Chinese Medicine PaBing-II Protects Human iPSC Derived Dopaminergic Neurons from Oxygen Stress


 BackgroundPaBing-II Formula (PB-II) is a traditional Chinese medicine developed to treat Parkinson's disease (PD). However, due to the complexity of PB-II and the difficulty of culturing human dopaminergic neurons (DAn) in vitro, the mechanism of PB-II to treat PD remains unclear. MethodsWe established the human induced pluripotent stem cells (iPSCs) and derived DAn from hiPSCs to study the protective effects of PB-II on DAn after oxidative stress, which plays an important role in PD pathogenesis. ResultsWe found that serum derived from rats that had ingested PB-II significantly protect hiPSC-derived DAn from reactive oxygen species (ROS). In addition, PB-II dependent serum can activate nuclear erythroid-derived factor 2 (Nrf2) responses, which are required for the neutralization of ROS. In addition, PB-II can activate the Nrf2/ARE signal pathway of midbrain dopaminergic neurons of PD rats induced with 6-hydroxydopamine (6-OHDA) injury, rescue DAn cells, and improve the symptoms of PD rats. ConclusionsPB-II significantly protects the DA neurons from oxidative stress by activating the Nrf2 pathway.

Acorus 5 g, Gastrodia 10 g, Tortoise plate 10 g, and Roasted licorice 3 g. Wumei nourishes kidney, liver, blood, softens tendons, and balances Yin and Yang to treat the symptoms of involuntary tremor. PaBing-II Formula has signi cant therapeutic effects on PD patients, especially during the early stage [7,8]. PB-II can relieve PD patients' motor and non-motor symptoms, improve the therapeutic effects of dopaminergic drugs while reducing drug's side effects, and improve patients' quality of life [9,10]. Previous studies report that gavaging PD rats induced by 6-hydroxydopamine (6-OHDA) with PB-II improve the rotational behavior [11], and protected the dopaminergic neurons from apoptosis [12,13].
The 6-OHDA can damage the dopaminergic neurons in substantia nigra of midbrain, and lead to the symptoms of PD in mice. The unilaterally lesioned 6-OHDA-lesioned rat model of PD has proved to be invaluable in advancing our understanding of the mechanisms underlying parkinsonian symptoms, and is widely used in PD research [14]. However, the impact of PB-II on human DAn and mechanisms underlying its therapeutic bene t remain to be established.
Based on the pluripotency of iPSCs to differentiate into all cell types in the body, the iPSC technology has provided an essential model to study human diseases in dishes [15]. Here, we generated iPSC from human broblast and differentiated them into dopaminergic neurons. Using this model, we investigated the mechanisms underlying the therapeutic bene ts of PB-II on PD patients.

PB-II medicated serum preparation
For PB-II original recipe, refer to previous reports. We prepared slices of Chinese crude drugs from the pharmacy of Guangdong Province Hospital of Chinese medicine, which were decocted twice with 10 and 8 times of water, ltered, and concentrated with water bath at 80℃ to 1.6 kg/L (measured by rude drug weight/volume). Rats took gavage of PB-II twice a day at 32 G•kg -1 (according to the crude drug meter) according to the body surface area converted from clinical dosage. After gavage for 2 weeks, the rats with 10% chloral hydrate anesthesia were bled of arterial blood and separated for medicated serum in accordance with the method reported in the previous studies [16]. The medicated serum and control serum were inactivated at 56 ℃ for 30 min, then stored at -80 ℃ before use.

LC-MS
LC-MS was conducted through the combined application of Dionex Ultimate 3000 UHPLC (Thermo Fisher Scienti c, Waltham, MA, USA) and Q Exactive Orbitrap mass spectrometer. After screening, the analysis was carried out using the Waters TM UPLCTM HSS T3 C18 (2.1×100 mm, 1.7μm). Chromatographic conditions: Gradient elution was performed with acetonitrile (A) -0.1% formic acid water (B). The elution procedure was: 0 min, 10% A; 5 min, 20% A; 20 min, 60% A; 25 min, 90% A; 28 min 90% A; 29-33min 10% A. The ow rate was 0.2 mL/min. Mass spectrometry conditions: The samples were ionized by ESI ion source and then analyzed by Q ExactiveOrbitrap high-resolution mass spectrometer. The main parameters of the ESI ion source are: spray voltage 3500V (Anion voltage -3500V), capillary temperature retention time of 0.5-29 min was selected by the automatic switching valve for mass spectrometry analysis.

