The Wnt pathway in ventral midbrain dopaminergic differentiation

Abstract

Parkinson’s disease (PD) is a debilitating neurodegenerative disease in which dopaminergic (DA) neurons in the substantia nigra are lost. Current treatments such as administration of levodopa (L-DOPA) are initially effective but the disease eventually progresses, highlighting the need for a treatment that restores and maintains the otherwise lost functions. Cell-replacement therapy (CRT), where DA neurons and/or precursors are grafted into the striatum in order to restore the lost striatal DA transmission, are considered a promising treatment. In order to implement CRT, large numbers of correctly specified ventral midbrain (VM) dopaminergic (DA) neurons should be obtained. Therefore, a lot of effort has been put into identifying cell sources from which high numbers of DA neurons can be generated and also into the study of the intrinsic and extrinsic factors that regulate DA neuron development. In our lab, we have focused on the Wnt pathway and its extracellular modulators which we believe to be of value for CRT as it does not imply genetic modification of the cells.

In this thesis, we have investigated the role of Wnt pathway components such as Wnt1, Wnt5a, Lrp6 and Dkk1 in VM DA neuron development and in stem cell DA differentiation.

The Wnt pathway regulates several important processes such as cell proliferation, fate determination, differentiation and patterning. The ligand Wnt1 plays a crucial role in DA VM midbrain patterning and specification: in its absence most of the midbrain fails to be formed; Wnt1 promotes the specification of the DA progenitor domain and differentiation in vivo, and the proliferation of DA precursors in vitro. Prior to this thesis, the role of the Wnt receptor Lrp6 in VM DA development and specification had not been elucidated. To determine the role of the Wnt/β-catenin pathway in DA specification, mouse embryonic stem cells (mESC) lacking the Wnt1 ligand or the Lrp6 receptor were induced to differentiate towards a VM DA phenotype using established protocols. Our results revealed that an impaired Wnt/β-catenin pathway, at the ligand or receptor level, improved neurogenesis and DA differentiation from mESC, Moreover, addition of Dickkopf 1 (Dkk1), a Wnt inhibitor, mimicked this effect in a mESC line, confirming that mESC-derived DA differentiation is improved by impairing the Wnt/β- catenin pathway and that current protocols to induce DA differentiation can be enhanced by addition of soluble factors such as Dkk1.

We next sought to investigate the role of Lrp6 in VM DA neuron development in vivo. Analysis of Lrp6-/- embryos revealed a normal midbrain patterning and a decreased differentiation of DA neurons, which later recovered. Thus, in the absence of Lrp6 there is a delay in DA neuron differentiation.

Given the results obtained with mESC, we decided to further assess the role of Dkk1 in DA development. Our results revealed that Dkk1 is expressed in the VM just prior to the onset of DA neurogenesis. In Dkk1+/- mice we could not detect any changes in the number of DA progenitors and neurons at E11.5. However, at E13.5 there was a significant reduction in the number of DA neurons. DA progenitors and precursors were not affected, there were no differences in cell death and other ventral midbrain populations were unchanged, suggesting a DA specific differentiation impairment. At later stages, we could still detect a 30% reduction in the number of DA neurons and an abnormal distribution in the VM. Analysis of the few surviving Dkk1-/- embryos at E17.5 revealed very few or almost absent DA neurons which, when present, were abnormally distributed and had a very immature morphology. Together, these results suggest a role of Dkk1 in VM DA differentiation and morphogenesis.

Finally, in order to address some of the limitations of using human fetal VM cells for CRT, we have evaluated whether a protocol that allows the generation of large and functional numbers of DA neurons from mouse VM cells could be applied to human cells. Cells were expanded in the presence of DA-appropriate factors for 2 weeks and induced to differentiate to assess their DA potential. Our results show that human fetal VM cells can be successfully expanded 2 to 3-fold, retain their VM DA identity during expansion and give rise to large and increasing numbers of DA neurons. Moreover, addition of Wnt5a, which has been shown to promote DA differentiation in several systems, also promoted a significant increase in the numbers of DA neurons after 2 weeks of expansion.

In sum, the results presented in this thesis describe new functions of the Wnt/β-catenin pathway in DA differentiation from embryonic stem cells and in vivo, a novel regulator of DA differentiation and an efficient protocol for expansion and differentiation of human fetal VM cells. We believe this knowledge can be successfully applied and improve current and future stem cell therapies in PD.