Genistein-3′-sodium Sulfonate Attenuates Cerebral Ischemia/reperfusion Injury in Rats by Down-regulating Microglial M1 Polarization Through α7nAChR-mediated Inhibition of NF-κb Signaling Pathway

Background: Microglial M1 depolarization mediated prolonged in�ammation contributing to brain injury in ischemic stroke. Our previous study revealed that Genistein-3′-sodium sulfonate (GSS) exerted neuroprotective effects in ischemic stroke and reduced pro-inammatory cytokines production. This study aimed to explore whether GSS protected against brain injury in ischemic stroke by regulating microglial M1 depolarization and its underlying mechanisms. Methods: We established transient middle cerebral artery occlusion and reperfusion (tMCAO) model in rats and used lipopolysaccharide (LPS)-stimulated BV2 microglial cells as in vitro model. Brain infarcted volume and neurological de�cit were evaluated by TTC staining and Garcia assessment, respectively. M1-depolarized markers, α7nAChR and NF-κB signaling proteins were determined using western blot. Real-time PCR was used to determine the expression of M1 depolarization markers. Morphological changes, IL-1β expression and the nucleus translocation of P65-NF-κB were measured using immuno�uorescent staining. The level of IL-1β was also determined using ELISA. Results: Our results showed that GSS treatment signi�cantly reduced the brain infarcted volume and improved the neurological function in tMCAO rats. Meanwhile, GSS treatment also dramatically reduced microglia M1 depolarization, reversed α7nAChR expression, and inhibited the activation of NF-κB signaling in the ischemic penumbra brain regions. These effects of GSS were further veri�ed in LPS-induced M1 depolarization of BV2 cells. Furthermore, pretreatment of α7nAChR inhibitor (α-BTX) signi�cantly restrained the neuroprotective effect of GSS treatment in both tMCAO rats and LPS-stimulated BV2 cells. Alpha-BTX also blunted the regulating effects of GSS on GSM1 depolarization and α7nAChR expression. Conclusions: This study demonstrates that GSS protects against brain injury and neurological de�cit in ischemic stroke by reducing microglia M1 depolarization to suppress neuroin�ammation


Background
Stroke is the third leading cause of death among human diseases, and over 80% of stroke results from brain ischemia [1,2] .Currently, therapeutic strategies for ischemic stroke are limited: administration of tissue plasminogen activator (tPA) or mechanical thrombectomy to gain brain reperfusion.[5] Thus, seeking novel drugs or other therapeutic strategies are still the concerns of neuroscientists.
Ischemic stroke activates in ammatory cascades in the acute phase, resulting in increased production of pro-in ammatory cytokines and chemokines, aggregation and adhesion of in ammatory cells, impairing of the blood-brain barrier (BBB), and subsequently exacerbating brain damage [6,7] .Post-stroke neuroin ammation is considered as a favorable target for the treatment of ischemic stroke [1,3,7,8] .
Microglia, one of the essential participants in neuroin ammation, is activated in a few minutes of ischemic stroke hitting [9,10] .Activated microglial cells polarize to M2 phenotype in the early stage of ischemic stroke and subsequently switch to M1 phenotype [11] .The roles of M2 and M1 polarization of microglia in neuroin ammation are antagonistic: M2 phenotype microglial cells exert anti-in ammatory and neuroprotective effects by secreting transforming growth factor β (TGF-β) and interleukin-10 (IL-10), while M1 polarization produces pro-in ammatory cytokines such as interleukin-1β (IL-1 β), IL-6 and tumor necrosis factor α (TNF-α), which contribute to the disruption of BBB and the deterioration of brain injury [3,12,13] .[16][17][18] .Cholinergic anti-in ammatory pathway (CAIP) is mediated by the stimulation of vagal nerve, which releases acetylcholine (ACh) to modulate cerebral in ammation mainly through α7 nicotinic acetylcholine receptor (α7nAChR) [14,19,20] .This receptor is expressed in the membrane of microglia, neurons, and endothelial cells in the brain.Activation of α7nAChR reduces microglial activation, inhibits the production of pro-in ammatory cytokines, and attenuates brain injury and functional de cits [6,21,22] .Moreover, it was reported that activation of α7nAChR inhibited nuclear factor kappa B (NF-κB) pathway including blocking the phosphorylation of IκB and thereby restraining the nucleus translocation of p65-NF-κB, a subunit of NF-κB [23,24] .In ischemic stroke mice, α7nAChR agonist reduced the brain injury through decreasing the phosphorylation of p65-NF-κB protein in microglia [25][26][27] ,demonstrating that α7nAChRmediated inhibition of NF-κB signaling is a crucial neuroprotective mechanism against ischemic brain injury.
Estrogen displays potent neuroprotective effects, whereas long-term estrogen treatment increases the incidence risk of breast cancer [28][29][30] .Therefore, phytoestrogens are seeing as an alternative to estrogen.
Genistein is widely studied as a phytoestrogen which protects against ischemic stroke-induced brain injury.However, the bioavailability of genistein is limited because of its poor water and lipid solubility [31] .
Genistein-3'-sodium sulfonate (GSS) is a sulfonated product of genistein, which has a better water solubility than genistein.Our previous research indicated that GSS treatment signi cantly alleviated cerebral edema and brain infarction in transient middle cerebral artery occlusion (tMCAO) rats [32] .
Besides, we also found that Genistein has a therapeutic effect in tMCAO rat because it down-regulates in ammatory cytokines levels such as IL-1β, TNF-α and IL-6, demonstrating that GSS precursor has signi cant anti-in ammation effects [33] .17β-Estradiol and progesterone promote microglial depolarization from M1 to M2 phenotype in ischemic stroke [34,35] .Therefore, we hypothesized that GSS could inhibit microglial M1 depolarization through regulating α7nAChR-mediated inhibition of NF-κB pathway and thus suppress the neuroin ammation and protect against the brain injury in ischemic stroke.In this study, we investigated the effects of GSS on microglial M1 polarization and its underlying molecular mechanisms using tMCAO rats and lipopolysaccharide (LPS)-treated microglial cells.

