Ellagic acid protects from rotenone-induced dopaminergic neuronal damage via activation of Nrf2 signaling in astroglia


 BackgroundParkinson’s disease (PD) is the second most prevalent central nervous system (CNS) degenerative disease. Oxidative stress is one of key contributors to the degeneration of dopamine (DA) neurons in the pathogenesis of PD. Nuclear factor erythroid-2-related factor 2 (Nrf2) is considered to be a ‘master regulator’ of many genes involved in anti-oxidant stress to attenuate cell death. Therefore, activation of Nrf2 signaling provides an effective avenue to treat PD. Ellagic acid (EA), a natural polyphenolic contained in fruits and nuts, possesses amounts of pharmacological activities, such as anti-oxidant stress and anti-inflammation. Recent studies have confirmed EA could be used as a neuroprotective agent in neurodegenerative diseases. However, the underlying mechanisms have not yet been entirely elucidated. MethodsIn this study, mice subcutaneous injection of rotenone (ROT)-induced DA neuronal damage was performed to investigate EA-mediated neuroprotection. In addition, adult Nrf2 knockout mice and different cell cultures including MN9D-enciched, MN9D-BV-2 and MN9D-C6 cell co-cultures were applied to explore the underlying mechanisms. ResultsThe present study demonstrated EA protected DA neurons against ROT-induced neurotoxicity. Furtherly, activation of astroglial Nrf2 signaling pathway participated in EA-mediated neuroprotection as evidenced by the following observations. First, EA activated Nrf2 signaling pathway in ROT-induced DA neuronal injury. Second, EA generated neuroprotection with the presence of astroglia and silence of Nrf2 in astroglia abolished EA-conferred DA neuroprotection. Third, EA failed to produce DA neuroprotection in Nrf2 knockout mice. ConclusionThis study identified EA protected against DA neuronal loss via an Nrf2-dependent manner in astroglia.


Conclusion
This study identified EA protected against DA neuronal loss via an Nrf2-dependent manner in astroglia.

Background
To date, the pathogenesis of Parkinson's disease (PD) is intangible [1]. It has been verified that the increased and prolonged production of reactive oxygen species (ROS) plays critical roles in progress of PD [2]. ROS compromises the balances between oxidant and anti-oxidant systems [3]. The anti-oxidant defense system accounted for one of the underlying mechanisms that prevent oxidative stress-induced cell damage.
Nuclear factor E2-related factor 2 (Nrf2) regulates the physiological and pathophysiological processes of various diseases [4]. Importantly, Nrf2 increases the gene expressions of endogenous anti-oxidative enzymes, such as phase II detoxifying enzymes, quinone oxidoreductase 1 (NQO1) and NAD(P)H, heme oxygenase-1 (HO-1) [5]. A number of studies confirmed that Nrf2 is located and maintained in the cytoplasm in physiological conditions via a Kelch-like ECH-associated protein 1 (Keap1)-dependent ubiquitinationproteasomal degradation [6]. Upon stimulated by oxidants or electrophiles, Nrf2 compromises Keap1/Nrf2 interactions and then Nrf2 dissociates from Keap1 complex and finally enters the nucleus. This is followed by binding to the anti-oxidant response elements (ARE), a consensus gene sequence that encodes anti-oxidant enzymes [7].
Amounts of studies identified that Nrf2 signaling participates in PD pathogenesis [8].
Activation of Nrf2 evoked dopamine (DA) neuroprotection and downregulation of Nrf2 rendered DA neurons susceptible to oxidative stress damage [9]. Insufficient Nrf2 activation was highly involved in the progress of PD [10]. Thus, potent Nrf2 signaling activation is thought to be a potential promising strategy for PD treatment.
Recently, several lines of evidence verified that EA conferred neuroprotection against aging, nerve dysfunction and neurodegeneration. In sporadic Alzheimer's disease animal model, EA generated neuroprotection and cognitive enhancement [13]. In addition, EA protected against sodium arsenate-induced neurotoxicity in rats [14]. Moreover, EA was indicated to produce neuroprotection against ischemic stroke [15]. However, the underlying mechanisms remain unclear.
In the present study, subcutaneous injection of rotenone (ROT)-elicited PD mouse model

Animals and Treatment
Wild type (WT) C57BL/6J male mice and homozygous Nrf2 knockout (Nrf2 − /− ) male mice 5 (22-28 g, 10-12 weeks) were purchased from the Nanjing University Biomedical Research Institute (Nanjing, China; Permit number: SCXK2015-0001). All animals were kept in accordance with Chinese animal health and welfare norms. This study was approved by the Animal Care and Use Committee of Zunyi Medical University (Zunyi, China). WT and Nrf2 knockout mice were randomly divided into Control, EA (100 mg/kg), ROT (1 mg/kg), ROT + EA (20 mg/kg) and ROT + EA (100 mg/kg) groups. Mice were given subcutaneous injection of ROT (1 mg/kg) 6 times a week for consecutive 5 weeks followed by daily intragastric administration of EA for consecutive another five weeks.

