Paeoniflorin Alleviates H2O2-Induced Oxidative Injury Through Down- Regulation of MicroRNA-135a in HT-22 Cells

Ailing Zhai1 · Zeng Zhang1 · Xiangjuan Kong1
Received: 20 June 2019 / Revised: 27 September 2019 / Accepted: 27 October 2019
© Springer Science+Business Media, LLC, part of Springer Nature 2019

Abstract

Paeoniflorin (PF) has been reported to possess neuroprotective influences on cognitive dysfunction illness. In current research, we attempted to probe into the protective influences of PF against H2O2-induced damage and the underlying regulating mechanisms on hippocampal HT-22 cells. HT-22 cells were pretreated with PF, and then induced by H2O2. Afterwards, the influences of PF pretreatment were examined using CCK-8 assay, apoptosis assay, western blot and ROS assay, respec- tively. In addition, the expression of microRNA-135a (miR-135a) was analyzed and altered by qRT-PCR and cell transfec- tion, respectively. After overexpression of miR-135a, the effects of miR-135a mimic on cell functions were detected again. Moreover, influences of H2O2, PF and miR-135a overexpression on JAK2/STAT3 and ERK1/2 signal pathways were further investigated. Further experiments verified that PF pretreatment alleviated H2O2-induced oxidative stress through increasing cell viability, inhibiting cell apoptosis, reducing ROS generation and activating JAK2/STAT3 and ERK1/2 pathways. Besides, expression of miR-135a was declined by PF pretreatment. Whereas, miR-135a mimic abrogated the protective effects trig- gered by PF pretreatment. These results indicated that PF can alleviate H2O2-induced oxidative stress by down-regulation of miR-135a via activation of JAK2/STAT3 and ERK1/2 pathways.

Keywords Paeoniflorin · Alzheimer disease · Cell apoptosis · ROS · JAK2/STAT3 pathway · ERK1/2 pathways

Introduction

Alzheimer disease (AD) is an neurodegenerative illness featured by cognitive dysfunction and progressive memory decline with personality and behavioral abnormalities [1]. The incidence of AD has been increasing with the popula- tion ages. Previous studies indicated that the chief pathologi- cal characteristic of AD are the amyloid deposits deposition around neurons, as well as neurofibrillary tangles of Tau proteins and neuronal apoptosis in nerve cells [2]. The medi- cal nursing work makes an important contribution to the management of AD treatment. In addition to this, major- ity of studies have focused on pharmacological treatment of AD, including donepezil, galantamine and rivastigmine. However, there were still many dropouts and adverse events occurred during use of these drugs, therefore, there is still an urgent need to find new and effective medicines for stabiliz- ing or slowing decline in cognition and function [3].
Paeoniflorin (PF) is the main effective constituent of Radix Paeoniae alba. Previous studies demonstrated that PF possessed varieties of biological activities, such as ame- lioration of inflammation in osteoarthritis [4] and colitis [5], recanalization of thrombosis [6], amelioration of ath- erosclerosis [7], as well as inhibition of tumorigenesis in myeloma [8], glioma [9], gastric cancer [10] and pancre- atic cancer [11]. Besides, a previous study conducted by the nervous system [12]. Gu et al. reported that PF exhib- ited neuroprotective effect on AD via influencing Bcl-2/ Bax protein expression and inflammation in transgenic mice models of AD [13]. Moreover, Kapoor et al. demon- strated that PF exerted the neuroprotective effects against glutamate-induced neurotoxicity via the Bcl-2/Bax pathway in PC12 cells [14]. Furthermore, Li et al. and Wang et al. verified that PF attenuated Aβ25−35-induced neurotoxicity via preventing mitochondrial dysfunction in PC12 cells and SH- SY5Y cells [15, 16]. However, the underlying mechanism by which PF performs its neuroprotective effects in HT-22 cells still remains unclear.
MicroRNAs (miRNAs) are widely considered to be the potential biomarkers of AD. Literature review showed that miR-135a expression was altered in blood, several cer- ebrospinal fluid (CSF) and the brain, consistently. There- fore, miR-135a has been recognized as one of the poten- tial biomarkers for AD [17]. Recent studies have reported that miR-135a possibly regulated the expressions of crucial AD-related proteins APP and BACE-1, besides, miR-135a expression in the serum and CSF of AD patients was sig- nificantly decreased [18], which was inconsistent with the results of subsequent research [19]. Research in atheroscle- rosis demonstrated that overexpression of miR-135a could inhibit oxidative stress through decreasing reactive oxygen species (ROS) level [20]. However, whether PF works in cells through regulating miR-135a remains unknown.
This research intended to explore the protective influ- ences of PF against H2O2-induced oxidative injury on HT-22 cells, and tried to probe into the potential molecular mecha- nism. The cell viability, apoptosis and their related factors in HT-22 cells and miR-135a mimic transfected HT-22 cells were detected after H2O2 and PF treatment. The possible mechanism was uncovered by determining the expressions of key factors involved in JAK2/STAT3 and ERK1/2 path- ways. These observations may offer a new perspective for future AD treatment.

