Rea regulates microglial polarization and attenuates neuronal apoptosis via inhibition of the NF‐κB and MAPK signalings for spinal cord injury repair

Abstract Inflammation and neuronal apoptosis aggravate the secondary damage after spinal cord injury (SCI). Rehmannioside A (Rea) is a bioactive herbal extract isolated from Rehmanniae radix with low toxicity and neuroprotection effects. Rea treatment inhibited the release of pro‐inflammatory mediators from microglial cells, and promoted M2 polarization in vitro, which in turn protected the co‐cultured neurons from apoptosis via suppression of the NF‐κB and MAPK signalling pathways. Furthermore, daily intraperitoneal injections of 80 mg/kg Rea into a rat model of SCI significantly improved the behavioural and histological indices, promoted M2 microglial polarization, alleviated neuronal apoptosis, and increased motor function recovery. Therefore, Rea is a promising therapeutic option for SCI and should be clinically explored.

neuronal inflammation and apoptosis by inactivating NF-κB and MAPK signalling.
Rehmannioside A (Rea, C 21 H 32 O 15 , Figure 1A) is a neuroprotective compound 25,26 isolated from Rehmanniae radix, an herb used in various Chinese medicine formulations. It remains to be elucidated whether Rea can promote functional recovery after SCI.
In the present study, we assessed the neuroprotective characteristics of Rea both in vitro and in vivo, along with pharmacological effects and underlying mechanisms. Our results indicated that Rea inhibited both NF-κB and MAPK signalling pathways and mitigated inflammation by promoting M2 (anti-inflammatory phenotype) polarization of the microglia, which eventually attenuated neuronal apoptosis. In the animal model of SCI as well, daily intraperitoneal injection of 80 mg/kg Rea significantly improved the behavioural and histological indices, and accelerated recovery of hindlimb motion. Overall, our experimental data indicated that Rea enhanced the polarization of M2 microglia and reduced neuronal apoptosis and promoted functional recovery through the regulation of MAPK and NF-κB signals after SCI. Thus, Rea is a promising drug for SCI treatment and ought to be investigated further.

| Reagents and antibodies
Rehmannioside A (purity > 98% by HPLC) was obtained from HUIJIA BIOTECHNOLOGY (Xiamen, China) and dissolved in DMSO.
Antibodies against CD68, NF-200, NeuN and Lipopolysaccharide (LPS) were bought from Abcam, and the anti-iNOS antibody was from Proteintech. Calcium fluorescein-AM/PI was provided by Keygen Biotech.

| Establishment of SCI model and treatment regimen
Thirty-six female SD rats (200-250 g, SLAC Laboratory Animal Company) were randomized into the Sham-operated, SCI and Immunofluorescence staining and quantification analysis for iNOS in BV2 cells treated above. Scale bar = 100 μm. #P < .05, ##P < .01 vs the BV2 cells were untreated. *P < .05, **P < .01 vs the BV2 cells were treated with LPS alone SCI + Rea groups. To induce SCI, the animals were anaesthetized via intraperitoneal injection of 1% (w/v) pentobarbital sodium (40 mg/kg), and the muscles tissues around the T9 spinous process were separated. The spinal cord was then exposed by T9 laminectomy and clamped with vascular clamp (15 g force, Oscar) for 1 minute to induce SCI. The Sham-operated rats underwent the same operation except for spinal cord compression. The bladder of the SCI model was manually emptied twice a day. In addition, 80 mg/kg Rea was injected intraperitoneally into the treatment group daily, and the other animals were injected with the same amount of normal saline. All protocols were consistent with the Animal Care and Use Committee of the Second Affiliated Hospital of Zhejiang University.

| Functional behaviour evaluation
The locomotive function was assessed on days 0, 1, 3, 7, 14 and 28 post-SCI with the Basso, Beattie and Bresnahan (BBB) scale and footprint analysis. The BBB rating scale ranged from 0 (complete paralysis of hind limbs) to 21 (normal movement). 27 Each rat was observed individually for 5 minutes and the BBB score was recorded.
Footprint analysis was performed by immersing the hindlimb in blue dye and the forelimb in red dye, 28 and the rats were allowed to walk through a pipeline lined with a white paper (10 cm width and 100 cm length). The resulting footprints are evaluated. All tests were conducted by three independent investigators in a blinded manner.

