Paeoniflorin improves functional recovery through repressing neuroinflammation and facilitating neurogenesis in rat stroke model

MCAO model and Paeoniflorin delivery

Animal experiments were conduct under the ethic approval obtained from the Committee for Animal Experiments at Wenzhou Medical University (wydw2019-0903). Forty adult male Sprague-Dawley (SD) rats (6–8 weeks, 200–250 g, n = 10 of each group) were obtained from the Laboratory Animal Centre Wenzhou Medical University and kept under the specific-pathogen-free (SPF).

MCAO procedures were performed on all groups (the sham-operated group, vehicle experimental group, vehicle Paeoniflorin 5 mg/kg group and vehicle Paeoniflorin 10 mg/kg group) as previously described (Hata et al., 1998; Yu et al., 2018). In short, each rat received intraperitoneal anesthesia using choral hydrate (350 mg/kg) before exposing the correct vessels by a midline skin incision. The middle cerebral artery (MCA) was occluded by inserting a coagulated external carotid artery into the internal carotid artery. The MCA perfusion was allowed by withdrawing the suture after two hours. Verification of the occurrence of ischemia was examined by measuring the blood flow of MCA through a Laser Doppler flowmetry (LDF, Perimed, PeriFlux 5000). The same operations were performed on the sham-operated group rats exclude the MCA occlusion. All rats were kept on a heating pad (Malvern, UK) at 37.0 °C until the skin incision enclosure. After the previous procedure, the vehicle group was treated with (phosphate-buffered saline consists of 150 µl saline and 20% dimethyl sulfoxide [DMSO]) by intraperitoneal injection; while the paeoniflorin group was injected with Paeoniflorin (5 mg/kg and 10 mg/kg) twice per day for 14 days.

Animals were housed in a colony room under controlled temperature (22 °C), and a 12:12 h light-dark cycle, with food and water available. All rats were sacrificed by overetherization and decapitation. 

Behavioral tests

Neurological function assessment was performed by the mNSS test in accordance with the previous method (Chen et al., 2001). This test was carried out at day 0, 1, 3, 7 and 14 after MCAO. This test, including the measurements for the motor, sensory, beam balance, and reflexes (Schaar, Brenneman & Savitz, 2010). This test was graded on a scale of 0–18. More serious injuries usually lead to higher scores.

The sensorimotor function was evaluated via the foot-fault test as the method reported by Zhang et al. (2015a). The number of foot-fault errors, paw slipping numbers, were recorded when the rat is trained to cross an elevated grid at day 0, 1, 3, 7 and 14 after MCAO.

Cell culture

The murine BV2 microglial cells were procured from the ATCC (Manassas, VA, USA), and then maintained in DMEM with 10% FBS, 1% penicillin, and streptomycin (Sigma), and were cultured at 37 °C in an incubator with 5% CO₂. The oxygen-glucose deprivation (OGD) was performed by exposing BV2 cells to DMEM without glucose or serum and incubating in a specific environment (5% CO₂ and 95% N₂) for 6 h. BV2 microglial cells were then incubated in Paeoniflorin (10 µM) for 6 h before TNF-α, IL-1β, IL-6 levels were measured by qPCR analysis.

Quantitative real-time PCR (qPCR)

The total RNA was isolated from BV2 cells or OGD model cells in the absence or presence of 10 µM Paeoniflorin treatment for 6 h with Trizol kit (Solarbio, Beijing, China) following the manufacturer’s manual. Reverse transcription (RT) was implemented by Prime Script™ Master Mix (Katara, Japan). qPCR was performed with 2 × SYBR Green Mix (Solarbio) on an Applied Biosystems 7500 Fast Real-Time PCR Systems (Thermo Fisher Scientific, MA). β-actin was chosen as an internal reference and the fold changes were calculated using the 2^−ΔΔCt method.