Methods and Materials

Cell Culture
Hippocampal HT-22 cells were gotten from the American Type Culture Collection (ATCC, Rockville, USA). The cells were sustained in Dulbecco’s modified Eagle’s medium (DMEM, BBI Life Sciences Corporation, Shanghai, China) as monolayers, complemented with 10% heat-inactivated fetal bovine serum (FBS, BBI Solution, Crumlin, UK). The cells were sustained in an incubator which comprising 5% CO2 at 37 ˚C. In order to avoid cell characteristics changes along with the prolonged culture time, all experimental cells used were between passages 15 and 25. To maintain expo- nential growth, trypsin/EDTA treatment was conducted with each cell suspension when subculturing every 2 days.

Cell Treatment

To induce oxidative injury, the cells were dealt with gradient concentrations of H2O2 (50, 100, 150, 200, 300 µM) for 3 hto determine the optimal concentration. The control group was dealt with fresh medium free of H2O2.
PF (purity > 99%) was purchased from the Aladdin (Shanghai, China). PF was firstly dissolved in DMEM medium to 2 mg/mL without FBS (BBI Solution, Crum- lin, UK) supplement and then was diluted to 50, 100, 150 and 200 µg/mL. Subsequently, HT-22 cells were exposed to series concentrations for determining the appropriate concentration. Afterwards, HT-22 cells were pretreated with optimal concentration of PF for 24 h and then induced by H2O2 treatment for determining the influences of PF pretreatment.

Hydrogen Peroxide Production
After PF pretreatment, the production of H2O2 in the cul- ture supernatant of HT-22 cells was measured employing Amplex Red hydrogen peroxide detection kit (Thermo Fisher Scientific, Massachusetts, USA) following the prod- uct instructions. In brief, cells were firstly incubated with PF for 24 h, subsequently the supernatant was removed and Kreb’s Ringers (HIMEDIA LABORATORIES, India) was supplied and co-incubated with cells for 4 h. Afterwards, the culture supernatant was collected and incubated with the Amplex Red horseradish peroxidase reagent at 37 °C for 4h. Finally, the fluorescence intensity of excitation/emission 530/590 nm was detected utilizing a fluorospectro photom- eter (HITACHI, Tokyo, Japan).

Cell Counting Kit‑8 (CCK‑8) Assay
The cell viability of HT-22 cells was assessed using CCK-8 (BBI Solution, Crumlin, UK) method. After stimulation, the reagent was added into the culture medium, and the mixture was maintained in an incubator comprising 5% CO2 + 95% air at 37 °C for 1 h. The absorbance at 450 nm was detected utilizing a Microplate Reader (Bio-Rad, Hercules, CA).

Apoptosis Assay
Cell apoptosis analysis was measured employing FITC- Annexin V/PI double-staining technique. Cells were firstly rinsed in phosphate buffered saline (PBS) and subsequently immobilized in 70% ethanol. Afterwards, the immobilized cells were rinsed twice in PBS and stained with PI/FITC- Annexin V with the existence of 50 µg/ml RNase A (Thermo Scientific, Massachusetts, USA), and afterwards maintained at 25 °C in dark situation for 1 h. The proportion of apoptotic cells was determined by flow cytometry and then calculated utilizing FlowJo software.

Western Blot
The proteins were taken out using RIPA lysis buffer (San- gon Biotech, Shanghai, China) in the presence of protease inhibitors (Sangon Biotech, Shanghai, China). The concen- trations of proteins extracted were detected utilizing the BCA™ Protein Assay Kit (Thermo Scientific, Massachu- setts, USA). The target proteins extracted were separated by sodium dodecyl sulfate polyacrylamide gel electro- phoresis (SDS-PAGE) and subsequently were transferred to polyvinylidene fluoride (PVDF) membranes (Solarbio, Beijing, China) for immunoblotting. After blocking in 5% bull serum albumin (BSA; Millipore) at 25 °C for 1 h, the spots were maintained overnight at 4 °C with primary anti- bodies directed against CyclinD1 (ab16663, abcam), p53 (ab131442, abcam), Cleaved-caspase 3 (ab49822, abcam), Cleaved-PARP (ab32064, abcam), t-JAK2 (ab39636, abcam), p-JAK2 (ab32101, abcam), t-STAT3 (ab137803, abcam), p-STAT3 (ab76315, abcam), t-ERK1/2 (ab17942, abcam), p-ERK1/2 (#4370, Cell Signaling Technology) and β-actin (ab8227, abcam). After rinsing with Tris-Buffered Saline Tween (TBST) 20 buffer, the PVDF membranes were maintained with secondary antibody (ab6721, abcam) labeled by horseradish peroxidase (HRP) at 25 °C for 1 h. After washing, the PVDF membrane was removed into the Bio-Rad ChemiDoc™ XRS system. Subsequently, Immo- bilon Western Chemiluminescent HRP Substrate (Millipore, MA, USA) was added encircling the membrane. The signals were detected and analyzed utilizing Image Lab™ Software (Bio-Rad, Shanghai, China).

ROS Assay
The cells were maintained with serum-free culture medium comprising 10 µM 2,7-dichlorofluorescein diacetate (DCFH- DA, aladdin, Shanghai, China) at 37 °C for 20 min in dark condition, then were plated into a 6-well plate. Subsequently, the cells were rinsed with PBS and digested with trypsin (Sangon Biotech, Shanghai, China) for sample collection. Then all samples were centrifuged and the supernatants were discarded. Precipitate was suspended in 500 µl PBS and fluorescent intensities were detected utilizing a flow cytometer (Beckman Coulter, USA).

