Soluble delta-like 1 homolog (DLK1) stimulates angiogenesis through Notch1/Akt/eNOS signaling in endothelial cells.

AIM: Delta-like 1 homolog (DLK1) is a non-canonical ligand of Notch signaling, which plays a pivotal role in vascular development and tumor angiogenesis. This study aimed to elucidate the function and mechanism of DLK1 in angiogenesis. METHODS AND RESULTS: By using in situ hybridization and immunohistochemical studies, expression analysis revealed a unique vascular tropism of DLK1 in vasculature of neuroblastoma and vascular tumors. Thus, it was hypothesized that DLK1 may be cleaved and then bound to endothelial cells, thereby regulating the endothelial function. To test such hypothesis, soluble DLK1 encompassing DLK1 extracellular domain (DLK1-EC) was generated and validated by its inhibitory function in adipogenesis assay. Recombinant DLK1-EC exhibited the preferential binding capability toward endothelial cells and stimulated the microvessels sprouting in aorta rings. Above all, implantation of DLK1-EC dose-dependently elicited the cornea neovascularization in rats. By using various angiogenesis assays, it was delineated that DLK1-EC stimulated the angiogenesis by promoting the proliferation, motility and tube formation of endothelial cells. By immunoblot and luciferase analysis, it was elucidated that DLK1-EC enhanced the expression and activities of Notch1/Akt/eNOS/Hes-1 signaling in dose- and time-dependent manners. Pharmaceutical blockage of Notch signaling using γ-secretase inhibitor DAPT abrogated the DLK1-EC-induced endothelial migration and Hes-1-driven luciferase activities. Furthermore, Notch1 inactivation by neutralizing antibodies or RNA interference reversed the DLK1-EC-induced angiogenesis. CONCLUSIONS: The present study unveils the pro-angiogenic function and mechanism of soluble DLK1 through activation of Notch1 signaling in endothelial cells.

DLK1 Expressed in Mouse Orexin Neurons Modulates Anxio-Depressive Behavior but Not Energy Balance.

Lateral hypothalamic area (LHA) neurons expressing the neuropeptide orexin (OX) are implicated in obesity and anxio-depression. However, these neurons release OX as well as a host of other proteins that might contribute to normal physiology and disease states. We hypothesized that delta-like homolog 1 (DLK1), a protein reported to be co-expressed by all OX neurons, contributes to the regulation of energy balance and/or anxio-depression. Consistent with previous reports, we found that all rat OX neurons co-express DLK1. Yet, in mice and humans only a subset of OX neurons co-expressed DLK1. Since human OX-DLK1 distribution is more similar to mice than rats, mice are a comparable model to assess the human physiologic role of DLK1. We therefore used a viral lesion strategy to selectively delete DLK1 within the LHA of adult mice (DLK1Null) to reveal its role in body weight and behavior. Adult-onset DLK1 deletion had no impact on body weight or ingestive behavior. However, DLK1Null mice engaged in more locomotor activity than control mice and had decreased anxiety and depression measured via the elevated plus maze and forced swim tests. These data suggest that DLK1 expression via DLK1-expressing OX neurons primarily contributes to anxio-depression behaviors without impacting body weight.

The non-canonical NOTCH1 ligand Delta-like 1 homolog (DLK1) self interacts in mammals.

Delta-like 1 homolog (DLK1) is an imprinted gene, which is widely expressed during mammalian development and plays a pivotal role in differentiation of various tissue types. Most recently, we have shown that DLK1 interacts with NOTCH1, yet several Notch independent mechanisms have previously been suggested as well, but only poorly confirmed in a mammalian context. In the present study, we employed the mammalian two-hybrid (MTH) system, a genetic in vivo protein-protein interaction system, to show robust DLK1-DLK1, DLK1-FnI (Fibronectin) and DLK1-CFR (cysteine-rich FGF receptor) interactions, whereas the proposed DLK1-IGFBP1 interaction was not supported by MTH. Very little has previously been described on the DLK1 self-interaction. Herein, we showed by immunoprecipitation as well as Sulfo-SBED label transfer that the DLK1-DLK1 interaction likely is part of Dlk1's function in preadipocytes. Furthermore our data suggest that DLK1 interacts with itself through EGF domain 4 and 5, which is distinct from the recently described NOTCH1-DLK1 interaction, which occurs between EGF domain 5 and 6. This opens up the possibility that Notch independent mechanisms like the DLK1-DLK1 interaction may modulate the non-canonical NOTCH1-DLK1 interaction further complexing this system.

Evidence of non-canonical NOTCH signaling: Delta-like 1 homolog (DLK1) directly interacts with the NOTCH1 receptor in mammals.

Canonical NOTCH signaling, known to be essential for tissue development, requires the Delta-Serrate-LAG2 (DSL) domain for NOTCH to interact with its ligand. However, despite lacking DSL, Delta-like 1 homolog (DLK1), a protein that plays a significant role in mammalian development, has been suggested to interact with NOTCH1 and act as an antagonist. This non-canonical interaction is, however controversial, and evidence for a direct interaction, still lacking in mammals. In this study, we elucidated the putative DLK1-NOTCH1 interaction in a mammalian context. Taking a global approach and using Dlk1(+/+) and Dlk1(-/-) mouse tissues at E16.5, we demonstrated that several NOTCH signaling pathways indeed are affected by DLK1 during tissue development, and this was supported by a lower activation of NOTCH1 protein in Dlk1(+/+) embryos. Likewise, but using a distinct Dlk1-manipulated (siRNA) setup in a mammalian cell line, NOTCH signaling was substantially inhibited by DLK1. Using a mammalian two-hybrid system, we firmly established that the effect of DLK1 on NOTCH signaling was due to a direct interaction between DLK1 and NOTCH1. By careful dissection of this mechanism, we found this interaction to occur between EGF domains 5 and 6 of DLK1 and EGF domains 10-15 of NOTCH1. Thus, our data provide the first evidence for a direct interaction between DLK1 and NOTCH1 in mammals, and substantiate that non-canonical NOTCH ligands exist, adding to the complexity of NOTCH signaling.