Western blotting

Total protein was purified from the brain tissues of 14 days after MCAO in the absence or presence of 10 mg/kg Paeoniflorin treatment using lysis buffer and then centrifuged at 12,000 rpm for about 15 min. The protein concentration was measured with the BCA kit (Pierce, IL). The western blotting was carried out as the method reported by Zhu et al. (2019). The cell was cultured with primary antibodies, anti-Iba-1 antibody (1 µg/ml; ab5076; Abcam, MA), anti-JNK antibody (1:1000; ab179461; Abcam, MA), anti-phosphorylated-JNK antibody (1:1,000; #9251; Cell Signaling Technology), anti-P65 antibody (0.5 µg/ml; ab16502; Abcam), anti-Histone H3 antibody (1:1,000; ab215728; Abcam) and anti-actin antibody (1:1000, ab6276, Abcam), at 4 °C overnight. After washing, horseradish peroxidase (HRP)-conjugated secondary antibodies were incubated with membranes for 1 h.

(Enzyme linked immunosorbent assay) ELISA

The ELISA assay was carried out to detect TNF-α, IL-1β and IL-6 content using ELISA kits (PT516/PI303/PI328; Beyotime, China) following the manufacture’s manual. In short, the brain tissues were extracted at 14 days after MCAO in the absence or presence of 10 mg/kg Paeoniflorin treatment, and then the supernatant of the brain tissue homogenate (1:20 dilution) was added to 96-well plates that are coated with indicated antibodies, in order to quantify the protein level of TNF-α, IL-1β, IL-6 in the tissue of the brain. After the reaction was completed, the absorbances of the sample were measured at 450 nm with a microplate reader. At least three repeated measurements were taken for each sample.

Immunofluorescence staining

The immunofluorescence staining for brain tissues was carried out as the method reported by Burton (Burton, Sparkman & Johnson, 2011), and the brain tissues were extracted at 14 days after MCAO in the absence or presence of 10 mg/kg Paeoniflorin treatment. The cell was cultured with specific primary antibody against Iba-1 (1:1,500; ab178846) at 4 °C overnight, and then incubated with Anti-rat Alexa Fluor 488 (1:1000; CST) as a secondary antibody. The samples were evaluated with the LEICA TCS SPE microscope (Leica, Germany).

Statistics

Data from the experiments were presented as mean ± standard deviation (SD). The difference between two groups was compared using two-tailed  student’s t-test, or one-way analysis of variance (ANOVA)  followed by the Scheffé test. Values less than 0.05 was considered significant differences.

Paeoniflorin improved the functional recovery in rat stroke model

The role of Paeoniflorin in functional recovery following cerebral ischemic injury was first tested in rat stroke model. The rats were subjected to MCAO and then Paeoniflorin was intraperitoneally administrated for 14 days at 2 h after MCAO. The Foot-fault test and mNSS were performed at day 0, 1, 3, 7 and 14 after MCAO. There were no remarkable differences in neurological functional deficits at day 1 after Paeoniflorin administration, while Paeoniflorin markedly improved functional recovery at day 3, 7 and 14 following MCAO, compared with vehicle experimental group (Fig. 1A). Meanwhile, Paeoniflorin administration significantly reduced the frequency of forelimb and foot faults at day 3, 7 and 14 following MCAO (Fig. 1B). Importantly, better protective effects were obtained with higher Paeoniflorin concentrations in rat stroke model (Figs. 1A and 1B). The current data verified that Paeoniflorin enhanced the functional recovery in rat stroke model.

Paeoniflorin repressed microglia activation in rat stroke model

Microglia are the resident innate immune cells in the brain. Microglia in resting state are responsible for routine immune surveillance (Guruswamy & ElAli, 2017). After cerebral ischemic injury, microglia are activated in morphological and phenotypical changes (Jayaraj et al., 2019), and the severity of cerebral ischemic injury is correlated with the activated state of microglia (Emmrich et al., 2015). To explore whether Paeoniflorin repressed the activation of microglia in rat stroke model, microglia were analyzed using immunofluorescence analysis after Paeoniflorin administration. As shown in Fig. 2A, rats subjected to MCAO showed a significant microglial proliferation in ipsilateral cortex, whereas additional Paeoniflorin treatment repressed MCAO-induced microglial proliferation. Moreover, Paeoniflorin inhibited MCAO-induced microglial activation as measured by the intensive ramified Iba-1-positive staining (Fig. 2A). To further assess the function of Paeoniflorin on microglial activation, the protein expression of Iba-1 in ipsilateral cortex was analyzed. Rats subjected to MCAO showed a significant increase of Iba-1 protein level in ipsilateral cortex, whereas Paeoniflorin treatment inhibited this increase compared with the control (Figs. 2B and 2C). Based on this result, the role of Paeoniflorin in regulating the secretion of pro-inflammatory cytokines was assessed in cerebral tissues. Figures 3A–3C showed that rats subjected to MCAO showed a significant increase in the production of IL-1β, TNF-α and IL-6, whereas Paeoniflorin administration repressed pro-inflammatory cytokines production in cerebral tissues. These results demonstrated that Paeoniflorin exerts the important role in repressing microglial activation and subsequent neuroinflammation in rat stroke model.