Rotarod Test
The rotarod test, in which animals must balance on the rod, was widely used to evaluate motor deficit in neurodegenerative disease models. Before the trial, all mice received rotarod training until they stayed on the stick at least for the cut-off time. In the test, the started speed was 10 rpm and every 30 s increased 5 rpm until mice slid off the steps.
Every mouse was tested 2 times a day and the average duration of stay on the rod was recorded.

Open Field Test
Open field test was one of the most popular ethological tests to detect the anxiety-like behavior in animals. In the present study, mice were placed on the open field and each animal was located in a separate area and mice behavioral parameters were recorded during 5 min [16]. Before each round of testing, the device should be cleaned with a 75% alcohol solution to eliminate odor interference. After the end of the experiment, the total distance of mice movement was calculated.

Immunohistochemistry Staining and Cell Counting
Mice brains were cut into 7 microns cross sections on a paraffin slicer and attached to a 6 glass slide. The brain slices were then dried and dewaxed. Fixed brain slices were separately treated with 3% hydrogen peroxide and 0.1 M citrate buffer, and blocked with goat serum. Next, it was cultured overnight with an anti-TH antibody at 4 °C.
Subsequently, brain sections were incubated with the secondary antibody working solution

Real-Time RT-PCR Assay
Total RNA in the mice midbrain and cells were separately extracted by RNAiso plus reagent and purified with RNeasy Kits. The SYBR Green PCR Master Mix was used for realtime PCR analysis. The primers used were designed by Sangon Biotech (Shanghai, China).

7
The relative differences of the target genes were first normalized to β-actin, and then calculated and expressed as a relative reduction or increase, setting the Control group at 100%.

Cell Culture and Treatment
BV-2 microglia cell lines, C6 astroglia cell lines and MN9D DA neuron cell lines were obtained from the Chinese Academy of Sciences Cell Bank (Shanghai). All cultures were maintained in DMEM/F12 media supplemented with 10% fetal bovine serum and 1% antibiotics. Three kinds of cell lines were cultured at 37 °C in a humidified atmosphere of 5% CO 2 and 95% air. Cells were seeded at 1 × 10 6

Statistical Analysis
Data analyses were performed as mean ± SEM. Statistical comparisons were analyzed using the SPSS statistical software by one-way ANOVA. Then, the Bonferroni's post hoc test was performed for all pairwise comparisons among means. p < 0.05 was considered as 9 statistically significant.

EA protected against ROT-induced DA neuronal damage
To explore the neuroprotective effects of EA on PD, we established ROT-induced DA neuronal lesion in mice and determined whether EA could attenuate ROT-induced DA neurotoxicity. First, we examined behavior changes by open field and rotarod tests. As shown in Fig. 1a and b, ROT caused an apparent reduction in total movement distance and the time mice stayed on the rod. EA exhibited a significant improvement in locomotor dysfunction. Furthermore, ROT decreased TH-positive neuronal number and TH protein expression and EA ameliorated ROT-induced DA neuronal damage ( Fig. 1c and d).

