Expression and function of GDNF family ligands and receptors in the nervous system

Abstract

The search for novel neurotrophic factors promoting dopamine uptake and the survival of central dopaminergic neurons led to the discovery of a novel neurotrophic factor termed glial cell-line derived neurotrophic factor (GDNF). GDNF, neurturin (NTN), artemin (ART) and persephin (PSP) belong to the GDNF family ligands, and are potent survival-promoting factors for subsets of central and peripheral neurons. GDNF has been shown to elicit a variety of effects at different developmental stages and on several different cell types. In motor neurons as well as enteric neurons, GDNF stimulates neurite growth. The GDNF family ligands signal through the tyrosine kinase RET receptor and a glycosyl- phophatidylinositol (GPI)-anchored receptor (GFRalpha1-alpha4) that binds ligand with high affinity. RET is unable to bind GDNF on its own but can be activated in complex with GFRalpha1.

After the first report that GDNF signals trough GFRalpha1 and RET, we aimed to search for new members. By homology search, we identified and cloned two additional members; GFRalpha2 and GFRalpha3. We characterised the expression pattern of the so far identified ligands and receptors and provided the first evidence of GDNFGFRalpha1 pairing in vivo, showing that GDNF-/- mice were selectively deficient in GFRalpha1-expressing neurons in the trigeminal ganglion.

In cranial motor neuron subpopulations, we found a dynamic and subnuclei-specific expression of GFRalpha1 and GFRalpha2, which corresponded to their dependency in GDNF null mutant mice. These findings provided evidence for a role of GDNF for specific motor neuron subpopulations in vivo.

One of the most well characterized population of peripheral neurons in term of neurotrophic factor dependencies are the sensory neurons of the dorsal root ganglia (DRG). Embryonic DRGs depend on NGF for their survival and we found that although they express GDNF family receptors embryonically, they do not respond to their ligands by survival in culture, indicating that GDNF family ligands may play other roles during development that for neuronal survival. Supporting evidence for this notion has been found when looking at the GDNF family ligand- and receptor expression by in situ hybridisation in order to study their role in the embryonic development of facial cutaneous sensory innervation. A dynamic spatial and temporal regulation of gdnf, ntn, gfralpha1-alpha3 within the follicle sinus complex was found that correlated with the development of distinct subclasses of sensory nerve endings.

Because postnatal, but not embryonic, DRGs respond to their ligands in vitro, we aimed to explore the transcriptional program in embryonic and postnatal DRG in response to GDNF stimulation by oligonucleotide microarrays and real-time PCR.

The results from these experiments confirmed that GDNF plays distinct roles embryonically as compared to postnatal stages. Interestingly, one of the most consistent findings was that a number of cytoskeletal, structural and adhesion-related genes involved in neurite outgrowth were downregulated with GDNF. This lead us to examine the role of GDNF and its receptor GFRalpha1 in neurite outgrowth culture experiments. Dissociated Bax-/- DRG neurons were cultured with GDNF, soluble or bound GFRalpha1 in different combinations. The reduced or increased axonal growth of embryonic primary,sensory neurons cultured with soluble and bound GFRalpha1-protein, respectively, in the absence of GDNF, combined with our immunohistochemical data showing GFRalpha1 in Schwann cells but not axons, supports the hypothesis of terminal Schwann cells as a source of locally administered GFRalpha1. The different outcome depending on the configuration of GFRalpha1 and GDNF indicate a high degree of versatility of these molecules, pointing to a requirement for a strict control of both expression of these ligands and receptors as well as cleavage and release of GFRalpha1.