Generation of human DA neurons from iPS cells
We selected an iPS cell line from a healthy human skin cell line preserved in our laboratory. According to the previous reports [17], we differentiated the DA neuron from the iPSC with the necessary media and related cytokines. iPSCs were plated at 4×10 4 cells/cm 2 on Matrigel (BD)-coated tissue culture dishes for differentiation. N2B27-CDM + bFGF (20ng/ml) was used to culture for 3 days. The differentiation was performed in the KSR medium. On day 3, the cell should be almost con uent (over 80%). The culture was gradually changed to the N2 medium, supplemented at day 0-5 with SB431542 (5 μM)+LDN-193189 (100 nM)+Iwp2 (1 μM) to get retinal progenitor, and then split in 1:3 ratio for the next six passages using Accutase. Neural induction media supplemented with 3 μM CHIR99021 and 2 μM on X-ray inactivated MEF feeders or Matrigel-coated plates was used to culture the cells. On day 6-10, induction factors were withdrawn, meanwhile adding PD173074 (0.2 μM)+DAPT (10 μM) for retinal ganglion cell inductions. On day 10-15, the medium was changed to N2, B27 and 300 mμg/mL cAMP (Sigma-Aldrich) adding 100 ng/mL SHH (C24 ) and 100 ng/mL FGF8b. We then added 10 ng/mL BDNF, 10 ng/mL GDNF, 10 ng/mL IGF-1, 1 ng/mL TGF-β, and 0.5 mM db-cAMP and continued culturing cells for 30 days. We collected the cell and detected the cells with dopaminergic neuronal markers, such as TH and TUJ1. We used primary antibodies, which were as follows: Tyrosine Hydroxylase Antibody (CST-2791), β3-Tubulin (TU-20) Antibody (CST-4466), DAPI (Sigma-D9542). The primers of DA neuron-speci c genes are shown in the following table.
The immuno uorescence method Cells were xed using 4% v/v paraformaldehyde (Alfa Aesar), washed three times with PBS containing 0.2% v/v Tween (PBST) (Fisher Scienti c), and permeablized using 0.15% v/v TritionX-100 (Sigma-Aldrich) in PBS for 1 hr at 25 ℃. After gentle removal of PBST, cells were incubated with the primary antibody in PBST overnight at 4 ℃. After that, cells were washed three times with PBST and stained with the secondary antibody for 1 hr at 37 ℃. The cells were washed three times in PBST, stained with DAPI, and viewed with a Laser scanning confocal microscope (Carl Zeiss-710). Dopaminergic neuron-speci c TH antibody, TUJ1 staining, To observe the ratio of dopaminergic neurons, we statistically analyzed the proportion of TH+/TUJ1+ double positive in all the cells.
The establishment of DA neurons oxidation model Cultured DA neurons were treated with 100 μM H 2 O 2 for 12 h in accordance with the method previously reported [18]. After the treatment, we carried out examinations of apoptosis and the ROS levels in the cells. By ow cytometry analysis, we found that ROS and apoptosis increased signi cantly. And the IF data showed that the ratio of TH/TUJ1 double-positive cells in H 2 O 2 treated cells decreased signi cantly, which is considered to be a DA neuron model of oxidative damage.

