Wnt signalling in the development of ventral midbrain dopaminergic neurons

Abstract

Dopaminergic (DA) neurons in the ventral midbrain (VM) are one of the major cell types lost in Parkinson's disease (PD). Proof-of-principle exists for cell replacement therapies for PD, but wider application is halted by the unavailability of abundant sources of DA neurons. Stem cells might constitute one of these sources. However, efficient protocols promoting their specific differentiation into a DA neuronal phenotype are required. In this thesis, evidence is presented for a major contribution of the Wnt family of glycolipoproteins in the development of VM DA neurons.

In the first study, we found that Wnt-1, Wnt-3a, and Wnt-5a expression is differentially regulated during VM development and that partially purified Wnts distinctively regulate DA neuron development. Wnt-1 and Wnt-5a increased the number of VM DA neurons in rat embryonic day (E) 14.5 precursor cultures by two distinct mechanisms. Wnt-1 predominantly increased the proliferation of nuclear receptor-related factor 1 (Nurr1) positive precursors. In contrast, Wnt-5a primarily increased the proportion of Nurr1 precursors that acquired a neuronal DA (tyrosine hydroxylase (TH) positive) phenotype. These findings indicate that Writs are key regulators of proliferation and differentiation of DA precursors during VM neurogenesis and that different Wnts have specific and unique activity profiles.

In a second study, we found that VM glia, but not cortical glia, secrete high levels of Wnt-5a and increase the differentiation of Nurr1 precursors into DA neurons. Post-natal VM glial cells, including astrocytes and radial glia, were found to retain a specific regional transcriptional code and to be the source of instructive signals required by neural precursors or stem/ progenitor cells (neurospheres) to acquire a DA phenotype. Moreover, Wnt-5a and VM glia were able to induce a TH positive phenotype in cortical precursors expressing Nurr1.

The Wnt signal transduction can be modulated by secreted ligands of the dickkopf (Dkk) family, which can bind with high specificity to the low density lipoprotein receptor related protein (LRP) -5 and -6. In the third study, we investigated the role of Dkks and LRPs in the induction of a DA phenotype. We found a dynamic expression pattern in the developing rat VM, with a sequential expression of Dkk-1, Dkk-2 and LRP-6 (both peaking at the time of DA neurogenesis) and Dkk-3. Dkks distinctively regulated DA differentiation from E14.5 VM precursors.

Dkk-1, a Wnt/beta-catenin signalling inhibitor, decreased the number of TH positive neurons, while Dkk-2 increased the conversion of Nurr1 positive precursors into DA neurons. LRP-6 null mutant mice showed a delay in DA differentiation, with 50% less DA neurons at E11.5 and a 25% reduction at E13.5. Our results identified Dkk-2 and LRP-6 as key regulators of DA differentiation in the developing VM. Wnts and Dkks could thus serve as potential tools in the treatment of PD. However, Wnts are poorly soluble and act mainly as short range signalling molecules.

In our fourth study, we investigated whether intracellular canonical Wnt signalling components could modulate the development of DA neurons. Two chemical inhibitors of glycogen synthase kinase-3 beta (GSK3beta), indirubin-3-monoxime (I3M) and kenpaullone (KP), were found to stabilize beta-catenin and to increase DA differentiation in VM precursor cultures, thereby mimicking an effect of Wnts. The three to five fold increase in DA differentiation of precursor cells by GSK-3beta inhibitors suggests that such compounds may contribute to improve stem/precursor cell therapy approaches to PD.