Animal and Drugs
Animal experimental protocols were approved by the Gannan Medical University Animal Care and Use Committee.All animal experiments were performed following the Guidelines for the Care and Use of Laboratory Animals of China Medical University.Sprague-Dawley (SD) rats (Male, 250-300g) were purchased from Hunan SJA Laboratory Anima Company (Hunan, China) and were housed at 50-60% relative humidity and a temperature of 22-25℃.The rats drank sterile running water and took SPF grade chow freely.Before performing experiments, all rats were acclimatized to the laboratory environment for at least one week.

Transient middle cerebral artery occlusion and reperfusion (tMCAO) rat model and drug treatments
Rat tMCAO model was established, as described previously [32,36] .Brie y, rats were anesthetized with 1% pentobarbital sodium (50 mg/kg, i.p.), and brain ischemia was induced by inserting a 5-cm-long nylon lament (diameter, 0.24-0.28mm) into the middle cerebral artery for 2 h.Then, the lament was removed to allow reperfusion of the brain for 24 h.Sham rats were performed a comparable surgery as tMCAO rats but without the occlusion of the middle cerebral artery.The rats were randomly divided into ve groups: sham group, tMCAO group, GSS group (tMCAO rats treated with GSS), GSS+α-BTX group (tMCAO rats pretreated with α-BTX and subsequently treated with GSS), and α-BTX group (sham rats were administrated with α-BTX).Since α-BTX is not able to pass through the blood-brain barrier, we injected it into the lateral ventricle at a dose of 0.5 µg/kg body weight 30 min before tMCAO surgery.GSS (1.0 mg/Kg) was administrated via sublingual vein injection 10 min after ischemia.Rats without α-BTX and GSS treatments were parallelly given equivalent vehicle (saline).The schematic diagram of animal treatments is shown in Figure S1A and B in supplementary materials.The doses of GSS and α-BTX were chosen based on previous study and our preliminary experiments [32] .lateral ventricular injection Rats were anesthetized with 1% pentobarbital sodium (50 mg/kg, i.p.) and xed on a stereotactic apparatus (ZH-1-Lanxing, RWD, China).When bregma was exposed, a burr hole was prepared on the skull in the right hemisphere (1.6 mm lateral and 0.9 mm posterior to bregma) using a power drill (68605, ϕ 1.4 mm, RWD, China).Then an injector (5-μL micro-syringe, Hamilton, USA) connected with micro-pump (KDS310, Harvard Apparatus, USA) was placed its needle into the right lateral ventricle at 3.5 mm depth of sub-dural through the micro-hole.Placement of needle was performed using a stereotactic apparatus.The vehicle or the α-BTX was then administered into the lateral brain ventricle at a speed of 0.25 μL/min.The needle stayed in place for 5 min after injection to prevent back ow.