Cell Transfection
HT-22 cells were respectively transfected with miR-135a mimic and negative control (NC) mimic, synthesized by Sangon Biotech (Shanghai, China), utilizing Lipo- fectamine 3000 reagent (Invitrogen) according to the product instructions.

Quantitative Real‑Time Polymerase Chain Reaction (qRT‑PCR)
Total RNAs were taken out from cells utilizing TRIzol rea- gent (Invitrogen) with the presence of DNaseI (Promega). The concentrations of the extracted RNAs were quantified with a Nanodrop 2000 system (Thermo Scientific, Massa- chusetts, USA). The cDNA was synthesized utilizing the MultiscribeRT kit (Applied Biosystems) and oligo (dT) primers. The reverse transcription conditions were 25 °C for 10 min, 48 °C for 30 min, and a final step 95 °C for 5 min. The relative expression of miR-135a was normalized to U6.

Statistical Analysis
The results of these experiments, each repeated for three times, were presented as the mean ± standard deviation (SD). Statistical analyses were completed utilizing SPSS 19.0 statistical software (SPSS, Inc., Chicago, IL, USA). The P-values between two groups and muti-groups were respec- tively calculated using LSD-t test and one-way analysis of variance (ANOVA). P < 0.05 was considered as statistically significant.

Results

H2O2 Exposure Induced Oxidative Damage in HT‑22 Cells
To determine the adequate H2O2 concentration for subse- quent experiments, cell viability analysis was conducted following a 0–300 µM H2O2 treatment in HT-22 cells. CCK-8 assay results suggested that cell viability was sig- nificantly decreased at 100, 150, 200 and 300 µM (P < 0.05 or P < 0.01 or P < 0.001) H2O2 treatment and exhibited as a concentration-dependent decrease. Because the cell viabil- ity was the closest to the value of IC50 when treated with 200 µM H2O2 (Fig. 1a), thus we chose this concentration to conduct the following experiments. The expressions of proliferation-related CyclinD1 and p53 were also exam- ined, and the results demonstrated that H2O2 treatment significantly decreased CyclinD1 expression (P < 0.05) whereas notably increased p53 expression (P < 0.01) com- pared with control (Fig. 1b, c). In Fig. 1d, the percentage of apoptotic cells in H2O2-treatment group was significantly higher than control (P < 0.001). Besides, the expressions of apoptosis-related proteins Cleaved-PARP and Cleaved- Caspase-3 was detected. The results revealed that both the expressions of Cleaved-Caspase-3 (P < 0.001) and Cleaved- PARP (P < 0.01) were markedly up-regulated by H2O2 treat- ment compared to control (Fig. 1e, f). Meanwhile, H2O2
Fig. 1 H2O2 treatment induced HT-22 cells oxidative injury. a H2O2 treatment (100, 150, 200 and 300 µM) notably decreased cell viabil- ity and the results showed in a concentration-dependent manner. b, c H2O2 treatment remarkably decreased the expression of apoptosis- related CyclinD1 and markedly enhanced the expression of p53. D. H2O2 treatment dramatically increased the proportion of apoptotic cells. e, f H2O2 treatment dramatically enhanced the expressions of apoptosis-related Cleaved-caspase-3 and Cleaved-PARP. g H2O2 treatment markedly increased the generation of ROS. CTRL control, PARP poly-ADP-ribose polymerase, ROS reactive oxygen species. compared to non-treated group (P < 0.01, Fig. 1g). These data suggested that the injury model of AD was preliminar- ily constructed in HT-22 cells.

