Background
Estrogen confers strong neuroprotection against cerebral ischemic injury in rodents and humans via estrogenic receptors α and β (ERα, ERβ) and G protein-coupled receptor 30 (GPR30) [
1‐
3]. However, the peripheral actions of estrogen on reproductive organs and the potential to increase the risk of endometrial and breast cancer limit its clinical application [
4].
GPR30 is a novel estrogenic receptor that is highly expressed in the brain [
5,
6]. Our team members originally found that the GPR30 agonist G1 reduced ischemia reperfusion injury in ovariectomized (OVX) mice [
7]. Moreover, an acute bolus dose of G1 following ischemic injury significantly enhanced neuronal survival and decreased cell apoptosis [
8,
9]. Thus, GPR30 plays an important role in the acute neuroprotective effect of estrogen. However, the explicit molecular mechanisms underlying the acute neuroprotective effects of GPR30 remain unclear.
Toll-like receptor 4 (TLR4) is primarily expressed in microglia and mediates microglial activation [
10]. The inhibition of TLR4 has beneficial effects on tissue homeostasis and attenuates both myocardial and spinal cord ischemic injury [
11,
12]. E2 (17-β-estradiol) and G1 reduced the TLR4 levels in peripheral macrophages, resulting in anti-inflammatory effects [
13], suggesting that GPR30 participates in the peripheral anti-inflammatory pathway. Therefore, we investigated the roles of microglial GPR30 in the inflammatory pathway of the central nervous system (CNS), the neuroprotective effects of estrogen against ischemic injury, and the potential molecular mechanisms.
Methods
Animals
One hundred and seventy 6- to 8-week-old female C57BL/6 mice (20 ± 2 g) were obtained from the Laboratory Animal Center of Fourth Military Medical University. All animal experimental procedures were approved by the Ethics Committee for Animal Experimentation of the Fourth Military Medical University and followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals (
https://grants.nih.gov/grants/olaw/Guide-for-the-Care-and-use-of-laboratory-animals.pdf).
Cell cultures and neuron-microglia co-cultures
Primary mouse microglial cultures were harvested from 1- to 2-day-old neonatal C57BL/6 pups. Briefly, after removal of the meninges and hippocampus, the cortical tissues were subjected to enzymatic digestion and mechanical isolation. Then, the mixed cortical cells were passed through a 70-μm nylon mesh cell strainer and seeded into a cell culture flask in Dulbecco’s Modified Eagle’s Medium (DMEM, HyClone) containing 10% FBS (Gibco) and 1% penicillin/streptomycin. Fourteen to 15 days later, the mixed glial cultures were shaken on an orbital shaker at 220 rpm for 1 h. Then, the detached microglial cells in the supernatant were collected and reseeded into cell culture containers. The purity of the microglia in culture was more than 95%, as confirmed by staining with the microglial marker Iba1 (see Additional file
1: Figure S1).
Neuronal cultures were harvested from the cerebral cortices of E16–E17 mouse embryos. Dissociated cells were obtained and then seeded onto poly-D-lysine-coated culture plates or glass cover slips. The Neurobasal Medium (Gibco) contained 2% B27 (Invitrogen), 1% glutamine, and 1% penicillin and streptomycin. The purity of the neuronal cultures was more than 95%, as confirmed by staining with neuronal and glial markers.
For the indirect neuron-microglia co-cultures, neurons were seeded in 24-well plates and incubated for 3 days. Then, primary microglia (microglia:neurons = 1:2) were added to 0.4-μm pore-sized Transwell inserts (Costar, USA). The neurons and microglia were co-cultured for another 2 days before oxygen-glucose deprivation (OGD) treatment.
Groups and drug treatment
In vivo, all mice were ovariectomized (OVX) to eliminate the influence of circulatory estrogen. After 1 week, all mice except for those in the sham group were subjected to middle cerebral artery occlusion (MCAO) injury. Then, mice were randomly divided into the following groups: sham, vehicle (cottonseed oil), E2, G1, and ICI182780+E2. The drugs E2 (17-β-estradiol), G1 (GPR30 agonist), and ICI182780 (ERα/β antagonist) were administered immediately upon reperfusion. In particular, 235 μl of E2 (1 mmol/l) was administered via intraperitoneal injection, and 2 μl of G1 (0.1 g/l) and 2.8 μl of ICI182780 (2 mmol/l) were administered via left intracerebroventricular infusion.
