Although the adult SN lacks the capacity to generate dopaminergic neurons, generation of non-neuronal cells has been robustly shown in the adult nigra [5, 6]. The origin and functional role of these cells are still under debate . These cells mainly express the neuro-glial antigen NG2, which marks oligodendrocytic precursors and also a number of cells with the potential to differentiate into neurons or microglia [19, 26, 27, 29]. The main hallmark of PD is dopaminergic neurodegeneration in the SN and the goal for potential restorative therapies in the future might be the replacement of dying tissue by potential neural precursors. We asked, if the generation and differentiation of nigral NG2+ cells depends on dopamine and if and how physical activity and environmental enrichment would influence these cell populations and their potential differentiation. In the first step we quantified all BrdU+ cells in the SNpc and SNpr. As we found that in both areas the numbers of BrdU+ cells were equally significantly increased following MPTP but the absolute numbers of these cells were very low, we did not further differentiate between SNpc and SNpr in the phenotypic analyses of BrdU+ cells. The small numbers of BrdU+ cells in the SN might be due to the usage of only one proliferation marker and its systemic application. Other groups reported the local, intrastrial application of BrdU and consecutively higher numbers of labelled cells . However, intrastriatal application requires a surgical intervention which per se already induces changes in the local microenvironment and might be responsible for changes in the numbers of newborn cells. Thus, the de facto number of BrdU+ cells in the SN might be higher than detected by our methods. The analyses of all BrdU+ cells in the SN over time showed an initial significant increase in the numbers of BrdU+ cells in MPTP-treated mice compared to controls. Within a few days the numbers decrease significantly indicating an initial transitory reactive cell proliferation following the lesion. This hypothesis is supported with the finding, that a large number of these cells were microglia at this time-point and the percentage of microglial cells of all BrdU+ cells did not play a role at any later time point or treatment. One could assume, that after an initial activation these cells either die or migrate away from the lesion site. This migration-hypothesis at the early time point in MPTP-mice would also support the finding, that in healthy control animals the numbers of BrdU+ cells remain stable over a long period of time (70 days in our study) without a significant effect of e.g. aging on the numbers of BrdU+ cells in the SN.
Regarding the phenotypic distribution we found a significant number of the newborn nigral cells co-expressing NG2. These cells were robustly regulated by physiological stimuli in healthy animals as expected from previous work [6, 14, 15]. In contrast to studies in the 6-OHDA PD rat model, we did not detect an influence of activity on the numbers of NG2+ cells of parkinsonian MPTP-treated mice. This might be mainly explained by species and animal model differences. Not till additional treatment with levodopa as dopamine rescue, nigral NG2+ cell proliferation in mice with access to physical activity or environmental enrichment was significantly enhanced. These results suggest a function for dopamine in activity-induced NG2+ cell generation. If these new cells could in the end indeed play a role in neuronal restoration following dopaminergic neurodegeneration remains to be clarified, but their reactive proliferation following MPTP and their vulnerability to physiological stimuli and dopamine suggests a potential function role in the dopamine-depleted brain.
Therefore, in the next step, we were interested if activity had also a dopamine-dependent effect on long-term maturation of NG2+ cells. The majority of the NG2+ cells in the adult brain in vivo are oligodendrocytic precursors [32, 33], but NG2+ cells in the SN were also described to maturate into microglia or in vitro into neurons and secrete neurotrophic factors as a potential endogenous neuroprotective mechanism following neurodegeneration [19, 29, 34]. The numbers of NG2+ cells in our study decreased significantly over time. This decline implicates a physiological differentiation process of NG2+ cells into mature oligodendrocytes or other resident cell types in the healthy SN. We thus quantified the numbers of BrdU+ nigral oligodendrocytes by CNPase-expression over time depending on MPTP-treatment and physiological stimulation. In the short-term MPTP- and saline-treated groups, no CNPase+ cells were detected indicating the appearance of mature cell populations rather at later time points. The first BrdU+/CNPase+ cells in the SN were detected 10 days after MPTP or saline in both groups supporting the hypothesis of an ongoing maturation process. In controls the numbers of new CNPase+ cells increased over time in line with a continuous decrease of NG2+ cell numbers. In contrast, MPTP-treated animals showed unchanged numbers of new CNPase+ cells in the SN, despite a continuous decrease of NG2+ cells as a possibly sign of a disturbed oligodendrocyte maturation process in the SN following dopamine depletion in line with previous studies showing disturbances in oligodendrocytic homeostasis following MPTP . Another reason for the decrease of NG2+ cells in MPTP mice and a lack of increase of CNPase+ cells could be the maturation of NG2+ cells in other neural cell types, but also changes due to the aging brain between the earliest and latest time points of our investigations. However, in our model we did not observe any BrdU+ cell population increasing over time associated with the NG2+ cell decrease.
The numbers of CNPase+ oligodendrocytes were not altered by physiological stimulation or levodopa rescue in any group and at any time point. As activity robustly increases functionally relevant neurogenesis in neurogenic regions of the healthy brain [35, 36], our data add somewhat to the differentiation between neurogenic and non-neurogenic regions and also show a different regulation of nigral cells following physiological stimuli and pharmacological treatment in line with comparable data from the dentate gyrus . In the SN activity regulated cellular plasticity might occur on precursor cell level, however the stimulus is not sufficient to induce or enhance the further maturation process of these cells. This might be of interest in the context of a suggested neuroprotective role for oligodendrocytes in the lesioned SN most likely by protective factor secretion .
The maturation of NG2+ cells into microglia with neuroprotective capacities has been discussed recently . Apart from a reactive general glia proliferation 3 days after MPTP, we did not detect any significant changes of Iba1-positive microglia or other glial cells in any group at any time point in this present study. To address if other neural precursors are generated in the SN in a dopamine or activity dependent manner, we applied MPTP to transgenic Nestin-GFP mice and visualised new neural precursors in vivo[38, 39]. No relevant increase in the numbers of new Nestin-GFP cells was detectable at any time point and treatment paradigm. This is in agreement with the quantitative colony-forming assay results showing no increase of isolatable NPCs following MPTP treatment for 3 days. The acute and transient increase of Nestin-GFP-cells and Iba-1 positive microglia 3 days after MPTP possibly reflects an unspecific reactive post-toxic proliferation [40–42].
After a primary reduction of TH-immunoreactivity, the numbers of immunohistochemically detectable TH+ cells in the MPTP-treated mice increased significantly over time [43, 44]. The mechanisms underlying this transient degeneration remain widely unclear. No BrdU+/TH+ neurons have been detected in our study at any time. Thus, the recovery of nigral neurons detected by TH after MPTP administration might be due to a down-regulation of TH-expression on neuronal cells in the SN by MPTP followed by a re-expression of TH over time, rather than by de facto degeneration of the neurons or reactive proliferation of new dopaminergic neurons. Both, physiological stimuli and levodopa rescue had no effect on TH-expression in the SN at any time point. One recent report showed forced treadmill-exercise induced increase of the numbers of TH+ neurons in the mouse SN . On the other hand, exercise-induced neuroprotection, but no neurogenesis in the SN was reported by others in line with our results . Although one study reported neuronal generation in the SN following MPTP , the majority of the studies are in line with our findings of cell proliferation but not neo-neurogenesis in the SN [2, 5, 6].
Taken together, the activity-induced generation of NG2+ precursor cells in the SN critically depends on the presence of dopamine. Based on this data, additional studies will investigate the possible neuro-restorative role of NG2+ nigral cells in PD-brain in dependence of dopamine.