Introduction and Overview

Notch receptors are modular, single-pass transmembrane receptors that respond to transmembrane ligands expressed on neighboring cells. Signals transduced by Notch receptors influence a wide range of cell fate decisions both during development and in the mature organism.

Notch receptors are activated by ligand-induced proteolysis, which ultimately releases the intracellular portion of the receptor from the membrane, allowing it to enter the nucleus and regulate target gene transcription (Figure 1). Notch receptors are synthesized as large precursor glycoproteins (Figure 2), which are typically processed by a furin-like protease at site S1 into two non-covalently associated subunits during maturation. A series of N-terminal EGF-like modules constitutes the part of the receptor responsible for ligand binding. The receptors are normally resistant to activating proteolysis until ligand binding occurs. The ligand-dependent proteolysis step, which occurs at a site called S2, is carried out by metalloproteases of the ADAM family.

A negative regulatory region C-terminal to the EGF-like repeats maintains the receptors in the "off" state prior to ligand-induced activation. This region consists of three Lin12/Notch repeats virtually unique to Notch receptors and a heterodimerization domain divided by S1 cleavage that also houses the regulated metalloprotease cleavage site. The metalloprotease-cleaved form of Notch becomes a substrate for S3 cleavage by the intramembrane protease gamma-secretase, which releases the intracellular portion of Notch from the membrane. The untethered intracellular fragment of Notch then migrates to the nucleus where it participates in the activation of transcription of Notch-responsive genes. The long-term objectives of our studies are to understand the molecular logic of Notch signaling, using structural, molecular, and biophysical approaches.

Mechanism of activation and the "on/off" switch

The laboratory recently determined the structure of the negative regulatory region of Notch2 in its protease-resistant conformation. This structure revealed the basis for Notch autoinhibition (Figure 3). We are currently investigating the mechanism by which ligand engagement renders normal receptors sensitive to proteolysis at site S2, and studying how cancer-associated activating mutations in human Notch1, which occur in more than half of human T cell acute lymphocytic leukemia/lymphomas, bypass normal restraints on activation. Structures of the Notch2 and Notch1 negative regulatory regions will also inform an ongoing search for small molecules that can selectively prevent ligand-induced activation of Notch1 in vivo.

Complexes of ICN with effectors and modulators of signaling

The primary known effector of Notch1 signaling is the DNA-binding protein CSL. When the intracellular fragment of Notch1 (ICN) binds to CSL, ICN converts this target molecule from a transcriptional repressor to a potent transcriptional activator. The ankyrin repeat region (ANK) and an N-terminal flanking region of ICN called RAM mediate association with CSL. Transcriptional activation depends on additional recruitment of Mastermind proteins (MAML1-3 in humans), which associate into complexes only when CSL and the ankyrin domain of ICN are both present.

We recently determined X-ray structures for a MAML1-ANK-CSL-DNA complex using human proteins. This structure showed that the key region of the MAML1 coactivator binds in a kinked helical conformation to a composite interface created by docking of the ANK domain against the Rel-homology region of the CSL transcription factor (Figure 4). A concomitant publication of a homologous structure using components from C. elegans by our colleagues identified the same composite interface in worm complexes as well.

A striking feature of the promoter regions of a number of well-characterized Notch-responsive genes, like the Hairy/Enhancer-of-split genes in Drosophila and a number of their mammalian homologues, is that they often have more than one binding site for the CSL transcription factor. Indeed, pairs of CSL binding sites in these promoters are frequently positioned in a characteristic head-to-head arrangement with a typical spacing of 16 or 17 nucleotides between sites. These sites are typically located on opposite strands of canonical Notch target genes, like HES-1 in mammals and many of the enhancer of split genes in Drosophila.

Our X-ray structure of the human Notch transcription complex core contained contacts between the convex surfaces of ANK domains near a two-fold symmetry axis in the crystals, such that the interacting complexes are positioned head-to-head at a distance roughly equal to that needed to occupy both recognition elements of an canonical paired site (Figure 5). Using a combination of biochemical and molecular approaches, we recently established that cooperative formation of dimeric Notch transcription complexes on promoters with paired sites is required to activate transcription. Our findings identified a new mechanistic step that can account for the exquisite sensitivity of Notch target genes to variation in signal strength and developmental context. One emphasis of our future studies of nuclear complexes will be to understand the structural basis for the cooperative loading of these Notch transcription complexes at these paired DNA binding sites. We are also interested in investigating other cooperative interactions that regulate transcription of other target genes that do not have clustered CSL binding sites in their promoters, but that are nevertheless identified as direct Notch targets.