Neurological test
All rats were received Garcia assessment at 24h after I/R to evaluate neurobiological function.The evaluation includes six tests scoring on a scale from 3 to 18 [37] : spontaneous activity, symmetry in four limb movements, forepaw outstretching, climbing, body proprioception, and response to vibrissae touch.
In brief, rat brain was consecutively sliced into six coronal sections (2 mm thickness) and then stained with 0.5% TTC solution for 30min at 37℃ in a water bath shaker in the dark.During the staining, the slices were turned over every 5min.After the staining, the slices were washed with PBS and xed in 4% paraformaldehyde for 6h.The infarcted area and brain area in TTC-stained brain slices were measured using ImageJ 1.37a.The corrected percentage of infarct volume was calculated as the formula: infarct volume (%) = (the area of contralateral hemisphere -the area of non-infarcted ipsilateral hemisphere) /(2 * the area of contralateral hemisphere) *100% [37] .
Cell culture and treatments BV2 microglial cells were purchased from the Cell Bank of Type Culture Collection of Chinese Academy of Sciences (Shanghai, China).BV2 cells were grown in Dulbecco's modi ed Eagle's medium (DMEM) containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin in a cell incubator with 5.0% CO 2 at 37°C.Cells were seeded in culture dishes at a density of 1×10 6 cells/mL and treated with LPS (1.0 μg/mL) for 24 h to induce M1 polarization.A schematic diagram for in vitro experiments is shown in Figure S1C and D in supplementary materials.
To determine the effects of GSS on M1 polarization, cells were randomly allocated into four groups: control group (treated with vehicle), LPS group (treated with LPS), LPS+GSS group (treated with LPS plus GSS treatment, 10 μM), and GSS group (treated with GSS alone).Cells were treated with GSS for 24 h.
In order to determine whether α7nAChR-mediated NF-κB signaling was involved in the effects of GSS on microglial M1 polarization, cells were randomly divided into six groups: control group (treated with vehicle), LPS group (treated with LPS), LPS+GSS group (treated with LPS plus GSS treatment, 10μM), GSS group (treated with GSS alone), α-BTX+LPS+GSS group (pretreated with α-BTX followed by LPS and GSS treatment), and α-BTX+LPS group (pretreated with α-BTX followed by LPS incubation).Cells were pretreated α-BTX (10nM) 30min before given LPS incubation.

Western blot
Ischemic penumbra brain tissues were collected for western blotting.Brain tissues were added protein lysis buffer and homogenized on ice for 1h.BV2 cells were also lysed with protein lysis buffer on ice for 1h.Then these samples were centrifuged at 12 000 g for 15 min to obtain the supernatants.The protein concentration in samples was measured by BCA kit according to the manufacturer's instructions.

Quantitative PCR (qPCR)
Total RNA was isolated from tissues or cells using TRIzol reagent according to the manufacturer's instructions.4 μg total RNA was used to synthesize the rst strand cDNA using Reverse Transcription Master Mix kit (SYBR Ò Select Master Mix).Then PCR reaction was performed as following reaction system: cDNA 2 μL, 1x SYBRÒ 10 μL 10 μM primer 2 μl adding ddH 2 O to a total volume of 20 μL.
Parameters for PCR reaction were as bellows: pre-denatured at 95℃ for 20 min follow by 40 cycles (desaturated at 95℃ for 10s, annealed at 61℃ for 20s, and extended at 72 ℃ for 25s).PCR results were normalized to GAPDH and expressed as folds of sham or control group.The sequence of primers used in this study was shown in Table 1.

ELISA assay
The concentration of IL-1β in the culture supernatant was determined using enzyme-linked immunosorbent assay (ELISA) kit (R&D systems, Minneapolis, MN) according to the manufacturer's instructions.

