Notch signaling is a critical form of cell communication used throughout development. Over 90 years ago, mutant fruit flies were identified which exhibited distinct notches on their wings; the unknown, underlying gene was therefore named Notch. Since then, this gene has been discovered to code for a transmembrane protein which serves as a critical regulator of development in flies, worms, mammals, and just about everything else! In short, the Notch protein is a receptor displayed on one cell (receiving cell) that is activated by ligands displayed on a neighboring cell (signaling cell). This activation results in the transportation of part of the Notch protein into the nucleus where it cooperates with other proteins to regulate the transcription of numerous genes which affect processes such as neuronal cell fate determination, vasculogenesis, and even inner ear development (just to mention a few!). The Notch signaling described below is a sort of combination of available information from various model organisms. While each organism may not necessarily perform each of the steps identically, the overall picture is the same. Specific functions of Notch in development are not addressed, but a very thorough description of these events (in Drosophila) can be found HERE. As always, any suggestions/corrections are greatly appreciated…just CLICK HERE.
1 - The Receptor
The Notch receptor is a ~300-350 kDa single-pass transmembrane protein composed of several functional units. The N-terminus is located extracellularly (when Notch is displayed on the cell surface) and is comprised primarily of Epidermal Growth Factor (EGF)-like repeats which are important for binding to the appropriate ligand. Following the EGF-like repeats is a small region known as the Negatively Regulatory Region (NRR) which normally blocks S2 cleavage when Notch is in its unbound state…it sort of acts like an on/off switch during signaling (more on this later…”Receptor Meets Ligand”). The rest of the Notch protein passes through the cell membrane and extends into the cytosol. This Notch IntraCellular Domain (NICD) actually gets cleaved and transported to the nucleus when a ligand is bound (see “S3 Cleavage”).
Figure 1 - Basic Structure of the Notch Receptor
This is a good time to mention the different flavors of Notch receptor. Drosophila has one Notch receptor (called Notch or dNotch) compared to two in C. elegans (Glp-1 and Lin-12), and 4 in mammals (NOTCH1-4). A major difference between these various forms of Notch are the number of EGF-like repeats in the extracellular domain (Figure 2).
Figure 2 - Structural Homology of the Notch Receptor (Gordon et al 2008)
2 - S1 Cleavage and Glycosylation
After its synthesis in the ER, the Notch holoprotein (holoprotein referring to the raw translated protein before any cleavage) is modified during its transport to the cell surface.
First, while in the endoplasmic reticulum, the EGF-like repeats can be glycosylated. The glycans which are bound to Notch are small and come in a variety of flavors, but the primary modifications are O-fucosylation and O-glucosylation (Figure 3; Stanley et al 2009).The enzymes which add these glycans to Notch (and other proteins) are called glycosyltransferases. One such enzyme which modifies Notch is the O-fucosyltransferase, Pofut1 in mammals, and, Ofut1 in Drosophila. Another important enzyme at this step is Fringe (mammals have three orthologues with crazy names like Lunatic Fringe and Manic Fringe). Fringe comes along after Ofut1 and elongates the O-fucose modifications. Bottom line, by the time Notch gets to the surface it is covered with sugars. These sugar modifications provide yet another means to tweak Notch signaling. To get more details on glycosylation check out Stanley et al 2009 and Takeuchi et al 2010.
Figure 3 - "Glycans on Notch. A diagram representing the ECDs of mouse Notch1 and Drosophila Notch which contain 36 EGF repeats (white ovals) and 3 Lin repeats (blue ovals). Symbols in the EGF repeats identify consensus motifs for O-fucose (A), O-glucose (B), O-GlcNAc (C), O-xylose (D), and N-glycans (E) that have the potential to contain the sugars shown in the structures below the diagram." See reference for more explanation (Stanley et al 2009).
A second modification that occurs while in the Golgi is the cleavage of the Notch N-terminal portion. In 1998, Logeat et al demonstrated that Furin, a calcium-dependent convertase, is responsible for cleaving within the HD domain to generate a “free” N-terminus and a membrane embedded C-terminus. This process is termed S1 cleavage (Figure 1) and the resulting N-terminal fragment is non-covalently bound to the C-terminal portion to form a heterodimer. There has been controversy over the actual importance of S1 cleavage and whether it even exists in Drosophila (see Lake et al 2009).
3 - The Ligands
Once the Notch receptor reaches the cell surface, it remains in an inactive state until it interacts with a DSL ligand (receiving their name from the two Drosophila Notch ligands, Delta and Serrate, and the C. elegans orthologue, Lag-2). Similar to the Notch receptor, DSL ligands are bound to the cell surface (with some exceptions) and possess a large N-terminal extracellular domain. The DSL ligands consist of an N-terminal domain followed by a DSL domain, and multiple EGF-like repeats. The DSL domain along with the first two EGF-like repeats (termed the DOS motif) are critical for binding to the Notch receptor. As with the Notch receptor, the DSL ligands display a wide range of variability mainly due to differences in the number and spacing of the EFG-like repeats (Figure 2). The mammalian Delta-like 1, 3, and 4 are homologous to the Drosophila Delta, whereas the mammalian Jagged-like 1 and 2 are homologous to the Drosophila Serrate.
