Birth dating of midbrain dopamine neurons identifies A9 enriched tissue for transplantation into Parkinsonian mice

Abstract

Clinical trials have provided proof of principle that new dopamine neurons isolated from the developing ventral midbrain and transplanted into the denervated striatum can functionally integrate and alleviate symptoms in Parkinson's disease patients. However, extensive variability across patients has been observed, ranging from long-term motor improvement to the absence of symptomatic relief and development of dyskinesias. Heterogeneity of the donor tissue is likely to be a contributing factor in the variable outcomes. Dissections of ventral midbrain used for transplantation will variously contain progenitors for different dopamine neuron subtypes as well as different neurotransmitter phenotypes. The overall impact of the resulting graft will be determined by the functional contribution from these different cell types. The A9 substantia nigra pars compacta dopamine neurons, for example, are known to be particularly important for motor recovery in animal models. Serotonergic neurons, on the other hand, have been implicated in unwanted dyskinesias. Currently little knowledge exists on how variables such as donor age, which have not been controlled for in clinical trials, will impact on the final neuronal composition of fetal grafts. Here we performed a birth dating study to identify the time-course of neurogenesis within the various ventral midbrain dopamine subpopulations in an effort to identify A9-enriched donor tissue for transplantation. The results show that A9 neurons precede the birth of A10 ventral tegmental area dopamine neurons. Subsequent grafting of younger ventral midbrain donor tissue revealed significantly larger grafts containing more mitotic dopamine neuroblasts compared to older donor grafts. These grafts were enriched with A9 neurons and showed significantly greater innervation of the target dorso-lateral striatum and DA release. Younger donor grafts also contained significantly less serotonergic neurons. These findings demonstrate the importance of standardized methods to improve cell therapy for Parkinson's disease and have significant implications for the generation and selectivity of dopamine neurons from stem cell based sources.

Introduction

Parkinson's disease (PD) is characterized by the progressive degeneration of midbrain dopaminergic (DA) neurons. To date, pharmacotherapy using l-dihydroxyphenylalanie (l-DOPA) or dopamine agonists remains the most effective therapy for restoring dopamine transmission, however chronic treatment results in waning efficacy due to disease progression and the development l-DOPA-induced dyskinesias (LID) (Winkler et al., 2005). In contrast, cell replacement therapy (CRT) offers the prospect of long-term symptomatic relief, whereby grafted DA neurons replace degenerating endogenous DA neurons. While proof of principle for CRT has been achieved in clinical trials, demonstrating the ability of new neurons to functionally integrate and alleviate motor symptoms, the outcomes of these trials have been highly variable. It is well recognized that lack of standardization of an inherently variable donor cell source is likely to be an important contributing factor.

In order to restore dopamine transmission and motor function, research has shown that the grafted neurons must possess the properties of ventral midbrain (VM) DA neurons, enabling them to re-innervate the appropriate striatal target (Fricker-Gates and Dunnett, 2002). However dissections of VM fetal tissue represent a mixed population of neurons including DAergic and GABAergic neurons as well as the frequent presence of serotonergic neurons from the adjacent ventral hindbrain. Furthermore, the DA composition of VM tissue will comprise progenitors for different subtypes of midbrain DA neurons. The developed midbrain includes three major sub-populations of neurons: the A8 DA neurons of the retro-rubal field (RRF), A9 neurons of the substantia nigra pars compacta (SNpc), and the A10 neurons of the ventral tegmental area (VTA) (Dahlstrom and Fuxe, 1964). These subclasses of DA neurons can be readily identified and categorized based upon their morphology, location within the midbrain and efferent projection patterns (Bjorklund and Dunnett, 2007). The A8 neurons are the most posteriorly localized VM DA population that innervate both limbic and striatal areas, as well as providing local DA innervation within the midbrain. The A9 neurons, positioned lateral to the midline and identified by the expression of G-protein-gated inwardly rectifying K⁺ channel subunit 2 (Girk2), project via the nigrostriatal pathway to the dorso-lateral striatum and regulate motor function. A10 neurons are the most medial population of VM DA neurons, co-express Calbindin, and project via the mesocorticolimbic pathway to cortical and limbic structures to regulate reward-related behavior.

In PD, A9 DA neurons are the first to degenerate and contribute to the various motor anomalies, while adjacent A8 and A10 DA neurons are relatively spared early in the disease (Agid et al., 1993, Brooks et al., 1990). More specifically, the ventral tier of the SNpc and ventrolateral VTA, identifiable by labeling for the Pitx3 target gene aldehyde dehydrogenase (Ahd2), are the most susceptible (Jacobs et al., 2007, McCaffery and Drager, 1994). Hence, it is not surprising that recent studies have shown that the A9 DA neuronal composition of VM grafts are responsible for the functional benefits following transplantation (Grealish et al., 2010, Kuan et al., 2007, Mendez et al., 2005, O'Keeffe et al., 2008). Fetal grafts from mice lacking A9 neurons showed poor striatal innervation and functional recovery compared to comparative control grafts containing both A9 and A10 DA neurons (Grealish et al., 2010). In support, neural stem cell-VM co-grafts, enriched in Girk2 + DA neurons, improved motor function (O'Keeffe et al., 2008). In addition to restoring motor deficits in PD models, A9 neurons present within grafts have also been implicated in alleviating LID, by normalization of striatal protein levels of cyclin-dependent kinase 5 (Cdk5) and dopamine- and cAMP-regulated phosphoprotein of 32 kDa (DARPP-32) (Kuan et al., 2007).

To date, clinical trials in PD patients have utilized human fetal VM tissue ranging from 6 to 10 weeks of gestation. Additionally, the current ‘gold-standard’ for rodent transplantation studies is to isolate VM tissue from E12.5 mice (or the equivalent E14.5 in rats), perceived to be the peak of DA neurogenesis. However the relative contribution of the different subpopulations of DA neurons during these ages remains unknown. Given the important ability of A9 neurons within VM donor material to alleviate motor symptoms and LID, efforts to enrich for this subpopulation of DA neurons may improve the efficacy and consistency of clinical outcome after transplantation. In this regard, we set out to enrich for A9 neurons within fetal grafts by identifying the birth dates of the various VM DA subpopulations during mouse development.