PF Pretreatment Alleviated H2O2‑Induced Oxidative Damage in HT‑22 Cells
HT-22 cells were dealt with gradient concentrations of PF (0, 50, 100, 150 and 200 µg/mL) and then CCK-8 assay was carried out to assess the influences of PF treatment. The results illustrated that PF treatment had no significant influ- ence on cell viability (Fig. 2a). Then, the influences of PF on H2O2-stimulated HT-22 cells were studied. H2O2-induced HT-22 cells were pre-treated with series of concentrations of PF (0, 50, 100 and 200 µg/mL). As shown in Fig. 2b, cell viability was notably increased by PF pretreatment com- pared with H2O2-treated group and exhibited in a concentra- tion-dependent way (P < 0.05 or P < 0.01). Because the most effective concentration was 200 µg/mL, hence, we chose this concentration to conduct subsequent experiments. Detection of expressions of proliferation-related proteins showed that PF treatment itself prominently promoted the expression of CyclinD1 (P < 0.05), while notably repressed the expression of p53 (P < 0.01). Meanwhile, further results illustrated that PF pretreatment alleviated H2O2-triggered suppression effect on CyclinD1 expression (P < 0.05) and the promotion effect on p53 expression (P < 0.05, Fig. 2c, d). Besides, PF individually treatment had no significant influence on cell apoptosis,
Fig. 2 PF pretreatment alleviated H2O2-induced oxidative injury in HT-22 cells. a PF treatment (0, 50, 100, 150 and 200 µg/mL) had no significant influence on cell viability. b PF pretreatment remark- ably alleviated H2O2-induced inhibitory effect on cell viability and exhibited in a dose-dependent way. c, d PF treatment notably increased the expression of apoptosis-related CyclinD1 and decreased the expression of p53. Besides, PF pretreatment markedly relieved H2O2-induced inhibitory effect on CyclinD1 expression and the pro- motion effect on p53 expression. e PF treatment had no notable influ- ence on cell apoptosis. While PF pretreatment remarkably alleviated
H2O2-induced promotion effect on cell apoptosis. f, g PF treatment notably decreased the expressions of apoptosis-related Cleaved- caspase-3 and Cleaved-PARP. PF pretreatment markedly relieved H2O2-induced promotion effects on the expressions of Cleaved-cas- pase-3 and Cleaved-PARP. h PF treatment significantly reduced the generation of ROS. PF pretreatment observably relieved the promo- tion effect on ROS generation. i PF treatment observably reduced the production of H2O2. CTRL control, ns no significance, PF Pae- oniflorin, PARP poly-ADP-ribose polymerase, ROS reactive oxygen species.*P < 0.05; **P < 0.01; ***P < 0.001 while notably reduced the expressions of apoptosis-related Cleaved-caspase-3 and Cleaved-PARP compared to control group (both P < 0.05). Additionally, PF pretreatment dra- matically relieved the promotion effects on cell apoptosis (P < 0.05, Fig. 2e), and remarkably eliminated the enhanced expressions of Cleaved-caspase-3 and Cleaved-PARP com- pared to H2O2-treated group (both P < 0.05, Fig. 2f, g). For ROS assay, we observed that the H2O2-triggered increased effect on ROS generation was significantly attenuated by PF pretreatment (P < 0.05). In addition, PF exposure itself sig- nificantly decreased ROS production (P < 0.05, Fig. 2h) and H2O2 generation (P < 0.01, Fig. 2i). In brief, these results led to a conclusion that PF possessed the cytoprotection and ROS attenuation effects on HT-22 cells.

PF Pretreatment Promoted Activation of JAK2/ STAT3 and ERK1/2 Pathways
To uncover the potential regulation mechanism, we investi- gated some of the key factors involved in the JAK2/STAT3 and ERK1/2 pathways. Results displayed that the expres- sions of p/t-JAK2 and p/t-STAT3 were notably enhanced by PF exposure (both P < 0.01), while were markedly inhibited by H2O2 treatment (both P < 0.05). However, the inhibitory effects were remarkably reversed by PF pretreat- ment (P < 0.05, Fig. 3a, b). The same expression pattern also appeared in the ERK1/2 signaling pathway exhibiting as PF exposure notably elevated the expression of p/t-ERK1/2 (P < 0.05), while H2O2 treatment observably suppressed p/t-ERK1/2 expression (P < 0.01), whereas, this suppression effect was markedly reversed by PF pretreatment (P < 0.05, Fig. 3c, d). Therefore, these results pointed out that PF pre- treatment could promote the activation of the JAK2/STAT3 and ERK1/2 pathways.