Paeoniflorin repressed microglial viability and inflammatory cytokines production in vitro

To test the role of Paeoniflorin in microglial activation in vitro, the viability of murine BV-2 microglial cells on oxygen and glucose deprivation (OGD) was assayed in the presence or absence of Paeoniflorin. Figure 4A showed that the viability of BV-2 cells was reduced after Paeoniflorin treatment under normoxia or OGD, indicating that Paeoniflorin inhibited BV-2 cells activity. The effect of Paeoniflorin on pro-inflammatory cytokines production in vitro was further assessed. Here OGD model was used as an in vitro model of ischemia. The mRNA expression level of IL-1β, TNF-α and IL-6 was assayed by Real time-PCR under OGD in the presence or absence of Paeoniflorin, and 18s was used as internal gene to control the different amount of RNA input due to the different cell viability. Figures 4B–4D showed that IL-1β, TNF-α and IL-6 expression level was upregulated in BV-2 cells under OGD, but additional treatment of Paeoniflorin repressed OGD-induced increase in IL-1β, TNF-α and IL-6 expression. These data suggested that Paeoniflorin improved functional recovery, at least in part, by suppressing microglia-mediated inflammatory response.

Paeoniflorin facilitated neurogenesis and vasculogenesis in rat stroke model

The marker of endothelial cells of cerebral blood vessels, vWF, was next assessed after Paeoniflorin administration. As shown in Figs. 5A and 5B, Paeoniflorin treatment resulted in a significant upregulation of the vWF expression in peri-infarct zone compared with control in rat stroke model. The indication of migrating neuroblasts, doublecortin, was also assessed in peri-infarct zone after Paeoniflorin treatment. Figures 5C and 5D showed that Paeoniflorin treatment resulted in a significant increase of the doublecortin expression in peri-infarct zone compared with control. Furthermore, Paeoniflorin reduced the TUNEL-positive neurons compared with control (Figs. 5E and 5F). These results demonstrated that Paeoniflorin promoted neurogenesis and vasculogenesis after ischemic stroke.

Paeoniflorin repressed JNK and NK-κB signaling activation

Given the regulatory role of Paeoniflorin in Jun N-terminal kinase (JNK) signaling in many diseases (Liu et al., 2019; Wu et al., 2019a; Zhang et al., 2017a) and the effect of JNK on regulating inflammatory response (Chen et al., 2018), we next investigated whether Paeoniflorin repressed JNK signaling activation in rat stroke model. The results from western blot analysis showed that the level of p-JNK (phosphorylated JNK) was markedly upregulated in cerebral tissues at 4 h after MCAO, whereas additional treatment of Paeoniflorin weakened the p-JNK level in cerebral tissues compared with control, suggesting that ischemia-induced activation of JNK signaling was repressed by Paeoniflorin administration (Figs. 6A and 6B).

It is well-known that JNK could activate NF-κB signaling to aggravate the progression of inflammatory response (Pan et al., 2013; Zheng et al., 2020). We thus tested whether neuroinflammation-relieving role of Paeoniflorin was associated with NF-κB signaling activation. Figures 6C and 6D showed that the nuclear protein level of p65 was markedly upregulated in rat stroke model, whereas additional Paeoniflorin treatment weakened the nuclear p65 level in cerebral tissues compared with control.

Immunofluorescence analysis also showed that Paeoniflorin administration partially inhibited MCAO-induced nuclear translocation of NF-κB p65 in primary microglia (Fig. 6E). Taken together, the current results demonstrated that Paeoniflorin improves functional recovery through repressing neuroinflammation and facilitating neurogenesis by inhibiting JNK- NF-κB signaling in rat stroke model.