| Cell culture and co-culture
BV2 and PC12 cells were cultured in high glucose DMEM supplemented with 10% foetal bovine serum (FBS), streptomycin (100 μg/ mL) and penicillin (100 U/mL) at 37°C under 5% CO 2 . BV2 cells were pre-treated overnight with Rea (0, 20, 40 and 80 μmol/L) and then incubated for 24 hours with or without 1 μg/mL LPS. For the coculture system, BV2 cells pre-treated overnight with/out 80 μmol/L Rea, and then BV2 cells with/out LPS stimulation were rinsed with PBS to eliminate the effects of remaining LPS and Rea. Next, BV2 cells were seeded into inserts (pore size = 0.4 μm; Corning). The latter were placed on the PC12 monolayer at the bottom of the well and cultured for 24 hours.

| Cell viability assay
Cell viability was evaluated using the CCK-8 (Beyotime) kit according to the manufacturer's instructions. Briefly, BV2 cells were seeded in 96-well plates at the density of 10 000/well, and with 0, 5, 10, 20, 40, 80 or 160 μmol/L Rea for 24 hours. In addition, BV2 cells pretreated overnight with Rea (0 or 80 μmol/L) and then with LPS (0 or 1 μg/mL) for 24 hours were co-cultured with PC12 as described. Ten microlitres CCK-8 reagent was added to each well, and the absorbance at 450 nm was measured after a 2 hours incubation.

| Reverse transcription quantitative polymerase chain reaction (RT-qPCR)
RNA was extracted from the suitably treated cells using Trizol reagent (Invitrogen) according to the manufacturer's instructions, and RT-qPCR was performed as previously described. 29 Each experiment was repeated thrice, and the relative gene expression levels were calculated using the 2 −ΔΔCt method. The primer sequences are shown in Table 1.

| Calcein AM/propidium iodide (PI) staining
Apoptosis was evaluated by calcein AM/PI staining as per established protocols. Briefly, PC12 cells were seeded in a 6-well plate, and co-cultured as described with BV2 cells that were pre-treated overnight with Rea (0 or 80 μmol/L), followed by LPS (0 or 1 μg/ mL) for 24 hours. After co-culturing for 24 hours, the medium was removed, and the PC12 cells were rinsed thrice with PBS. The cells were then incubated with calcein AM/PI at 37°C in the dark for 30 minutes. The stained cells were observed under a fluorescence microscope (Olympus).

| Immunofluorescence Staining
In vitro study, BV2 cells was pre-treated overnight with or without Rea (80 μmol/L) before stimulation with or without 1 μg/mL LPS for 24 hours. The cells were rinsed gently with PBS, fixed with 4% paraformaldehyde (PFA) at room temperature for 15 minutes, and For vivo study, the differentially treated rats were perfused with 0.9% sodium chloride and 4% paraformaldehyde (PFA) after anaesthetization, and a segment of the spinal cord was cut approximately 1 cm on each side of the damaged region. Then, the tissues were fixed in 4% PFA for 48 hours, dehydrated and embedded in paraffin. Longitudinal 5μm-thick sections were cut from the paraffin blocks, dewaxed and boiled in citrate buffer for antigen retrieval as previously described. 30 After blocking with 5% BSA at 37°C for 1 hour, the tissue (1:500) at 4°C. The following steps were same as above.

| Histological staining
The tissue sections were processed as described and stained with haematoxylin and eosin (H&E) and observed under a light microscope (Olympus).

| Statistical analysis
All data were presented as the mean ± standard deviation. The different groups were compared by one-way analysis of variance and Tukey's post-hoc test using GraphPad Prism 7.0. P < .05 was considered statistically significant. These data demonstrated that LPS up-regulated pro-inflammatory mediators in BV2 cells, which was attenuated by Rea pre-treatment.

