Molecular diversity in the Notch receptor family

Abstract

We have studied the Notch family of transmembrane receptors, which play a crucial role in cell fate determination. The Notch signalling pathway represents a conserved mechanism to mediate signalling between adjacent, equivalent cells and to direct them to adopt different cell fates. This process, called lateral inhibition, is involved in many processes during development e.g. bristle development in Drosophila and retina and pancreas development in vertebrates. The current view is that the Notch receptor is cleaved intracellularly upon activation by Delta or Serrate ligands on neighbouring cells. The intracellular Notch domain then translocates to the nucleus, binds to the DNA- binding factor Suppressor of Hairless (CSL in mammals), and acts as a transactivator of Enhancer of Split (HES in mammals) gene expression.

Several studies have demonstrated that the intracellular domain alone functions as a constitutively-active receptor. There are four known mammalian Notch receptors with overlapping patterns of expression. The role of the different receptors is poorly understood, but mutations in the receptors result in distinct genetic disorders like CADASIL and a variety of tumours CADASIL (Cerebral Autosomal Dominant Arteriopathy with Subcortical. Infarcts and Leukoencephalopathy) is an inherited form of stroke and dementia. The histopathological hallmarks are degeneration of vascular smooth muscle cells (VSMCs) and accumulation of granular osmiophilic material (GOM)between the degenerating VSMCs. CADASIL is caused by missense mutations in the human Notch3 gene. The mutations always affect cysteine residues located in the extracellular domain of the Notch3 receptor. There is currently no biochemical data available on the mechanism by which the mutated Notch3 receptor causes CADASIL.

The aim of this study has been to understand the extent of functional diversity in the Notch receptor family. More specifically, we have investigated the molecular differences between the Notch1 and Notch3 receptors in terms of transcriptional activity and interactions with transcriptional activators and repressors. We have also extended the invitro knowledge in studying the effects of Notch3 signalling during pancreas development and in CADASIL pathogenesis. We show that the intracellular domain of Notch3 (Notch3 IC), unlike Notch1 IC, is a weak activator of HES gene expression both in cell-based systems and in vivo (Paper 1). Furthermore, we find that the low transcriptional activity of Notch3 IC is dominant over Notch1 IC-mediated activation. Hence, Notch3 IC behaves as a functional repressor of Notch1 signalling with respect to the HES genes. To learn, in detail, which domains are responsible for the differences between the receptors, we have molecularly dissected the Notch1 and Notch3 ICs (Paper II). We make two important observations.

First, we have identified a region, RE/AC, which is required for both activation and repression. Loss of this region in both ICs results in complete loss of function, i.e. Notch1 IC loses its activator function and Notch3 IC loses its ability to act as a repressor. Second, we show that the origin of the ankyrin-repeat regional one determines the transcriptional activity. Two lines of in vivo experiments corroborate these results. First, HES-5 expression is reduced in transgenic mice that misexpress the Notch3 IC in the neural tube, indicating that Notch3 IC also in vivo can affect Notch1 mediated activation of HES-5 (Paper I). Second, disruption of Notch1 signalling by loss of either the ligand Delta-like1 (Dll1) or RBP-Jk results in accelerated differentiation of pancreatic endocrine cells. A similar phenotype was observed in mice over-expressing the intracellular form of Notch3 (Paper III).

These data also show that Notch3 functions to block lateral inhibition by repressing Notch1 function. In adult humans, Notch3 is expressed in VSMCs, the degeneration of which is the key pathogenic feature of CADASIL. The role of Notch3 in CADASIL has been studied in a transgenic mouse model (Paper IV). The wildtype Notch3 allele was replaced with a CADASIL (Rl40C)-mutated Notch3 gene by homologous recombination in ES cells. Mice heterozygous for the CADASIL-mutated allele have a distinct vascular phenotype with disturbed vessel wall organisation. Interestingly, some ultrastructural changes observed in CADASIL patients, GOM have not been observed in our transgenic mice, suggesting that GOM could be a secondary effect of the pathogenesis.