Immuno uorescence
Rats were anesthetized and xed the whole animal with 4% paraformaldehyde through transcardial perfusion.The brain was then taken out, immersed in 4% paraformaldehyde, dehydrated with 10% sucrose, and cut into brain slices (30 μm thickness) by freezing microtome.Dried brain slices were hydrated with PBS at room temperature (RT) for 20min, and then the slices were incubated in 0.3 % Triton X-100/PBS solution at RT for 10-15 min.After washed with PBS for 5 min×3 times, the slices were blocked with 3% BSA at RT for 30 min and incubated with anti-CD11b antibody (1:200) at 4℃ overnight.Then the slices were washed with PBS and incubated with AlexaFluor488 mouse secondary antibody (1:2000).Finally, the slices were stained with DAPI and sealed with 50% glycerin.The immuno uorescent images were captured under a uorescence microscope (Carl Zeiss Lsm880, Germany).

Statistical analysis
Data were presented with mean ± standard deviation (SD).Statistical analysis was performed using SPSS 20.0 software.One-way ANOVA with Newman-Keuls test was used to compare the differences between the means of more than two groups.The value of P<0.05 was considered statistically signi cant.

Results
GSS treatment reduced brain injury and neurological de cit in tMCAO rats.
TTC staining and neurological de cit evaluation results showed that vehicle-administrated tMCAO rats (tMCAO group) developed signi cant brain infarction and neurological de cit at 24 h after I/R compared with sham rats, which were signi cantly inhibited by GSS treatment (Figure 1).

GSS treatment inhibited microglial M1 polarization in the brain peri-infarct area in tMCAO rats
Next, we used CD11b as a cell surface marker to determine microglial activation.As shown in Figure 2A, the morphology of microglia in sham group was rami ed with a small cell body, which was in a resting phenotype, whereas most microglial cells were activated in the peri-infarct area of brain of tMCAO rats, characterized by a rod-shape morphology with a larger cell body, a fusion of cell nucleus, and thick or less number of cellular processes, which were signi cantly inhibited by GSS treatment.In the meantime, Realtime qPCR and western blot results showed that mRNA and protein expression of M1 microglial markers (CD11b, CD40 and CD68) signi cantly increased in tMCAO group.Whereas, these changes were signi cantly attenuated by GSS treatment in tMCAO rats (Figure 2B-I).Besides, GSS treatment also signi cantly inhibited the increase of IL-1β protein expression in the peri-infarct brain region induced by tMCAO insult (Figure 2E and F).
GSS treatment restored α7nAChR protein expression while suppressed NF-κB signaling in tMCAO rats NF-κB signaling pathway is regulated by α7nAChR, attributing to M1 polarization of microglia [39,40] .Realtime qPCR results showed that mRNA expression of IKK, IκB and P65-NF-κB in the ischemic penumbra signi cantly increased in tMCAO group compared with sham group; GSS treatment signi cantly reduced mRNA expressions of these genes in tMCAO rats (Figure 3A-C).GSS treatment signi cantly inhibited tMCAO-induced increases of phosphorylated IKK and phosphorylated p65 protein, but restored IκB and α7nAChR protein expression in the ischemic penumbra region (Figure 3D-H).α7nAChR inhibitor blocked the protective effects of GSS against brain injury after tMCAO Next, we applied a speci c α7nAChR inhibitor, α-BTX, to illustrate whether GSS inhibited microglial M1 polarization and brain injury through activating α7nAChR signaling.We found that α-BTX pretreatment alone did not change the infarction area and neurological de cit in tMCAO rats but signi cantly suppressed the therapeutic effects of GSS on brain injury and neurological function (Figure 4).Pretreatment with α-BTX also reversed the effects of GSS on the morphological change of microglia (Figure 5A) as well as mRNA and protein expression of M1 microglial markers (CD11b, CD40 and CD68) and IL-1β in the ischemic penumbra of tMCAO rats (Figure 5B-J).As shown in Figure 6A and B, we observed that α-BTX pretreatment signi cantly blocked the inhibitive effects of GSS on NF-κB signaling to exhibit as increasing mRNA expressions of IKK and P65-NF-κB and phosphorylation of IKK and p65 while reducing Iκ B protein expression to a comparable level as tMCAO group.