Experimental grouping
The neural cell culture and DA neuronal cells were randomly divided into 4 groups, including the control group (Ctrl), oxidative damage model (ODM), blank serum group (BS), and medicated serum group (MS).
Control cells were cultured in normal conditions, and the other cultures were treated separately. We added 10% mock serum in the BS group and 10% medicated serum in the MS group but continued normal condition without supplementing the model sample for 24 hours. On the following day, the 3 treatment groups, including the BS group, MS group, and the ODM group, were treated at 100 μM H 2 O 2 for another 12 hours. Finally, all cell samples were examined for cell apoptosis, DA neuronal activity, ROS, and Nrf2 signal pathway gene expressions.

Flow cytometry analysis
We used ow cytometry to analyze the TH positive ratio, apoptosis, and ROS levels. For TH detection, we used intracellular staining. The cells were digested, xed with 4% paraformaldehyde, and blocked with BSA. The TH-antibody (ab75875, 1/100 dilution) was then for 30 min at 22ºC. The secondary antibody used was DyLight-488 goat anti-rabbit IgG (H+L) (ab96899) at 1/500 dilution for 30 min at 22ºC.
Acquisition of >5,000 events was performed. For apoptosis detection, we used the KEYGEN apoptosis kit (#KGA108-1) according to the manufacturer's instructions. We also checked Annexin V-FITC/PI staining through ow cytometry with software (BD). ROS were detected with the Reactive Oxygen Species kit (#KGT010-1) according to the manufacturer's instructions. The brief principle for the detection of ROS was based on the uorescent probe DCFH-DA. Intracellular ROS can oxidize non-color DCFH into uorescent DCF. Thus, ow cytometry could be used to detect the uorescence intensity for ROS levels.

Nrf2/ARE signal detection
Western blotting was carried out for the detection of Nrf2 protein in each group of cells. ImageJ software was used to analyze the protein gray value. The RT-PCR detection was for Nrf2 downstream gene mRNAs, such as HO-1, NQO1, MRP2, and GPX2. The primer sequences are shown in the following table.
Quanti cation and statistical analysis Data are represented as mean ± SEM unless otherwise indicated, and Student's t-test was used for comparing two groups. F-test was used for comparing variances. For comparing multiple groups, oneway ANOVA or two-way ANOVA were used. n was indicated in gure legends. GraphPad Prism 5 software was used for statistical analysis. Differences between two groups were considered signi cant when the P-value was less than 0.05 (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001; n.s., not signi cant. Error bars indicate means ±s.d.).

Results
Quality control test of medicated serum We used liquid chromatography and mass spectrometry (LC-MS) method to analyze the active ingredients in the medicated serum, and analyze the main six compounds in the whole recipe (citric acid  (Figure 1-6)). Perform identi cation and simultaneous detection to provide a reference for the quality control of PB-II. The detailed map is shown in Figure 1.

Generation of hiPSC and their differentiation into DAn
We derived an iPSC line from healthy human skin broblasts as previously described. Brie y, we used a Nucleotransfection Kit (P2 primary cell 4D-Nucleotector X Kit, Lonza) to transfect the Y4 episomal vectors  (Figure 2A-C). These data con rmed the pluripotency of the hiPSCs.
Using the protocol described previously (Li W et al., 2011, PANS), we differentiated hiPSCs into dopaminergic neurons. We con rmed the presence of dopamine neurons in the differentiating culture using DAn-speci c markers. DAn-speci c genes such as Tyrosine hydroxylase , TUBB3 (TUJ1), FOXA2, and Engrailed 1 (EN1) were signi cantly increased in iPSC-derived DAn culture ( Figure 2D). In addition, the cell morphology con rmed the morphological characteristics of DA neurons ( Figure 2E). Immuno uorescence (IF) analysis demonstrated that about 60% of iPSC-derived cells expressed the neuron-speci c markers TUJ1, and about 40% of iPSC-derived neurons were TH + /TUJ1 + , con rming the presence of hiPSC-derived DAn ( Figure 2F). .