EA activated Nrf2 signaling pathway in vivo
To determine whether EA could activate Nrf2 signaling pathway, the expressions of Nrf2, HO-1, and NQO1 were detected by real-time RT-PCR and western blot assay. As shown in Fig. 2a, the mRNA expressions of Nrf2, HO-1 and NQO1 were increased in the ROT and EA groups, in which EA induced higher mRNA expressions of these genes than ROT. To further determine the effects of EA on Nrf2 distribution, the cytosolic components and nuclear fractions were collected, respectively. The cytosol marker HSP90 was used to verify the purity of nuclear protein and the nuclear fraction was not contaminated by cytosol proteins as confirmed by the absence of HSP90. As shown in Fig. 2b, compared with control group, ROT activated Nrf2 and induced Nrf2 to transfer from cytosol to nucleus.
Compared with ROT group, the increase in translocation of Nrf2 from cytosol to nucleus was more pronounced in EA group. Meanwhile, the higher protein expressions of HO-1 and NQO1 in ROT + EA group were indicated than those in ROT group (Fig. 2c).
EA targeted astroglia to protect from ROT-induced DA neuronal damage 10 To further explore which cell type EA targeted to produce DA neuroprotection, three cell culture systems including MN9D-enriched, MN9D-BV2 and MN9D-C6 co-cultures were performed using transwells. Cultures were treated with EA for 30 min followed by ROT treatment for 24 h. As shown in Fig. 3a, EA-mediated neuroprotection was indicated in MN9D-C6 co-culture system but not in either MN9D-enriched or MN9D-BV2 co-cultures.
Similarly, from TH-positive neuronal immunofluorescence staining and TH protein detection assays shown in Fig. 3b  whether EA activated Nrf2 signaling in astroglia, the mRNA and protein expressions of Nrf2 signaling in C6 cells were examined. As shown in Fig. 4b, Nrf2, HO-1 and NQO1 mRNA levels were up-regulated in ROT and ROT + EA treatment groups, whereas EA enhanced these genes mRNA expressions compared with ROT administration. In addition, western blot assay showed the translocation of Nrf2 from cytoplasm to the nucleus was more obvious after EA treatment that that in ROT group (Fig. 4c). Also, EA induced the higher protein expressions of HO-1 and NQO1 than ROT (Fig. 4d).

EA protected DA neurons by the activation of astroglial Nrf2
To explore the role of Nrf2 in EA-mediated DA neuroprotection, Nrf2 siRNA in vitro and Nrf2 knockout mice were performed. As shown in Fig. 5a, Nrf2 gene and protein levels were down-regulated in C6 cells transfected with Nrf2-siRNA with the silence ratio of 70%.
Furtherly, in TH-positive neuronal counting and TH protein expression assays, EAgenerated DA neuroprotection from ROT-induced neurotoxicity was neutralized by Nrf2-siRNA application (Fig. 5b and c).
Next, Nrf2 knockout mice were employed to confirm EA evoked neuroprotection via an Nrf2-dependent manner in vivo. First, the knockout efficiency was verified by Nrf2 protein level detection (Fig. 6a). Then, there was no significant difference of the genes and proteins expressions of HO-1 and NQO1 in the midbrain of Nrf2 knockout mice after EA treatment ( Fig. 6b and c). In addition, EA failed to improve the time mice stayed on rod and locomotor distance in Nrf2 knockout mice ( Fig. 6d and e). Meanwhile, TH-positive neuronal counting and TH protein expression detection showed EA-mediated DA neuroprotection was abolished in Nrf2 knockout mice ( Fig. 6f and g). Collectively, these results demonstrated EA protected DA neurons through the activation of astroglial Nrf2 signaling.