In vitro, primary microglial cells were randomly divided into the following groups: control, vehicle (dimethyl sulfoxide, DMSO), E2, G1, and ICI182780+E2. All cells except for those in the control group were subjected to OGD treatment. E2 (10 nmol/l), G1 (10 nmol/l), and ICI182780 (1 μmol/l) were administered immediately upon reintroduction. In addition, primary microglial cells were randomly divided into three groups: control, vehicle (DMSO), and G1. All cells except for those in the control group were exposed to lipopolysaccharide (LPS, 10 ng/ml) for 12 h to induce microglial inflammation. G1 (10 nmol/l) was administered immediately upon the end of the LPS exposure.
The drugs E2 (#10006315, Cayman Chemical, USA), G1 (#10008933, Cayman Chemical, USA), and ICI182780 (#10011269, Cayman Chemical, USA) were dissolved in cottonseed oil and dimethyl sulfoxide for in vivo and in vitro experiments, respectively, and LPS (Sigma) was dissolved in cell culture medium. All drugs were administered as previously described [
7,
14‐
17]. All experiments followed the principles of randomization and double blindness.
Ovariectomization of the female mice (OVX)
All mice were subjected to OVX to eliminate the influence of circulatory estrogen on the results, and the operation was conducted as previously described [
7]. Briefly, mice were anesthetized with 10% chloral hydrate (0.03 ml/kg), and the ovaries were then exposed via a midline dorsal skin and muscle layer incision at the right or left side of the vertebral column. After ligating the uterine horn, the ovaries were cut, and the wounds were closed.
Intracerebroventricular cannulation and infusion
The lateral ventricle was targeted − 0.2 mm anterior-posterior, + 1.0 mm medial-lateral, and − 2.2 mm dorsal-ventral from the bregma. Cannulas (62003, RWD Life Science) were held in the left hemispheres of the brain [
7]. Mice were given 7 days to recover from the surgery. Afterward, the mice were subjected to MCAO injury, and drugs were injected into the lateral ventricle at a rate of 1 μl/min immediately upon reperfusion.
Middle cerebral artery occlusion (MCAO)
MCAO was conducted as previously described [
7]. General anesthesia was induced by 10% chloral hydrate (0.03 ml/kg), and the body temperatures of the OVX female mice were maintained at 37.0–37.5 °C. A midline skin incision was performed to expose the left common carotid artery (CCA). Then, a nylon monofilament was inserted from the left CCA to the origin of the MCA. The regional cerebral blood flow (rCBF) was monitored using laser Doppler flowmetry (Periflux System 5000, PERIMED, Stockholm, Sweden). Mice were included only if blood flow was reduced to more than 20% of the preischemic baseline level during ischemia and was restored to up to 70% of the preoperative level during reperfusion. After 1 h of MCAO, cerebral blood flow was recovered by removing the filament. The physiological parameters are provided in Additional file
1.
Neurobehavioral evaluation
Neurobehavioral evaluation was performed 24 h after reperfusion according to a previous report [
7]. The neurological evaluation was scored using a 5-point scale: 0, no neurological deficit; 1, failure to extend left forepaw fully; 2, circling to the left; 3, inability to bear weight on the left; 4, no spontaneous walking with depressed level of consciousness. The performer was blind to the treatments.
Assessment of infarct volume
Infarct volume was assessed by 2,3,5-triphenyl-2H-tetrazoliuM chloride (TTC) staining as previously described [
7]. Briefly, mice were decapitated, and their brains were rapidly removed and cut into 1-mm-thick coronal brain slices. Then, the slices were immersed in 2% TTC at 37 °C until an obvious color change appeared, transferred to 4% paraformaldehyde for 6 h of fixation, and photographed using a digital camera (Canon IXUS 220HS). The percent infarct was calculated as [(
VC − VL)/
VC] × 100, where
VC is the volume of the control hemisphere and
VL is the volume of the noninfarcted tissue in the lesioned hemisphere.
Nissl staining
Nissl staining was applied to observe morphological changes in cells within the ischemic penumbra. The brains were perfused with cold 4% paraformaldehyde. Then, the brains were removed after post-fixation and dehydrated in 20 and 30% sucrose solutions. Subsequently, 10-μm-thick sections were prepared using the Leica CM1900 frozen slicer. The experimental steps were strictly performed according to the manufacturer’s manual of the Nissl staining kit (#G1430, Solarbio, China). The cells with large cell bodies, a rich cytoplasm, and one or two large, round nuclei were intact neurons, while the cells with shrunken cell bodies, condensed nuclei, dark cytoplasm, and empty vesicles were damaged neurons, as described in the manufacturer’s manual. Cells in five different fields were counted.
Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining
TUNEL staining was performed using an In Situ Cell Death Detection Kit (Roche Diagnostics, Mannheim, Germany) according to the manufacturer’s instructions. TUNEL-positive neurons/DAPI were regarded as an apoptosis index.
OGD/R
Briefly, the cells were seeded in DMEM without glucose (Gibco), transferred into a modular incubator chamber, and flushed with 3 l/min of a 95% N2 and 5% CO2 gas mixture for 15 min at room temperature. The chamber was then sealed and placed in a 37 °C container. OGD was carried out for 1, 2, or 4 h. Following OGD, the cells were incubated with normal media for an additional 12 h of reperfusion under normal conditions.
CCK-8 cell viability assay
A Cell Counting Kit (Seven Sea Biotech, Shang Hai, China) was used to assess cell survival. The experimental steps were strictly performed according to the manufacturer’s manual. Briefly, 50 μl of CCK-8 solution was added to 500 μl of medium solution in each neuronal culture well of a 24-well plate and incubated for 4 h at 37 °C. The absorbance at 450 nm was measured with a microplate reader (Infinite M200, TECAN).
LDH release assay
The LDH-Cytotoxicity Colorimetric Assay Kit II (#K313-500, BioVision, USA) was used to detect cell injury. The experimental steps were strictly performed according to the manufacturer’s manual. Briefly, medium solution (10 μl per well) was added to each neuronal culture well of a 24-well plate in an optically clear 96-well plate. Then, 100 μl of LDH Reaction Mix was added to each well, mixed, and incubated for 30 min at room temperature. The absorbance at 450 nm was measured with a microplate reader (Infinite M200, TECAN).
Cell fraction assay
Twelve hours after primary microglia were subjected to G1 treatment, the cytoplasmic and nuclear extracts of primary microglial cells were prepared in accordance with the protocol provided in the NE-PER Nuclear and Cytoplasmic Extraction Reagent Kit (Pierce, Rockford, IL, USA). Briefly, 200 μl of ice-cold cytoplasmic extraction reagent I was added to the washed microglia and incubated on ice for 10 min, and 11 μl of ice-cold cytoplasmic extraction reagent II was then added and incubated on ice for 1 min. The tube was then centrifuged for 5 min at maximum speed in a microcentrifuge (16,000×g). The supernatant fraction (cytoplasmic extract) was immediately transferred to a clean pre-chilled tube, and 100 μl of ice-cold nuclear extraction reagent was added to the insoluble fraction by vortexing for 15 s every 10 min for a total of 40 min. The tube was centrifuged at maximum speed in a microcentrifuge for 10 min. The nuclear extract fraction was then moved to a clean pre-chilled tube. All extracts were analyzed by Western blotting.
Quantitative (real-time) polymerase chain reaction (real-time PCR)
Primary microglial cells were collected from each group. Total RNA was isolated using a TaKaRa MiniBEST Universal RNA Extraction Kit (code no. 9767, TaKaRa, Japan) and quantified. Then, 1000 ng of total RNA was reverse transcribed in a 20-μl reaction at 37 °C for 15 min followed by 85 °C for 5 s using PrimeScript RT Master Mix (code no. RR036A, TaKaRa, Japan). Real-time PCR was performed in a 25-μl reaction according to the manufacturer’s manual for SYBR Premix Ex Taq (Tli RNaseH Plus) (#RR820A, TaKaRa, Japan). The TLR4 mRNA primer sequences [TaKaRa Biotechnology (Dalian) Co., Ltd.] were as follows: F: 5′-GGGCCTAAACCCAGTCTGTTTG-3′, R: 5′-CTTCTGCCCGGTAAGGTCCA-3′. The reaction was performed at 95 °C for 3 min followed by 40 cycles of 95 °C for 10 s and 55 °C for 30 s on the IQ5 Multicolor Real-Time PCR Detection System (Bio-Rad, USA). The primers and sequences derived from the relative mRNA expression were analyzed using the formula 2−(Ct target gene − Ct reference gene).
ELISA
The levels of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) were determined by ELISAs (R&D Systems, Minneapolis, MN, USA). The experimental steps were strictly performed according to the manufacturer’s manual.
