevelopmental effects the fragment released would contain GAL4 and should