GSS treatment inhibited LPS-induced M1 polarization of microglial cells
We further used LPS-stimulated BV2 microglial cells to rea rm the observed effects of GSS on M1 microglial polarization in vivo.ELISA results showed that LPS (1μg/mL) stimulating for 24 h signi cantly upregulated the concentration of IL-1β in the cell culture supernatant compared with vehicle-treated cells.In contrast, the increase of IL-1β protein in the supernatant was signi cantly reduced by GSS (10 μM) treatment (Figure 7A).We also applied cell immuno uorescent staining and qPCR to validate the nding of ELISA.Consistently, the upregulation of IL-1β protein (Figure 7B) and mRNA expression (Figure 7C) in BV2 induced by LPS were signi cantly inhibited by GSS treatment.Besides, LPS stimulus signi cantly increased CD40 and CD68 mRNA expression in BV2 cells, which was signi cantly inhibited by GSS treatment (Figure 7D and E).Similarly, GSS treatment also signi cantly suppressed the increase of CD40 and CD68 protein expression induced by LPS (Figure 7F-H).GSS treatment alone without LPS did not signi cantly affect the expression of M1 polarization makers (IL-1β, CD40 and CD68) compared with control (Figure 7).

GSS treatment restored α7nAChR expression while blocked NF-κB signaling in LPS-stimulating BV2 cells
We determined the intracellular localization of P65-NF-κB using immuno uorescent staining.As shown in Figure 8A, P65-NF-κB dominantly located in the cell nucleus after LPS stimulation, while it distributed mainly in the cytoplasm after GSS treatment in LPS-stimulated BV2 cells.Meanwhile, α7nAChR and IκB protein expression were signi cantly decreased, while expressions of p-IKK and p-P65-NF-κB were signi cantly increased in LPS-treated BV2 cells compared with vehicle-treated cells, which were signi cantly blocked by GSS treatment (Figure 8B-F).α7nAChR blocker antagonized the protective effects of GSS against microglial M1 polarization and NF-κB signaling Finally, we used α-BTX (an α7nAChR blocker) to determine whether the anti-M1 polarization effects of GSS was attributed to α7nAChR-mediated signaling.Both immuno uorescent staining and ELISA results showed that α-BTX pretreatment signi cantly reversed the effects of GSS on IL-1β protein expression in LPS-stimulated BV2 cells (Figure 9A and B).At the same time, α-BTX pretreatment restored the protein expressions of CD40 and CD68, which were signi cantly decreased by GSS treatment in LPS-stimulated BV2 cells (Figure 9C-E).GSS treatment signi cantly inhibited the nucleus translocation of P65-NF-κB in LPS-stimulated BV2 cells.Whereas, these effects of GSS was signi cantly attenuated by α-BTX pretreatment (Figure 10A).Western blot results also showed α-BTX pretreatment insigni cantly blunted the inhibitive effects of GSS on NF-κB signaling.As shown in Figure 10B-E, GSS treatment signi cantly reduced the levels of p-IKK and p-P65-NF-κB but increased IκB protein expression compared with LPSstimulated group.In contrast, α-BTX pretreatment dramatically blocked these effects of GSS.