Immunofluorescence staining assay
Immunofluorescence staining was performed on frozen coronal sections of mouse brains or on primary microglia plated on cover slips. The mouse brains were fixed with 4% paraformaldehyde. After post-fixation and concentration gradient dehydration, the brains were cut into 10-μm-thick sections using a Leica CM1900 frozen slicer. The cells were fixed with 4% paraformaldehyde for 30 min. The brain sections and cell cover slips were washed three times with PBS and then incubated overnight at 4 °C in a humidified atmosphere with primary antibodies. The following primary antibodies were used: mouse anti-TLR4 (1:100; Santa Cruz, USA), rabbit anti-TLR4 (1:100; Santa Cruz, USA), rabbit anti-GPR30 (1:200; Abcam, England), goat anti-Iba1 (1:200; Novus, USA), rabbit anti-Iba1 (1:500; Wako, Japan), and mouse anti-NeuN (1:200; Millipore, USA). Then, the samples were incubated with mixtures of Alexa-488 (red, Invitrogen) or Alexa-594 (red, Santa Cruz) and Alexa-647 (green, Invitrogen)-conjugated donkey anti-goat or anti-rabbit and donkey anti-mouse secondary antibodies for 2 h in the dark at room temperature. Finally, the sections were photographed using an Olympus BX51 (Japan) fluorescence microscope. The average area of single microglial cells was measured by ImageJ software.
Western blotting
Expression of the GPR30, Iba1, and TLR4 proteins in penumbra and primary microglia was measured by Western blotting as previously described [
2]. The extracted proteins were separated by 10% SDS-PAGE and electrically transferred to polyvinylidene difluoride membranes. Then, the membranes were blocked with TBST containing 5% nonfat dry milk for 1 h at room temperature. The following primary antibodies were used: rabbit anti-GPR30 (1:1000; Abcam, England), rabbit anti-Iba1 (1:500; Wako, Japan), rabbit anti-TLR4 (1:200; Santa Cruz, USA), rabbit anti-NF-κB (Cell Signaling Technology; 1:500), mouse anti-β-actin (1:1000; Cell Signaling Technology, USA), rabbit anti-GAPDH (1:1000; Cell Signaling Technology, USA), and rabbit anti-Histone H2A.X (Cell Signaling Technology; 1:1000). The membranes were shaken at 60 rpm/min at 4 °C overnight and incubated with an IRDye secondary anti-rabbit or mouse antibody (Thermo Scientific, USA) for 1 h. Protein bands were visualized using the LI-COR Odyssey System (LI-COR Biotechnology, USA).
Statistical analysis
GraphPad Prism 7.0 was used for statistical analyses. Data were collected by two independent investigators who were blind to the drug treatments. Comparisons among multiple groups were performed using one-way ANOVA followed by Tukey’s post hoc test. Neurological deficit scores are presented as medians and were analyzed using two-tailed Mann-Whitney U tests, and the other values are presented as the mean ± SD. p values < 0.05 were considered statistically significant.
Discussion
We investigated the roles of microglial GPR30 in the neuroprotection of estrogen against ischemic injury in vitro and in vivo. We found that GPR30 was highly expressed in microglia and significantly increased after ischemic injury. Meanwhile, the activation of GPR30 significantly improved the neurological deficit and alleviated neuronal injuries. Moreover, the activation of GPR30 significantly reduced the TLR4 expression levels and NF-κB activity, relieved microglial activation, and decreased the release of TNF-α, IL-1β, and IL-6 in ischemic penumbra and microglia. Thus, microglial GPR30 exerts neuroprotective effects against cerebral ischemic injury by inhibiting TLR4-mediated microglial inflammation, suggesting that GPR30 may be a potential therapeutic target in the treatment of ischemic stroke.
GPR30 was expressed in microglia, which was confirmed by RNA-sequencing transcriptome studies using primary microglia or microglial cell lines and immunofluorescence staining performed in our study [
18,
19]. GPR30 participates in the acute neuroprotection of estrogen and regulates neuronal functions, such as neurotransmitter release and hippocampal synaptic plasticity [
20]. An acute application of the GPR30 agonist G1 following ischemic injury reduced neuronal damage in both aged rats and adult mice [
7,
9]. GPR30 activation also improved neuronal survival in the hippocampus and striatum following global cerebral ischemia [
8]. The acute neuroprotective effects of GPR30 have great clinical significance because estrogenic compounds could be used to specifically activate GPR30 without the undesirable side effects of estrogen.