PF Pretreatment Promoted Down‑Regulation of miR‑135a
To clarify the function of miR-135a in AD, we tested the expression of miR-135a in H2O2, PF and H2O2 + PF treated HT-22 cells, respectively. The results turned out to be that miR-135a expression was remarkably inhibited by PF expo- sure compared to control (P < 0.05), while was markedly enhanced in the H2O2-treated group compared to control (P < 0.01). However in the H2O2 + PF treated group, the promotion effect on miR-135a expression induced by H2O2 exposure was markedly alleviated by PF pretreatment (P < 0.05, Fig. 4). These results implied that PF exposure had significant influences on miR-135a expression.
Fig. 4 PF pretreatment down-regulated the expression of miR- 135a in H2O2-induced HT-22 cells. PF treatment notably decreased the expression of miR-135a. Besides, H2O2 treatment dramatically enhanced the miR-135a expression, while PF pretreatment observably relieved the promotion effect. CTRL control, PF Paeoniflorin, miR- 135a microRNA-135a. **P < 0.05; ***P < 0.01 PF Pretreatment Attenuated H2O2‑Triggered Oxidative Injury in HT‑22 Cells Through Down‑Regulation of miR‑135a For further investigation of the function of miR-135a in the
Fig. 3 PF pretreatment pro- moted activation of JAK2/ STAT3 and ERK1/2 pathways in H2O2-induced HT-22 cells. a, b PF treatment dramatically increased the expressions of proteins participated in JAK2/ STAT3 pathway including p/t- JAK2 and p/t-STAT3. Besides, H2O2 treatment markedly reduced the expressions of p/t- JAK2 and p/t-STAT3, while PF pretreatment remarkably attenu- ated H2O2-triggered inhibitory effects. c, d PF treatment notably increased the expression of p/t-ERK1/2 which involved in ERK1/2 pathway. Besides, H2O2 treatment remarkably sup- pressed the expression of p/t- ERK1/2, while PF pretreatment markedly attenuated this inhibi- tory effect. CTRL control, PF Paeoniflorin, JAK2 Janus kinase 2, STAT3 signal transducer and activator of transcription 3, ERK1/2 extracellular signal regulated kinase 1/2. *P < 0.05; process of PF attenuating H2O2-triggered oxidative dam- age, we overexpressed miR-135a in HT-22 cells (P < 0.01, Fig. 5a). Results demonstrated that miR-135a mimic mark- edly reversed the promotion effects on cell viability induced by PF pretreatment (P < 0.05, Fig. 5b), as well as the pro- motion effect on CyclinD1 expression and the inhibitory effect on p53 expression triggered by PF pretreatment (both P < 0.05, Fig. 5c, d). In addition, we found that miR-135a mimic markedly reversed PF pretreatment-triggered suppres- sion effect on cell apoptosis (P < 0.05, Fig. 5e), as well as the inhibitory effects on the expressions of Cleaved-caspase-3 and Cleaved-PARP (both P < 0.01, Fig. 5f, g). Further- more, in H2O2 + PF exposure group, miR-135a mimic also significantly eliminated PF pretreatment-triggered suppres- sion effect on ROS production (P < 0.05, Fig. 5h). These observations illustrated that PF pretreatment may achieved
Fig. 5 PF pretreatment allevi- ated H2O2-induced oxidative injury in HT-22 cells by reduc- tion of miR-135a. a miR-135a was successfully overexpressed in HT-22 cells. b miR-135a overexpression notably revered PF pretreatment-triggered pro- motion effect on cell viability. c, d miR-135a mimic remarkably abolished PF pretreatment- triggered promotion effect
on CyclinD1 expression and the inhibitory effect on p53 expression. e miR-135a mimic notably reversed the inhibitory effect on cell apoptosis. f, g miR-135a mimic remarkably abolished the inhibitory effects on the expressions of Cleaved- caspase-3 and Cleaved-PARP. h miR-135a mimic markedly reversed the inhibitory effect on ROS generation. CTRL control, PF Paeoniflorin, PARP poly-ADP-ribose polymerase, ROS reactive oxygen species, ns no significance, miR-135a microRNA-135a,
its protective roles in HT-22 cells through decreasing the expression of miR-135a. PF Pretreatment Promoted Activation of JAK2/STAT3 and ERK1/2 Pathways Through Down‑Regulation of miR‑135a To explore the underlying mechanism of PF regulation, we determined the expressions of key factors involved in JAK2/ STAT3 and ERK1/2 pathways after miR-135a overexpres- sion. Results suggested that miR-135a mimic notably abol- ished the promotion effects on the expressions of p/t-JAK2 (P < 0.05), p/t-STAT3 (P < 0.05) and p/t-ERK 1/2 (P < 0.05) triggered by PF pretreatment (Fig. 6a–d). These results indicated that PF pretreatment might promote activation of JAK2/STAT3 and ERK1/2 pathways through decreasing the expression of miR-135a.

Discussion

Previous studies have shown that PF possessed neuropro- tective effects on hippocampal dysfunction animal models [21], and could ameliorate cognitive dysfunction in diabetic rats [22], therefore, it might be a promising traditional Chi- nese medicine (TCM) for AD treatment. To address this speculation, we firstly constructed an injury model of AD in HT-22 cells via H2O2 treatment and then investigated the protective effects of PF on oxidative injured HT-22 cells induced by H2O2 treatment, and attempted to explore the potential regulating mechanisms. Our results specified that PF could alleviate the oxidative damage induced by H2O2 in HT-22 cells via promoting the activation of JAK2/STAT3 and ERK1/2 pathways through down-regulating the expres- sion of miR-135a.
To elucidate how PF exerts its therapeutic effects on H2O2-induced oxidative injured HT-22 cells, we investi- gated the potential mechanisms. Herein, our results showed that PF itself treatment had no significant influence on cell viability. Whereas, following trials suggested that PF had significantly alleviated H2O2-induced oxidative damage and exhibited a certain protective effect on HT-22 cells, includ- ing increasing cell viability, decreasing cell apoptosis and declining ROS generation. Our study was in accordance with a previous study performed in osteoblastic MC3T3-E1 cells which demonstrated that PF provided a protective effect against methylglyoxal-induced cell damage via reducing oxidative stress [23]. We further explored the underlying mechanism(s). CyclinD1 is an important oncogene protein that regulates cell cycle and promotes G1/S phase transi- tion [24]. p53 plays important role in stress responses, and it can trigger cell cycle arrest and apoptosis in response to
Fig. 6 PF pretreatment pro- moted activation of JAK2/ STAT3 and ERK1/2 pathways in H2O2-induced HT-22 cells by decreasing miR-135a expression. a, b miR-135a mimic markedly abolished PF pretreatment-triggered promo- tion effects on the expressions of p/t-JAK2 and p/t-STAT3 which was involved in JAK2/ STAT3 pathway. c, d miR-135a mimic remarkably reversed PF pretreatment-triggered promo- tion effect on p/t- ERK1/2 expression which participated in ERK1/2 pathway. CTRL control, PF Paeoniflorin, JAK2 Janus kinase 2, STAT3 signal transducer and activator of tran- scription 3, ERK1/2 extracellu- lar signal regulated kinase 1/2, NC negative control. *P < 0.05; **P < 0.01
diverse stresses, including hypoxia, oxidative stress, DNA damage, ribonucleotide depletion and nutrient starvation [25]. Our results demonstrated that PF treatment markedly enhanced the expression of CyclinD1 protein and down- regulated the expression of p53 protein which further veri- fied that PF was capable of alleviating the inhibitory effect of H2O2 on cell proliferation. However, the roles of PF in cancers were completely contrary to this, which exhibited as inhibiting cell proliferation [8–11]. Our research verified that PF also had a protective effect on AD. Earlier studies verified that Cleaved-caspase 3 acted as an “actuator” to cleave structural and regulatory proteins in the cell, thereby enhanced apoptosis [26]. Caspase cleavage of the PARP protein to polypeptide indicates that the cell is undergoing apoptosis [27]. Flow cytometry showed that cell apoptosis was significantly reduced by PF treatment. Molecularly, the expression of Cleaved-caspase-3 and Cleaved-PARP were dramatically down-regulated by PF treatment. Moreover, accumulating studies clarified that oxidative stress was one of the pathogenic factors in the progression of AD [28, 29]. The imbalance between the generation and elimination of ROS resulted in oxidative stress, which may further lead to nervous system dysfunction and neuronal apoptosis [30, 31], leading to nervous system diseases such as AD. Herein, our experiments confirmed that PF treatment significantly decreased ROS generation induced by H2O2 treatment. This was in accordance with earlier studies clarifying that PF alle- viated advanced oxidation protein product (AOPP)-induced oxidative injury through decreasing ROS production [32] and PF protected thymocytes by scavenging ROS to against irradiation-induced cell damage [33].

