Role of the atypical Notch ligand Dlk2, and it’s relationship with Dlk1 in mammalian neurogenesis

Abstract

Mammalian neurogenesis is the process by which the central nervous system is formed. Neural diversity is generated by differentiation from neural stem cells (NSCs). In the embryonic murine brain, neuronal and glial cells are generated in the ganglionic eminence. In the postnatal brain, neurogenesis persists in two active niches, the ventricular-subventricular zone (V-SVZ) lining the lateral ventricles and the subgranular zone of the dentate gyrus in the hippocampus. These niches remain as NSC reservoirs throughout life, ensuring continual neurogenesis. Notch signalling has been shown to play a critical role in embryonic and postnatal neurogenesis by regulating the maintenance of NSCs and cell fate decisions. Delta-like homologue 1 (Dlk1) and Delta-like homologue 2 (Dlk2) are vertebrate-specific non canonical Notch ligands. Dlk1 is an imprinted gene, generally expressed from the paternally inherited chromosome. Previous work has shown early expansion of the V-SVZ neurogenic niche in Dlk1 mutant mice in the early postnatal period with later depletion of NSCs and reduced adult neurogenesis. Despite expression of Dlk1 in the embryonic brain, embryonic neurogenesis is not compromised in Dlk1 mutant mice. Dlk2 is a paralog of Dlk1, however, it is not imprinted and very little is known about its function. According to the Allen Brain Atlas, Dlk2 is predominantly neural specific and expressed throughout the adult brain with no data on expression during embryogenesis. I hypothesise that there is functional redundancy between the two genes during neurogenesis. In particular, I propose that Dlk2 rescues a potential embryonic neurogenesis defect in Dlk1 mutants. Furthermore, I propose that Dlk2 is unable to rescue the Dlk1 mutant neurogenic phenotype postnatally and that the redundancy is therefore stage specific. This thesis aims to characterise the role of Dlk2 in mammalian neurogenesis, to assess its relationship with Dlk1 and to consider a role for functional redundancy between the two genes in both embryonic and adult neurogenesis. To address these aims, I first compared the predicted structures of DLK1 and DLK2 and compared expression patterns of both genes in the brain through development. AlphaFold analysis predicts that structurally, and unlike the more canonical Notch ligands, DLK2 is almost identical to DLK1 suggesting that they may have related functions. Expression analysis reveals that Dlk2 expression is upregulated in the brain from mid to late embryogenesis and maintained in multiple adult brain subregions in the mouse, indicating a functional role in neurodevelopment. Several Dlk2 mutant mouse lines were generated and four were selected for initial characterisation. Of these, a mouse homozygous for a 73 base pair deletion in Dlk2 was selected for further analysis since it does not express any Dlk2. Embryonic neurogenesis is altered in Dlk2 mutants with increased radial glial cells and NSCs within the ganglionic eminence of mutants. However, postnatally, normal proliferation and NSC numbers were evident in the V-SVZ. In Dlk1 mutant embryos, Dlk2 is significantly upregulated during embryogenesis but not postnatally. In contrast, in Dlk2 mutant embryos, Dlk1 is normally expressed but is significantly upregulated postnatally. These findings indicate a role for Dlk2 in mammalian neurogenesis. They suggest that Dlk2 may rescue an embryonic neurogenesis phenotype in Dlk1 mutants, while Dlk1 may protect postnatal neurogenesis in Dlk2 mutants highlighting potential functional redundancy at key stages of neurogenesis.