In this study, the pronounced reduction of TH immunoreactivity and decreased numbers of TH-immunoreactive dopaminergic neurons in the SNpc of the 6-OHDA-lesioned side were observed, demonstrating an obvious dopaminergic neuronal degeneration and complete nerve terminal denervation, which are necessary for a successful PD animal model. Our study also shows that simvastatin prevented 6-OHDA induced dopaminergic neuronal loss, strongly implying that simvastatin would provide a neuroprotective effect in PD. This result is consistent with Ghosh and Hernandez-Romero's study, which demonstrated that statins slowed down dopaminergic degeneration and may be of therapeutic benefit for PD patients 
. It was shown that, in the EPM, 6-OHDA lesioned rats spent less time in the open arm and an obvious decrease in the entries into the open arm compared to the controls (), reflecting 6-OHDA lesion-mediated anxiety-like behaviour. Our result is consistent with Tadaiesky and Espejo's studies demonstrating that 6-OHDA lesioned PD rats showed increased anxiety-like activityes 
. Increasing evidence indicates that before the motor features occur, Parkinson's patients usually present one or more nonmotor symptoms, typically as cognitive and neuropsychiatric dysfunctions 
. Among those neuropsychiatric dysfunctions, anxiety is very common in PD patients, with prevalence rates of up to 30% depending on the criteria used 
. Therefore, it is imperative to explore the mechanisms underlying the anxiety-like activity. The current animal study directly reflects this neuropsychiatric profile in clinical PD patients and suggests possible mechanisms.
Our study showed that 6-OHDA lesion in the MFB reduced NMDA receptor expression in the brain regions examined (), which is similar to other studies, demonstrating that NMDA receptors or its subunits were decreased in the brain following unilateral dopamine depletion 
. However, how and why NMDA receptors were decreased following the 6-OHDA MFB lesion remains to be conclusively determined. Several lines of evidence demonstrated that striatal dopaminergic denervation resulted in increased afferent glutamatergic input 
; therefore, we hypothesize that the downregulation of NMDA receptors in 6-OHDA lesioned rat brain is due to increased levels of striatal glutamate following nigrostriatal dopamine denervation. Notably, we cannot preclude that the downregulation of NMDA receptors in the examined regions may reflect NMDA hypo-innervations following 6-OHDA lesion. However, the precise reasons behind this phenomenon remain to be determined.
It is well documented that NMDA receptors in the brain have a close correlation with anxiety-like activity. In NMDA NR3B (N-methyl,D-aspartate receptor subunit 3B) receptor knockout mice, pronounced decrease in activity and increase in anxiety-like behaviour were observed, suggesting that the function of the NMDA receptor directly contributes to anxiety processing 
. Similarly, Johnson and Shekhar found that anxiety-like responses in rats were regulated by the NMDA NR1 subunit and NMDA receptor antagonists 
. Our current study showed that the NMDA receptor was significantly decreased in the striatum, hippocampus, CA1 and amygdala brain regions of the 6-OHDA lesioned side. This robust downregulation of NMDA receptor in the examined brain regions of 6-OHDA lesioned rats correlated with longer duration of open-arm activity in the EPM (), strongly suggesting that the NMDA receptor hypofunction in these brain regions explains, at least partially, the anxiety-like activity in 6-OHDA induced PD rats. This hypothesis could also be supported by the facts that the altered levels of NMDA receptors in the hippocampus and amygdala directly influence anxiety behaviours 
In the current study, as our previous work and Byrnes' study 
, the elevated plus maze test was used to measure the anxiety of rats following 6-OHDA lesion and simvatatin treatment. Two indicators, the duration spent in the open arm and entries into the open arm, were applied to evaluate the anxiety of rats. Increased time, and/or entries traveled in the open arms of the EPM are interpreted as reduced anxiety-like behavior. Our data showed that when compared to 6-OHDA-lesion PD rats, simvastatin only produced an increased tendency but not significant effect in the entries into the open arms (p
0.060, ). This result may be due to either the small numbers of rats used in this study, or the rats being reluctant to move following the 6-OHDA lesion. This increased tendency in the entries into the open arms following simvastatin treatment, at least partially, indicated that simvastatin could attenuate the 6-OHDA induced anxiety. Moreover, our results () also showed that simvastatin administration profoundly increased the reduced time spent by 6-OHDA lesioned rats in the open arm of the EPM (), reflecting the ability of simvastatin to produced a pronounced anxiolytic-like effect. Consistent with our hypothesis, in a retrospective cohort investigation Starr found that statins obviously ameliorated anxiety disorder from in people aged 11–80 
. Increasing evidence shows that statins have been used clinically to restore the cognitive deficits in different neurodegenerative disorders such as PD, AD and vascular dementia 
, and the cumulative reduction in the levels of anxiety risk for patients is independent of the statins' cholesterol-lowering effect 
. However, how statins affect anxiety and the underlying mechanisms remain unclear. This study showed that the down-regulation of NMDA receptors in these examined regions was obviously restored following simvastatin administration. The present study is consistent with our previous observation in which simvastatin upregulated NMDA receptors in the naïve rat brain, and further validates our proposal that simvastain may exhibit NMDA antagonist-like effects 
. Our results demonstrated that the upregulation of NMDA receptors in the hippocampus, CA1 and amygdala following simvastatin treatment had a significant positive correlation with the time spent in the open arm of the EPM (), implying that simvastatin ameliorated anxiety behaviour in 6-OHDA lesioned rats via NMDA receptor modulation. Because previous studies have found that simvastatin affected dopamine levels as well as its metabolism in vivo 
, and because there exists a close interaction between the regulation of NMDA receptors and the dopaminergic system 
, it is reasonable to speculate that simvastatin may exhibit an anxiolytic-like activity in 6-OHDA-lesioned rats by modulating the expression of NMDA receptors in the examined brain regions or influencing the interaction of NMDA receptors and the central dopaminergic system.