An increasing number of researches have indicated that
TCM works by regulating the expression of miRNAs. Li et al. reported that PF inhibited doxorubicin-induced car- diomyocyte apoptosis via down-regulating the expression of miR-1 [34]. Besides, previous studies verified that PF inhibited cell proliferation and induced apoptosis by medi- ating miR-16 expression in human glioma cells [9, 10] and PF played the above roles via mediating the expression of miR-29b in multiple myeloma cells [8]. An literature review reported that miR-135a had been identified as one of the potential biomarkers for AD [17]. Therefore, we speculated that PF may play its role through regulating the expression of miR-135a. Interestingly, we observed that PF treatment could down-regulate the expression of miR-135a. To fur- ther confirm whether the effects of PF on H2O2-induced oxidative injury were achieved through down-regulation of miR-135a in HT-22 cells, we overexpressed miR-135a through cell transfection. Further experiments verified that overexpression of miR-135a can decline cell viability and enhance cell apoptosis. Moreover, overexpression of miR- 135a down-regulated the expression of CyclinD1, while up-regulated the expressions of p53, Cleaved-caspase-3 and

Cleaved-PARP. A study revealed that miR-135a was one of the ROS-regulated miRNAs [35]. Therefore, we suspected that overexpression of miR-135a may have an influence on ROS production. Our results verified that, ROS generation was markedly increased by miR-135a overexpression. Our study was in line with an earlier study proving that miR-135a overexpression can inhibit oxidative stress in atherosclerosis by decreasing ROS level [13]. These results suggested that PF alleviated H2O2-induced oxidative injury in HT-22 cells through down-regulating miR-135a expression.
It has been reported that the JAK2/STAT3 pathway might have a role in cognitive function and may have effects in the AD-relevant memory impairment NSC 178886 [36]. Researchers previ- ously verified that ERK1/2 pathway mediated neuroprotec- tive effects against H2O2 insult [37, 38]. Considering these observations mentioned above, we examined the expressions of the key factors involved in JAK2/STAT3 and ERK1/2 pathway after PF treatment and cell transfection. Results indicated that the activation of these pathways induced by PF was abrogated by miR-135a mimic in H2O2-induced HT-22 cells. In other words, PF may achieve the activation of these pathways through decreasing the expression of miR-135a.
In conclusion, these results indicated that PF could allevi- ate H2O2-induced oxidative damage of HT-22 cells which might be achieved through mediating miR-135a expression and regulating JAK2/STAT3 and ERK1/2 pathways.