Discussion
The present study aimed at investigating the neuroprotective actions of EA on ROTinduced DA neuronal damage and the underlying mechanisms as well. Results demonstrated that EA conferred neuroprotection against ROT-induced DA neurotoxicity.
Activation of astroglial Nrf2 signaling was involved in EA-mediated DA neuroprotection, as evidenced by the following observations. First, EA activated Nrf2 signaling pathway in ROT-induced DA neuronal damage. Second, EA generated neuroprotection with the presence of astroglia and silence of Nrf2 in astroglia abolished EA-mediated neuroprotection. Third, EA failed to produce DA neuroprotection in Nrf2 knockout mice.
Together, these results suggested EA-generated DA neuroprotection might be attributable to the activation of Nrf2 signaling in astroglia.
Oxidative stress has been considered as the central pathological event in the development of PD [19,20]. For survival and development of neurons, the maintenance of redox homeostasis in the CNS was indispensable [21]. Nrf2 was an oxidative defense transcription factor which mitigated the toxic neuronal effects of parkinsonian agents, such as ROT, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and 6-hydroxydopamine (6-OHDA) in vivo and in vitro [ 22]. Here, this study found that EA protected DA neurons, activated Nrf2 signaling pathway and initiated the Nrf2 stressor mechanism for its transfer from cytosol to the nucleus (Figs. 1 and 2). Further study indicated that EA-elicited neuroprotection were abolished in Nrf2 knockout mice (Fig. 6). These results suggested that EA-enhanced DA neuronal survival was dependent on Nrf2 activation.
Encouraged by the activation of Nrf2 signaling by EA, we further investigated Nrf2 signaling in which cell type participated in EA-mediated neuroprotection. It is interesting to note that EA targeted astroglia to produce neuroprotection since this protection was just indicated in the presence of astroglia (Fig. 3). Astroglia is the most abundant cell type in the brain [23]. They interacted with neurons in various aspects of signaling transmission, immune and oxidative defense, ion and water homeostasis, metabolite supply and neuroprotection [24]. In neurological disorders, astroglia themselves formed a defense barrier and promoted neuroprotection by producing anti-oxidants against inflammation, especially for excessive extracellular release of the excitatory neurotransmitter glutamate [25]. Stimulation of sulforaphane also prolonged Nrf2mediated gene expression in astrocytes and prevented neuronal damage caused by superoxide [26]. In addition, besides anti-oxidants, astrocytes could release neurotrophic factors, such as BDNF and GDNF-mediated activation of the Nrf2 pathway in astrocytes, which increased the survival of DA neurons [27]. Recent research indicated that Nrf2-ARE pathway was preferentially activated in astrocytes, HO-1 and NQO1were strongly 13 expressed in astrocytes, sometime microglia, with more infrequent expression in neurons [28,29]. Nevertheless, Pharmacological activation and overexpression of Nrf2 in astrocytes also clearly demonstrated its therapeutic potential for PD. In the mouse model of α-synuclein mutation, the specific overexpression of astrocyte Nrf2 was regulated, which improved the movement and non-motor dysfunction of the whole CNS, and delayed the process of α-synuclein aggregation [30]. Also, Nrf2 overexpressed in astrocytes kept mice from mutated α-synuclein to protect DA neurons [31]. Additionally, transplantation of astrocytes overexpressing Nrf2 into mouse striatum protected from DA neuronal loss [32].
Therefore, this was an important strategy for the support of neuroprotective functions of astrocytes to enhance neuronal survival and improve PD symptoms.
A large amount of evidence suggested Nrf2 activators could protect neurons and decrease the accumulation of aberrant proteins in vitro and in vivo in different neurodegenerative mouse models. In clinic, melatonin was found to activate Nrf2 to rescue hippocampal bioenergetics and improve cognitive function [33]. At present, the clinical trials on Nrf2 activators for the treatment of neurodegenerative diseases were increasing [34]. For example, pinostrobin could release HO-1 through Nrf2 activation to alleviate MPTP-induced neuronal loss in SN, and relieved zebra fish behavioral deficit disorder, and provided potential neuroprotective effects in vivo [ 25]. Moreover, Glaucocalyxin B was discerned to inhibit TLR/NF-κB pathway activation and induce Nrf2/HO-1 activation and further produce neuroprotection [35]. Also, dimethyl fumarate enhanced Nrf2 activation, thereby playing an important role in the defense of MPTP-induced loss of DA neurons in the SN [36]. In short, new Nrf2-based treatments might be considered as a viable strategy for future potential. Thus, this study found EA exhibited a promising therapeutic efficacy for PD via activating Nrf2 signaling pathway. These findings suggested that modulation of Nrf2 signaling by potential candidates, such as EA, might be beneficial for PD treatment.
14 Despite this optimistic perspective for future administration of EA potential treatment for PD, most of the findings were derived from experimental models and in vitro data and therefore need be rigorously corroborated in clinical trials in future.

Conclusions
This study demonstrated that EA protected DA neurons from ROT-induced neurotoxicity by activating astroglial Nrf2 signaling. These findings suggested EA might be an exciting option for treating PD.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Acknowledgements
We are grateful to the technical support provided by Shanghai Baiqu Biomedical Technology Co., Ltd.

Funding
All experiments were supported by National Natural Science Foundation of China (no.

Author contribution
FZ conceived and designed the experiments. All the authors participated in the experiment performance and data analysis. FZ, GQW and YZW wrote, revised and checked the article. All authors revised and approved the final manuscript.

Ethics approval and consent to participate
All experimental procedures were carried out in accordance with Chinese Guidelines of Animal Care and Welfare and this study received an approval from the Animal Care and Use Committee of Zunyi Medical University (Zunyi, China).

Consent for Publication
Written informed consent for publication was obtained from all participants.