To explore the effects of simvastatin on PD in an in vitro model, 6-OHDA treated PC12 cells, an accepted PD in vitro model, were used in this study. The 6-OHDA incubated PC12 cultures exhibited an obvious decrease of cell viability and increased apoptosis (), indicating the establishment of a successful in vitro PD model. However, pre-incubation with simvastatin reduced cell viability and increased apoptosis, as determined using Hoechst 33342 and flow cytometry analysis. In addition, our results showed that LDH and glutamate were significantly increased in 6-OHDA-induced PC12 cells. These elevations were obviously prevented after simvastatin incubation, demonstrating that simvastatin induced pronounced neuroprotective effects. PC12 cells mainly express the functional NR1 receptor; therefore NR1 was chosen to detect the effects of 6-OHDA neurotoxicity and simvastatin in this study. It has been shown that the elevation of NMDA receptors is closely correlated with inflammatory responses and induced neuronal death 
. In the current study, the increased NR1 expression and excitatory glutamate concentration were observed following 6-OHDA incubation ( and ). This 6-OHDA induced elevation of glutamate excessively activated NMDANR1 expression, which further aggravated PC12 damage 
and may have increased the susceptibilityof PC12 cells to excitotoxicity. However, the addition of simvastatin significantly abolished this elevation of NR1 and glutamate as well as the reduction in PC12 cell death. Considering that the elevation of NR1 and glutamate will lead to excitotoxicity and neuronal cell death, it is reasonable to speculate that in the current study simvastatin prevented PC12 cell death, at least partially, by protecting against NR1-induced excitotoxicity. This result is similar to Wang's study, showing that the upregulation of NR1 was correlated with neuronal cell death and abolishing this NR1 elevation prevented neuronal loss 
. Interestingly, we observed that the changes of NMDA receptors following 6-OHDA and simvastatin treatment in vivo and in vitro PD models are contrary. These contrasting results may be that in vivo PD model the animals responded with auto-regulation to dopaminergic damage; while in vitro PD model only PC12 cells react to micro-environment changes following 6-OHDA and simvastatin treatment. However, the precise mechanisms need further study.
To explore whether inflammatory mediators in PC12 cells changed following 6-OHDA and simvastatin treatment, we measured the expression of TNF-a and MMP9. Our study showed increased expression of TNF-a and MMP9 in 6-OHDA-induced PC12 cells (), implying that these inflammatory mediators affected NMDA receptors expression. The elevation of NR1 and TNF-a and MMP9 was significantly abolished following simvastatin treatment, strongly suggesting a direct anti-inflammatory property of simvastatin through NMDA receptor modulation. The current result is consistent with several lines of evidence showing that the regulation of NMDA receptors is directly correlated with inflammatory mediators TNF-a and MMPs in pathological brain processes, including the mediation of neuronal death 
. To further verify that the alteration of NMDA receptors is associated with inflammatory cytokine TNF-a, we focused specifically on 6-OHDA-treated PC12 expressing NR1 protein and analyzed the pattern and distribution of the punctate extranuclear immunostaining of TNF-a proteins presenting along dendrites. We detected a significant increase in NR1 protein clusters after 6-OHDA exposure; this increase was abolished following simvastatin treatment, whereas TNF-a proteins displayed a similar pattern after 6-OHDA neurotoxicity and simvastatin treatment (). The changed trend of TNF-a and NR1 proteins in our study () indicated that NR1 proteins were closely associated with inflammatory cytokine TNF-a following 6-OHDA and simvastatin treatment. This result is consistent with other studies showing that pro-inflammatory mediator TNF-a is involved in simvastatin-mediated neuroprotection and associated with the altered expression of NMDA receptors 
. To the best of our knowledge, this is the first attempt to describe the TNF-a and NR1 in PC12 and their similar changes in expression following inflammation.
In summary, our study presents the first evidence demonstrating the effects of simvastatin on NMDA receptors in the brain of 6-OHDA-lesioned rats and reveals an NMDA-modulatory effect, providing an exciting new paradigm to ameliorate anxiety-like activity in PD. Based on the current results, we reasonably speculate that the improvement in anxiety-like activity due to chronic treatment with simvastatin in 6-OHDA-lesioned rats is partially correlated with a reversal of the declined in NMDA receptors expression. Through in vitro and in vivo studies, our results strongly demonstrated that simvastatin provided robust neuroprotection against dopaminergic neurodegeneration, partially via NMDA receptor mediated anti-inflammatory mechanisms such as regulating TNF-a and MMP9. Although it is not a complete phenocopy of human disease, this 6-OHDA-mediated in vivo or in vitro PD models provides a useful means to study the pathomechanisms of clinical PD patients, as the models recapitulates many of the hallmarks of PD. A better understanding of the roles and relationships among statins, NMDA, and the dopaminergic system may open new perspectives for the statin family in the modulation of psycho-neurodegenerative disorders such as PD.