| Rea facilitates the recovery of motor function after SCI
We performed behavioural and pathological measurements to assess functional recovery after SCI with or without Rea treatment.
The behavioural analysis was based on the Basso Beattie Bresnahan (BBB) motor scores and footprint test. Following SCI induction, the rats showed lax palsy and recovered in a time-dependent manner.
However, the mean BBB score of the Rea-treated rats was 11 compared to only 6 in the untreated SCI group ( Figure 5A), indicating that Rea accelerates recovery after SCI. Consistent with this, the untreated rats showed obvious dragging of the hindlimbs in the footprint test ( Figure 5B) whereas the Rea-treated group partially recovered the coordination of front and hind limbs on day 28 after SCI ( Figure 5B,D, red arrow). Furthermore, histological examination of the injured spinal cord showed numerous cavities in the longitudinal plane ( Figure 5E), which were significantly narrowed upon Rea treatment, which is consistent with Figure 5C. In conclusion, Rea can accelerate the recovery of motor function in rats after SCI, likely by inhibiting the release of inflammatory factors from the activated microglia, and increasing neuronal survival and tissue regeneration.

| Rea boosts M2 polarization and mitigates inflammation and neuronal apoptosis after SCI
SCI triggers microglia/macrophage activation and polarization to the M1 phenotype, resulting in inflammatory damage that impairs tissue repair and aggravates neurological symptoms. 31 Thus, re-polarizing the activated microglia/macrophage to the anti-inflammatory M2 phenotype can significantly improve recovery after SCI. 9 The number and distribution of the M1 macrophages was determined by immunostaining with CD68 and GFAP. As shown in Figure 6A

| Rea promoted neural repair and axonal mobilization after SCI
The effects of Rea on axon regeneration and axon repair were determined by immunostaining for MAP2, a structural protein of axonal microtubules. As shown in Figure 7A,

| D ISCUSS I ON
The incidence of spinal cord injury (SCI) due to accidents and other physical trauma has increased in recent years, and warrants novel, effective therapeutic strategies. [34][35][36] Several Chinese herbal medicine formulations have shown significant therapeutic effects on SCI. 37,38 Rea is a neuroprotective and anti-inflammatory compound present in the herbal extract of R radix, which is used to treat neurodegenerative diseases. However, no study so far has tested the potential therapeutic effects of this extract against SCI.
We The NF-κB and MAPK pathways are the key regulators of inflammation-related genes. 49 We hypothesized therefore that Rea exerts is anti-inflammatory effects by targeting the NF-κB and MAPK signalling pathways. NF-κB is a transcription factor that is activated by endotoxins, bacteria, cytokines, tumour antigens etc, 20,50 which then translocates to the nucleus and promotes expression of pro-inflammatory genes like iNOS, COX-2, IL-6, IL -1β etc Rea inhibited NF-κB activity by simultaneously inhibiting p65 transcription and phosphorylation even at the low concentration of 20µM. MAPKs also lie upstream of the inflammatory cascade 24,51 and activate the ERK, JNKs and p38 pathways. 19,52,53 Subsequently, these transcription factors activate downstream genes to promote the release of pro-inflammatory factors, leading to neuroinflammation. Rea treatment significantly decreased the phosphorylation levels of NF-κB p65, I-κBα, JNK, ERK and p38 in a dose-dependent manner. Therefore, its neuroprotective effects can be attributed to the inhibition of LPS-induced NF-κB and MAPK signalling.
To summarize, Rea can mitigate neuroinflammation and promote M2 polarization of microglia by inhibiting NF-κB and MAPK signalling pathways, thereby reducing neuronal apoptosis and improving functional recovery after SCI. Therefore, the NF-κB and MAPK cascades are suitable targets for reducing inflammation during neurological injuries. However, the long-term side effects of Rea remain uncertain and there is a need to develop novel delivery systems for sustained and targeted drug release in order to improve the therapeutic outcomes. 54,55 The mechanistic basis of Rea action also needs to be explored further.

| CON CLUS ION
Rea inhibited LPS-induced inflammation by targeting the NF-κB and MAPK signalling pathways, and prevented neuronal apoptosis by promoting M2 polarization of microglia/macrophage in vivo and in vitro. The above research prove that Rea may become a potential drug for SCI therapy.

ACK N OWLED G EM ENTS
This research was financially supported by the Zhejiang medical and health science and technology project (Grant No. 2017ZD017).

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.