References

1. Saine K, Cullum CM, Martin-Cook K, Hynan L, Svetlik DA, Weiner MF (2002) Comparison of functional and cognitive donepezil effects in Alzheimer’s disease. Int Psychogeriatr 14(2):181–185
2. Ittner LM, Gotz J (2011) Amyloid-beta and tau–a toxic pas de deux in Alzheimer’s disease. Nat Rev Neurosci 12(2):65–72. https://doi.org/10.1038/nrn2967
3. Tan CC, Yu JT, Wang HF, Tan MS, Meng XF, Wang C, Jiang T, Zhu XC, Tan L (2014) Efficacy and safety of donepezil, galan- tamine, rivastigmine, and memantine for the treatment of Alz- heimer’s disease: a systematic review and meta-analysis. JAD 41(2):615–631. https://doi.org/10.3233/jad-132690
4. Zhao L, Chang Q, Huang T, Huang C (2018) Paeoniflorin inhibits IL1betainduced expression of inflammatory mediators in human osteoarthritic chondrocyte. Mol Med Rep 17(2):3306–3311. https ://doi.org/10.3892/mmr.2017.8222
5. Zhang J, Dou W, Zhang E, Sun A, Ding L, Wei X, Chou G, Mani S, Wang Z (2014) Paeoniflorin abrogates DSS-induced colitis via a TLR4-dependent pathway. Am J Physiol Gastrointest Liver Physiol 306(1):G27–G36. https://doi.org/10.1152/ajpgi.00465
.2012
6. Ye S, Mao B, Yang L, Fu W, Hou J (2016) Thrombosis recanali- zation by paeoniflorin through the upregulation of urokinasetype plasminogen activator via the MAPK signaling pathway. Mol Med Rep 13(6):4593–4598. https://doi.org/10.3892/mmr.2016.5146
7. Li H, Jiao Y, Xie M (2017) Paeoniflorin ameliorates atheroscle- rosis by suppressing TLR4-mediated NF-kappaB activation. Inflammation 40(6):2042–2051. https://doi.org/10.1007/s1075 3-017-0644-z
8. Wang S, Liu W (2016) Paeoniflorin inhibits proliferation and promotes apoptosis of multiple myeloma cells via its effects on microRNA29b and matrix metalloproteinase2. Mol Med Rep 14(3):2143–2149. https://doi.org/10.3892/mmr.2016.5498
9. Li W, Qi Z, Wei Z, Liu S, Wang P, Chen Y, Zhao Y (2015) Pae- oniflorin inhibits proliferation and induces apoptosis of human glioma cells via microRNA-16 upregulation and matrix metal- loproteinase-9 downregulation. Mol Med Rep 12(2):2735–2740. https://doi.org/10.3892/mmr.2015.3718
10. Zheng YB, Xiao GC, Tong SL, Ding Y, Wang QS, Li SB, Hao ZN (2015) Paeoniflorin inhibits human gastric carcinoma cell proliferation through up-regulation of microRNA-124 and sup- pression of PI3K/Akt and STAT3 signaling. World J Gastroenterol 21(23):7197–7207. https://doi.org/10.3748/wjg.v21.i23.7197
11. Yang N, Cui H, Han F, Zhang L, Huang T, Zhou Y, Zhou J (2016) Paeoniflorin inhibits human pancreatic cancer cell apop- tosis via suppression of MMP-9 and ERK signaling. Oncol Lett 12(2):1471–1476. https://doi.org/10.3892/ol.2016.4761
12. Hu ZY, Xu L, Yan R, Huang Y, Liu G, Zhou WX, Zhang YX (2013) [Advance in studies on effect of paeoniflorin on nervous system]. Zhongguo Zhong yao za zhi = Zhongguo zhongyao zazhi = China. J Chin Mater Med 38(3):297–301
13. Gu X, Cai Z, Cai M, Liu K, Liu D, Zhang Q, Tan J, Ma Q (2016) Protective effect of paeoniflorin on inflammation and apoptosis in the cerebral cortex of a transgenic mouse model of Alzheimer’s disease. Mol Med Rep 13(3):2247–2252. https://doi.org/10.3892/ mmr.2016.4805
14. Kapoor S (2013) Neuroprotective effects of paeoniflorin: an emerging concept in neurology. Folia Neuropathol 51(1):92
15. Li J, Ji X, Zhang J, Shi G, Zhu X, Wang K (2014) Paeoniflorin attenuates Abeta25-35-induced neurotoxicity in PC12 cells by preventing mitochondrial dysfunction. Folia Neuropathol 52(3):285–290
16. Wang K, Zhu L, Zhu X, Zhang K, Huang B, Zhang J, Zhang Y, Zhu L, Zhou B, Zhou F (2014) Protective effect of paeoniflorin on Abeta25-35-induced SH-SY5Y cell injury by preventing mito- chondrial dysfunction. Cell Mol Neurobiol 34(2):227–234. https://doi.org/10.1007/s10571-013-0006-9
17. Nagaraj S, Zoltowska KM, Laskowska-Kaszub K, Wojda U (2019) MicroRNA diagnostic panel for Alzheimer’s disease and epige- netic trade-off between neurodegeneration and cancer. Ageing Res Rev 49:125–143. https://doi.org/10.1016/j.arr.2018.10.008
18. Liu CG, Wang JL, Li L, Xue LX, Zhang YQ, Wang PC (2014) MicroRNA-135a and -200b, potential biomarkers for Alzhei- mers disease, regulate beta secretase and amyloid precursor protein. Brain Res 1583:55–64. https://doi.org/10.1016/j.brain res.2014.04.026
19. Yang TT, Liu CG, Gao SC, Zhang Y, Wang PC (2018) The serum exosome derived microRNA-135a, -193b, and -384 were potential Alzheimer’s disease biomarkers. Biomed Environ Sci 31(2):87– 96. https://doi.org/10.3967/bes2018.011