PB-II protects DA neurons from oxidative damage induced by H 2 O 2
Immuno uorescence data showed more TUJ1 + TH + neurons in the PB-II medicated serum (MS) sample than those in the oxidative damage model sample (ODM) which was prepared by H 2 O 2 damage and blank serum (BS) samples ( Figure 3A). While TUJ1 + TH + neurons were obviously decreased in the ODM group after the treatment with H 2 O 2 , MS could signi cantly increase TUJ1 + TH + neurons, indicating that MS could protect TUJ1 + TH + neurons from oxidative stress ( Figure 3A). In support of this conclusion, ow cytometric analysis indicated that the percentage of TH + cells in the MS group was signi cantly higher than that in the ODM or BS group ( Figure 3B, 3C). We further analyzed the percentage of apoptosis in each experimental group, indicating that MS protected the neuronal apoptosis after H 2 O 2 treatment ( Figure 3D, 3E). PB-II activates the Nrf2/ARE signaling pathway and reduces cellular ROS In order to explore the mechanism of PB-II to protect DAn from oxidative stress neurons, we examined the ROS levels of hiPSC-derived neuronal culture after various treatments. While the ROS levels were similar between the ODM and BS treatment groups, MS signi cantly decreased cellular ROS in hiPSCderived neuronal culture, supporting a role of PB-II in reducing oxidative stress ( Figure 4A, 4B). Together, these ndings support the notion that PB-II can protect DAn from oxidative stress by reducing cellular ROS levels.
As reported before, the ROS triggers the redox system by activating Nrf2, which induces its downstream genes such as HO-1, NQO1, MRP2, and GPX2. After the treatment with H 2 O 2 (100μM, 12h), Nrf2 protein levels and its downstream gene expression was statistically similar to those of the Ctrl group ( Figure 4C-F). However, MS treatment signi cantly increases the expression of Nrf2 protein and its downstream genes such as NQO1 ( Figure 4C-F). Therefore, PB-II can activate the Nrf2/ARE signaling pathway to protect DAn from oxidative stress.

PB-II improves the symptoms of PD rats by activating the Nrf2/ARE signaling pathway
To further validate the ndings that PB-II can activate the Nrf2/ARE signaling pathway to protect DAn from oxidative stress, we tested the effects of PB-II on PD rat models by injecting 6-OHDA into the substantia nigra striatum of rats to induce the death of midbrain DAn (Model group). The Sham operation group was injected with the same volume of normal saline (Sham group). The PB-II group was given 32g / kg of PB-II by gavage. During the treatment course of 4 weeks, the behavioral symptoms of PD rats were measured weekly. The spinal behavior in PD rats was induced by the subcutaneous injection of APO in the back of the neck, and the number of rotations was recorded within 30 minutes. During the initial stage of treatment (0 weeks), the rats in the Ctrl group and the Sham group had no symptoms of insitu circles. However, the model group and PB-II group showed serious rotationary behavior with more than 210 rotations in 30 minutes, and there was no signi cant difference between the two groups. During the third week of the treatment, the number of rotations of rats was signi cantly reduced in the PB-II group ( Figure 5E). After 4 weeks of treatment, we euthanized the rats and obtained the tissues of the nigrostriatal region. The number of TH + neurons in the substantia nigra striatum of rats in the PB-II group was signi cantly higher than those in the Sham group and Model group ( Figure 5A). In addition, the levels of Nrf2 protein in midbrain dopamine neurons of rats in the PB-II group were signi cantly higher than those in the Sham group and Model group ( Figure 5B and Figure 5C). The expression of HO-1, NQO1, MRP2, and GPX2 in the midbrain dopaminergic neurons of the PB-II group was also signi cantly increased, indicating the activation of the Nrf2/ARE signaling pathway by PB-II ( Figure 5D). These data con rm that PB-II activates the Nrf2 / ARE signal pathway in the midbrain of PD rats by activating Nrf2 to reduce oxidative stress in DAn, and thus protecting DAn from oxidative stress induced apoptosis.