Figure 4 - Structural Homology of the DSL ligands (D'Souza et al 2010)
4 - Ligand Meets Receptor (S2 Cleavage)
When the Notch receptor reaches the cell surface it consists of two subunits which are non-covalently bound together at the HeteroDimerization (HD) domains. This Notch receptor is in its OFF state, and in order to be activated, it must be cleaved at the S2 cleavage site which is located within the HD domains. However, as long as the Notch receptor is unbound (no DSL ligand present), three Lin-12/Notch Repeats (LNR) interact with the HD domains to block S2 cleavage. Together, the LNR and HD domains constitute the Negative Regulatory Region (NRR) because they keep Notch signaling inactive when no ligand is present.
However, upon binding the appropriate DSL ligand, a conformational change occurs in the NRR which exposes the S2 cleavage site to a protease known as Kuzbanian in Drosophila and ADAM10 in mammals. ADAM creatively stands for, A Disintegrin And Metalloproteinase family, and there are a bunch of them in mammals. These membrane-bound metalloproteinases are known for their role in clipping off extracellular domains (a process called ectodomain shedding) (Edwards et al 2008); that’s exactly what happens when ADAM10 cleaves Notch at the S2 site…”Bye, bye” Notch extracellular domain and “Hello” S3 cleavage.
5 - Gamma Secretase (S3 Cleavage)
After S2 cleavage we are left with a small extracellular stub, followed by the membrane spanning region and the Notch IntraCellular Domain (NICD). Removal of the Notch extracellular domain results in immediate subsequent cleavage within the transmembrane domain by gamma-secretase, an intramembranous protease complex (Figure 5). The name gamma-secretase comes from its role in Alzheimer’s disease research; it is the third (alpha, beta, gamma) protease which functions during amyloid precursor protein processing. Gamma-secretase is composed of four protein subunits: Presenilin (Psn), Nicastrin, Anterior Pharynx Defective 1 (Aph-1), and Presenilin Enhancer 2 (Pen-2). Ectopic expression of these four proteins can reconstitute ?-secretase activity in yeast, while removing either one of these four inhibits the activity. In addition to cleaving Notch, ?-secretase is responsible for cleaving other type I transmembrane membrane proteins such as N-cadherin, E-cadherin, and many more.
Figure 5 - Gamma-Secretase Complex (Parks 2007)
S3 cleavage is thought to occur by default immediately after S2 cleavage has released the extracellular domain, but why doesn’t gamma-secretase act on the full length heterodimer before S2 cleavage has occurred? One likely possibility is that gamma-secretase recognizes the truncated extracellular N-terminus which is left over after S2 cleavage (Dries and Yu 2009). Bottom line, S3 cleavage results in the release of the Notch intracellular domain from the cell membrane, allowing it to be transported to the nucleus for transcription activation.
6 - Transcription Activation
When the Notch IntraCellular Domain (NICD) reaches the nucleus, it meets two other proteins which are critical for activating transcription. The DNA binding protein, CSL (CBF1[mammals]/Suppressor of Hairless [Drosophila]/Lag-1 [C. elegans]) recognizes specific motifs throughout the genome. In its default state (no Notch signaling) CSL is bound by corepressors and exerts a repressive effect on its target genes. When Notch signaling is activated and NICDs begin to flood the nucleus, the RAM and ankyrin repeats on the NICD bind to the CSL transcription factor and induce a conformational change. This conformational change causes CSL to favor binding to the protein Mastermind (Mam) over the corepressors. Mam is a coactivator which in turn recruits histone acetyltransferases thereby activating target gene transcription (Bray et al 2010). So CSL, NICD, and Mam all work together to turn ON the Notch target genes.
There are numerous genes targeted by Notch signaling, but the most well studied group is the HES/HEY family of transcription factors. These basic Helix-Loop-Helix (bHLH) proteins operate as transcription repressors. Other target genes include myc, reaper, EGFR, and Notch itself (Bray et al 2010). Because of its far-reaching impact on so many genes, it’s not surprising that Notch signaling seems to turn up almost everywhere you look in development!
Bray et al 2010 Chapter Eight - Notch Targets and Their Regulation
Edwards and Pennington et al 2008 The ADAM metalloproteinases
Gordon et al 2009 Effects of S1 Cleavage on the Structure, Surface Export, and Signaling Activity of Human Notch1 and Notch2
Lake et al 2009 In Vivo Analysis of the Notch Receptor S1 Cleavage
Logeat et al 1998 The Notch1 receptor is cleaved constitutively by a furin-like convertase
Pan et al 1997 Kuzbanian Controls Proteolytic Processing of Notch and Mediates Lateral Inhibition during Drosophila and Vertebrate Neurogenesis
Parks and Curtis 2007 Presenilin diversifies its portfolio
Sanchez-Irazarry et al 2004 Notch Subunit Heterodimerization and Prevention of Ligand-Independent Proteolytic Activation Depend, Respectively, on a Novel Domain and the LNR Repeats
Stanley et al 2009 Chapter Four – Roles of Glycosylation in Notch Signaling
Takeuchi et al 2010 Role of glycosylation of Notch in development