20. Du XJ, Lu JM (2018) MiR-135a represses oxidative stress and vascular inflammatory events via targeting toll-like receptor 4 in atherogenesis. J Cell Biochem 119(7):6154–6161. https://doi. org/10.1002/jcb.26819
21. Zhong SZ, Ge QH, Li Q, Qu R, Ma SP (2009) Peoniflorin attentuates Abeta((1–42))-mediated neurotoxicity by regulat- ing calcium homeostasis and ameliorating oxidative stress in hippocampus of rats. J Neurol Sci 280(1–2):71–78. https://doi. org/10.1016/j.jns.2009.01.027
22. Sun X, Li S, Xu L, Wang H, Ma Z, Fu Q, Qu R, Ma S (2017) Paeoniflorin ameliorates cognitive dysfunction via regulating SOCS2/IRS-1 pathway in diabetic rats. Physiol Behav 174:162– 169. https://doi.org/10.1016/j.physbeh.2017.03.020
23. Choi EM, Suh KS, Rhee SY, Kim YS (2014) Inhibitory effect of paeoniflorin on methylglyoxal-mediated oxidative stress in oste- oblastic MC3T3-E1 cells. Phytomedicine 21(10):1170–1177. https://doi.org/10.1016/j.phymed.2014.05.008
24. Fu S, Xu H, Gu M, Liu C, Wang Q, Wan X, Chen Y, Chen Q, Peng Y, Cai Z, Zhou J, Wang Z (2017) Adiponectin deficiency contributes to the development and progression of benign prostatic hyperplasia in obesity. Sci Rep 7:43771. https://doi. org/10.1038/srep43771
25. Hu W, Feng Z, Levine AJ (2012) The regulation of multiple p53 stress responses is mediated through MDM2. Genes Cancer 3(3–4):199–208. https://doi.org/10.1177/1947601912454734
26. Adams JM, Cory S (2002) Apoptosomes: engines for caspase activation. Curr Opin Cell Biol 14(6):715–720
27. Zhang W, Hirschler-Laszkiewicz I, Tong Q, Conrad K, Sun SC, Penn L, Barber DL, Stahl R, Carey DJ, Cheung JY, Miller BA (2006) TRPM2 is an ion channel that modulates hematopoietic cell death through activation of caspases and PARP cleavage. Am J Physiol Cell Physiol 290(4):C1146–C1159. https://doi. org/10.1152/ajpcell.00205.2005
28. Ahmad W, Ijaz B, Shabbiri K, Ahmed F, Rehman S (2017) Oxi- dative toxicity in diabetes and Alzheimer’s disease: mechanisms behind ROS/ RNS generation. J Biomed Sci 24(1):76. https:// doi.org/10.1186/s12929-017-0379-z
29. Cheignon C, Tomas M, Bonnefont-Rousselot D, Faller P, Hureau C, Collin F (2018) Oxidative stress and the amyloid beta peptide in Alzheimer’s disease. Redox Biol 14:450–464. https://doi.org/10.1016/j.redox.2017.10.014
30. Wilson C, Gonzalez-Billault C (2015) Regulation of cytoskel- etal dynamics by redox signaling and oxidative stress: impli- cations for neuronal development and trafficking. Front Cell Neurosci 9:381. https://doi.org/10.3389/fncel.2015.00381
31. Li H, Zhu YH, Chi C, Wu HW, Guo J (2014) Role of cytoskel- eton in axonal regeneration after neurodegenerative diseases and CNS injury. Rev Neurosci 25(4):527–542. https://doi. org/10.1515/revneuro-2013-0062
32. Song S, Xiao X, Guo D, Mo L, Bu C, Ye W, Den Q, Liu S, Yang X (2017) Protective effects of Paeoniflorin against AOPP- induced oxidative injury in HUVECs by blocking the ROS-HIF- 1alpha/VEGF pathway. Phytomedicine 34:115–126. https://doi. org/10.1016/j.phymed.2017.08.010
33. Li CR, Zhou Z, Zhu D, Sun YN, Dai JM, Wang SQ (2007) Protective effect of paeoniflorin on irradiation-induced cell damage involved in modulation of reactive oxygen species and the mitogen-activated protein kinases. Int J Biochem Cell Biol 39(2):426–438. https://doi.org/10.1016/j.biocel.2006.09.011
34. Li JZ, Tang XN, Li TT, Liu LJ, Yu SY, Zhou GY, Shao QR, Sun HP, Wu C, Yang Y (2016) Paeoniflorin inhibits doxorubicin- induced cardiomyocyte apoptosis by downregulating micro- RNA-1 expression. Exp Ther Med 11(6):2407–2412. https:// doi.org/10.3892/etm.2016.3182
35. Shin EA, Sohn EJ, Won G, Choi J-U, Jeong M, Kim B, Kim M-J, Kim S-H (2014) Upregulation of microRNA135a-3p and death receptor 5 plays a critical role in Tanshinone I sensitized prostate cancer cells to TRAIL induced apoptosis. Oncotarget 5(14):5624
36. Chiba T, Yamada M, Sasabe J, Terashita K, Shimoda M, Matsuoka M, Aiso S (2009) Amyloid-beta causes memory impairment by disturbing the JAK2/STAT3 axis in hippocampal neurons. Mol Psychiatry 14(2):206–222. https://doi.org/10.1038/mp.2008.105
37. Zheng W, Chong CM, Wang H, Zhou X, Zhang L, Wang R, Meng Q, Lazarovici P, Fang J (2016) Artemisinin conferred ERK mediated neuroprotection to PC12 cells and cortical neu- rons exposed to sodium nitroprusside-induced oxidative insult. Free Radic Biol Med 97:158–167. https://doi.org/10.1016/j.freer adbiomed.2016.05.023
38. Chong CM, Zheng W (2016) Artemisinin protects human reti- nal pigment epithelial cells from hydrogen peroxide-induced oxidative damage through activation of ERK/CREB signaling. Redox Biol 9:50–56. https://doi.org/10.1016/j.redox.2016.06.002