Estrogen provided neuroprotection by inhibiting inducible nitric oxide (iNOS) and TNF-α in microglia [
21,
22], suggesting that estrogen participates in the brain’s immune system [
23,
24]. However, the receptor mediating the anti-inflammation effect of estrogen remains controversial. Because both ERα and ERβ are expressed in immunocytes, several studies have indicated that ERα plays a key role in inhibiting NF-κB and iNOS expression in macrophages [
23,
25], while other studies have demonstrated that ERβ plays an anti-inflammatory role in microglia, macrophages, and astrocytes [
26]. Interestingly, several studies also indicated that E2 downregulated TNF-α via the GPR30-mediated Ca
2+ signaling pathway in macrophages [
27]. We found that the activation of microglial GPR30 reduced the release of TNF-α, IL-1β, and IL-6 and decreased microglial activation, further confirming that GPR30 also mediates the anti-inflammatory effect of estrogen. However, whether microglial GPR30 has a higher affinity for ERα or ERβ requires further investigation.
As innate immunocytes, microglia were excessively activated and released robust inflammatory cytokines, such as IL-1, IL-6, TNF, and iNOS, immediately after ischemic stroke, which contributed to neuronal apoptosis and thus aggravated brain injury [
28]. Studies have shown that TNF-α, IL-1β, and IL-6 expression is elevated 0.5, 6, and 3.5 h after stroke, respectively [
29]. The fact that inflammatory cytokines were produced by microglia centrally places these cells in potential future stroke therapy [
28]. In our study, we found that activating microglial GPR30 relieved inflammatory responses and alleviated neuronal injury. However, neuronal GPR30 activation may have indirect effects on microglia activation because GPR30 is also expressed in neurons and modulates synaptic plasticity in the mouse hippocampus [
30]. Therefore, a mouse model with a conditional knockout of the gpr30 gene in microglia is needed to better verify the specific neuroprotective effect of microglial GPR30 in the future. Moreover, whether the neuroprotection effect of microglial GPR30 has priority over neuronal GPR30 and the interrelation between the activation of neuronal GPR30 and microglial GPR30 requires further investigation.
GPR30 performed its functions by targeting downstream molecules. The PI3K/Akt pathway is involved in the anti-apoptotic effect of GPR30 in spinal neurons [
31]. G1 regulates hippocampal synaptic plasticity by inducing brain-derived neurotrophic factor (BDNF) release [
20]. Moreover, GPR30 ameliorated blood-brain barrier permeability by inhibiting vascular endothelial growth factor A expression [
32]. In our study, microglial GPR30 exerted neuroprotective effects by inhibiting TLR4-mediated microglial inflammation. Several potential pathways may be involved in GPR30-induced TLR4 downregulation in microglia. First, GPR30 may indirectly modulate the transcriptional activity of TLR4 by activating the second messenger signaling cascades. GPR30 activation has been shown to lead to the activation of adenylate cyclase and a subsequent increase in cAMP levels [
33]. GPR30 activation also induced the activation of PI3-kinase/Akt-dependent signaling and transactivation of the epidermal growth factor receptor via Src protein kinase [
33‐
35]. Moreover, GPR30 increased Bcl-2 and BDNF by regulating the PI3K/Akt and MAPK/ERK signaling pathways [
36]. Whether these second messenger signaling molecules are required for the GPR30-mediated downregulation of TLR4 in microglia needs to be determined. Second, GPR30 may inhibit TLR4 expression via “bridge molecules,” i.e., Rab proteins. Members of the Rab protein family are small GTPases that are widely expressed in the CNS [
37]. Rab8b is a GPR30-binding protein that has been confirmed by a mammalian two-hybrid system [
19]. More importantly, Rab proteins participate in TLR4-mediated inflammation. Rab8a and Rab10 decreased membrane TLR4 expression and its downstream pro-inflammatory cytokines [
38,
39], and Rab7b reduced TLR4 expression levels in macrophages [
37]. These studies indicate that Rab proteins may act as a molecular link between GPR30 and TLR4.
Our study does have some limitations. First, using a mouse model with a conditional knockout of the gpr30 gene in microglia could better verify the effect of microglial GPR30. Second, the mechanisms underlying the GPR30-mediated downregulation of TLR4 expression require further clarification.