Discussion
In the present study, we found that GSS treatment signi cantly reduced brain infarcted volume and improved neurological function by inhibiting microglial M1 depolarization-mediated in ammation in tMCAO rats, with an underlying mechanism through upregulating α7nAChR expression and thereby blocking NF-κB signaling.
Our previous study showed that GSS treatment reduced infarcted brain area and improved neurological recovery in tMCAO rats and the mechanism underlying the therapeutic effects was through protecting against neuronal apoptosis [32] .Besides, we also noticed that GSS treatment reduced MMP-3 and MMP-9 mRNA and protein expression [41] , provoking us to investigate the anti-in ammation effects and its underlying molecular mechanism of GSS in ischemic stroke.Our results showed that GSS treatment reversed IL-1β expression in brain peri-ischemic regions and alleviated brain injury and neurological de cits in tMCAO rats.These results demonstrate that the GSS inhibits in ammatory responses in ischemic penumbra brain regions during ischemic stroke.
Neuroin ammation occurs rapidly after the onset of ischemic stroke and lasts for the whole disease process [42] .It was reported that prolonged in ammation exacerbates brain injury and postpones brain function recovery [43] .Microglia is one of the resident immune cells in the brain.Their activated states account for the dual effects of neuroin ammation in ischemic stroke: Microglial M2 phenotype is dominant in the early stages of ischemic stroke; it gradually switches to M1 phenotype [1] .M2-depolarized microglia exert anti-in ammation effects by producing IL-10 and TGF-β.In contrast, M1-depolarized microglia is proin ammatory.Thereby, inhibiting microglia depolarization to M1 phenotype is considered as a potential strategy for the treatment of ischemic stroke [3] .In this study, we used the microglial marker CD11b to determine the activation of microglia in the ischemic brain [44] .Our results showed that microglia in ischemic penumbra regions was signi cantly activated, showing a rod-shaped morphology with a large cell body [44] .In comparison, GSS treatment signi cantly inhibited the activation of microglia in ischemic penumbra regions where most microglia were rami ed shape as showing in sham rats [45,46] .
In this study, we used CD40 and CD68 further veri ed microglial M1 depolarization [47][48][49] .Our results showed that CD40 and CD68 expression signi cantly increased in the petri-ischemic regions in tMCAO rats, while GSS treatment dramatically blocked the expression of these markers.Besides, we validated our ndings through in vitro experiments: GSS treatment signi cantly blocked LPS-induced M1 depolarization and IL-1β expression in BV2 microglial cells.The M1 depolarized microglia are characterized by a set of markers (CD16, CD32, CD40, CD68 and CD86) and secrete proin ammatory cytokines such as IL-1β and TNF-α [50,51] .Hence, our results demonstrate that GSS treatment suppresses microglial M1 depolarization in tMCAO rats.Our previous study showed that GSS increased IL-10 expression in peri-infarcted regions in ischemic stroke rats (unpublished data), implying that GSS might promote microglia depolarization to M2 phenotype, which requires future research.NF-κB, as a critical transcription factor, contributes to neuroin ammation through regulating microglia activation.In canonical NF-κB activation pathways, phosphorylated IKK protein promotes the degradation of IκB protein, thereby resulting in the release and subsequent nucleus translocation of P65-NF-κB protein, which regulates a variety of target genes expression [52] .This study showed that GSS treatment signi cantly increased IκB protein expression but reduced the phosphorylation levels of IKK and P65-NF-κB protein in both tMCAO rats and LPS-stimulated BV2 cells.Moreover, GSS treatment also remarkedly inhibited the nucleus translocation of P65-NF-κB induced by LPS in BV2 cells.Our results suggest that GSS inhibits microglial M1 depolarization through blocking NF-κB signaling activation.α7nAChR is a crucial receptor mediating cholinergic anti-in ammatory pathway.The stimulation of α7nAChR inhibits the activation of NF-κB signaling in microglia, reduced M1 microglia in the peri-infarct brain regions, and reduced brain injury in permanent MCAO (pMCAO)mice [2][25] .In this study, we found that α7nAChR protein expression signi cantly decreased in the ischemic penumbra regions in tMCAO rats, while it was restored by GSS treatment.Besides, GSS treatment also signi cantly inhibited the decrease of α7nAChR expression induced by LPS in BV2 cells.These results indicate that GSS upregulates microglial α7nAChR function in ischemic stroke.After pretreatment with α7nAChR inhibitor, α-BTX, we found that GSS lost the protection against brain infarction and neurological dysfunction and the inhibitive effects on microglial M1 depolarization and NF-κB signaling in both in vivo and in vitro experiments.These ndings demonstrate that α7nAChR-mediated inhibition of NF-κB signaling is a vital molecular mechanism underlying the regulation of microglial M1 depolarization by GSS.

Conclusions
Taken together, as shown in Figure 11, our results from in vivo and in vitro experiments showed that GSS treatment upregulated α7nAChR and thereby inhibited the activation of NF-κB and microglial M1 depolarization, contributing to the neuroprotective effects in tMCAO rats.This study uncovers a novel molecular mechanism underlying the anti-ischemic stroke effects of GSS, demonstrating that GSS protected against brain ischemic injury by suppressing neuroin ammation via α7nAChR-mediated inhibition of NF-κB signaling.

Figures
Figures

Figure 3 Effects
Figure 3

Figure 4 Alpha
Figure 4

Figure 5 Alpha
Figure 5