Discussion
It is commonly believed that oxidative stress can eventually lead to the death of dopaminergic neurons [19][20][21], and the activation of the endogenous antioxidant system may protect the cells from oxidative damage, which is a research hotspot at present [21]. Multiple studies in various organs have con rmed that the Nrf2-antioxidant response element (ARE) pathway can play a role of endogenous antioxidant to antagonize the oxidative stress injury [22]. In the central nervous system cells, such as dopaminergic neurons, astrocytes, and microglia, Nrf2 maintains the redox balances through the up-regulation of antioxidant gene expression [23]. Previous researches have veri ed that Nrf2 mostly translocates to the nucleus in the dopaminergic neurons in the substantia nigra of PD patients, while it is present in the cytoplasm in the matched normal control group of the same age [16,24,25]. Besides, studies have also demonstrated that overexpression of Nrf2 can reduce the damage of 6-OHDA in dopaminergic neurons [19,26]. Under physiological condition, Nrf2 protein expression levels were low in cells, mainly in the cytoplasm, where it can interact with Kelch-like ECH associated protein-1 (Keap1) [27,28]. When the occurrence of oxidative stress, Nrf2 phosphorylation, and Keap1 protein translocation into the uncoupling combine with ARE in the nucleus, regulation on downstream target genes, such as Heme Oxygenase-1 (HO-1) and NAD(P)H quinone dehydrogenase 1 (NQO1), is induced, to enhance the process of detoxi cation and antioxidant ability of cells [29][30][31]. In vitro experiments have also shown that the upregulation of HO-1 and NQO1 can protect cells against oxidative damage of glutamic acid, hydrogen peroxide, and amyloid beta-protein [32][33][34].
Our experiments showed that the PB-II medicated serum could effectively reduce ROS levels in the oxidation model of dopaminergic neurons, protecting dopaminergic neurons from apoptotic death. These results suggested the Nrf2/ARE pathway-mediated antioxidant mechanism might play the role of the PB-II in treating this neurodegenerative disease. When neurotoxic substances, such as H 2 O 2 , are transported by the dopamine transporter system into neuron cells, and induce oxidative stress and increase DA neuron apoptosis. Compared with the Model group, the Medicated serum group had highly up-regulated nuclear protein Nrf2 and its downstream HO-1, NQO1, MRP2, and GPX2 expression. The results suggest that PB-II plays a protective role in the oxidative stress in neurons through increased nuclear accumulation and phosphorylation of Nrf2, as well as the expression of Nrf2 downstream target genes. PB-II contains 14 kinds of Traditional Chinese Medicine compounds. Although the complex compound composition of these natural Chinese herbal medicine is not very clear, its function has been signi cantly observed and con rmed by a lot of clinicians in the clinical practice for many years [9,10,35,36]. Furthermore, several researchers have reported its protective effects on midbrain dopaminergic neurons against 6-OHDA toxicity in substantia nigra in rat models [11][12][13]. PB-II reduces the apoptosis of DA neurons in the PD rats model, promotes cell regeneration, and nally plays a role in improving rats' PD symptoms. This study may provide further evidence in iPSC-derived DA neurons and elaborate reversal of   PB-II protects PD rat DA neurons by activating the Nrf2/ARE signaling pathway. A. Immunohistochemical results showed that the TH-positive neurons in the substantia nigra striatum of the PB-II group were signi cantly increased compared with the Sham and Model groups (the scale is 120μm). B, C. Western blotting detection results showed that the expression of Nrf2 protein in cells of the nigrostriatal region of the midbrain in the PB-II group of rats increased signi cantly (n=3, *P≤0.05). D. RT-PCR results showed that the expression of Nrf2 downstream genes in the substantia nigra tissues of rats in the PB-II group was signi cantly increased, suggesting the activation of the Nrf2 signaling pathway of PB-II on DAn cells under 6-OHDA toxicity (n=3, **P≤0.01). E. Calculate the number of rotations of the rats in each experimental group within 30 minutes. From the third week, the number of rotations of the PD-rats in the PB-II group decreased compared with the Model group (n=5, * P≤0.05) and reached a very signi cant difference in the fourth week (n=5, ** P≤0.01).