Introduction
Traumatic brain injury (TBI) has always been a major cause of death and disability worldwide [
1]. Due to the complex pathophysiological mechanisms of TBI, effective pharmacotherapy is still lacking. A number of studies have identified that neuroinflammation following TBI plays an important role in secondary pathophysiological damage which might have a more serious and profound influence compared with that of the primary injury.
Microglia are the resident immune cells that initiate inflammation and immune response. Microglia are activated, migrate to the injury site rapidly and play a crucial role in neuroinflammation following TBI [
2]. Recent studies have discovered that the role of microglia in TBI is a double-edged sword based on their two polarizations: the pro-inflammatory (M1-like) phenotypes and anti-inflammatory (M2-like) phenotypes [
3]. Regulating the polarization of microglia, especially promoting the M2 phenotype transformation following TBI, may be a possible therapeutic strategy. Although some molecules have been identified as selectively increasing M2 polarization [
4], such as IL-4, IL-10 and TGF-β, effective medicine for the clinic is still lacking.
Astrocytes are the most abundant cell type in the central nervous system and play a crucial role in maintaining microenvironment stability and neural circuit function [
5‐
7]. Previous studies demonstrated that astrocytes and microglia are active participants in various pathological conditions including brain trauma. Activated astrocytes produce many regulatory factors that influence CNS immunity and provide negative feedback to activated microglia [
8,
9].
Recently, exosomes, a type of extracellular vesicles, have been identified as signalling conveyors in cell-to-cell communication [
10]. Via cargo proteins, RNAs and miRNAs, exosomes act as key players in triggering, transferring and regulating immune responses to neighbouring cells [
11]. Most cells in the CNS have been reported to secrete exosomes into the extracellular environment [
12]. Although some studies show that astrocyte-derived exosomes are powerful mediators involved in neuronal plasticity, immune response and neuronal survival under multiple pathological conditions [
13], little is known about their function in microglial activation and their role in mediating neuroinflammation after brain trauma.
In the present study, we used an in vitro TBI model to investigate the composition of astrocyte-derived exosomes and their effect on microglial activation, which could lead us to further understand the crosstalk between microglia and astrocytes. Furthermore, the potential functional miRNAs were investigated in a mouse TBI model in vivo to explore a new therapeutic target for neuroinflammation-associated injury.
Methods
Animal TBI model and experimental grouping
All experimental procedures were conducted following the existing rules of Huazhong University of Science and Technology and the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. Male adult mice (C57BL/10ScNJ; age, 10–12 weeks; weight, 20–22 g) were obtained from Huazhong Keji Co. All animals were housed in a controlled environment (temperature, 22 ± 3 °C, under a 12-h light/dark cycle) and were provided standard rodent nutrition and water. The mouse model of TBI was induced according to previous reports [
14]. Anaesthesia was surgically induced with chloral hydrate (400 mg/kg body weight) administered intraperitoneally (i.p.), and then the mice were subjected to a unilateral, moderately controlled cortical impact (CCI) of 2.0 mm depth at 3.5 m/s and 500 ms dwell time using the TBI 0310, a pneumatic impacting device (Precision Systems and Instrumentation, Fairfax Station, VA) with a hard stop Bimba cylinder (Bimba Manufacturing, Monee, IL). The size of the bevelled impactor was 5 mm. All craniotomies were placed midway between the bregma and lambda sutures in the left hemisphere of the brain. A total of 100 animals were randomly divided into 4 groups: sham group (
n = 20), sham + miR-873a-5p agomir group (
n = 20), TBI group (
n = 30), and TBI + miR-873a-5p agomir group (
n = 30). All investigators were blinded to the treatment groups during animal surgery, data collection and analysis.
Primary microglia and astrocyte culture
Primary microglia and astrocytes were obtained and purified from C57BL/6 mice on postnatal days 1 to 2 as previously described [
14]. Briefly, the cerebral cortex was collected and cut into 1-mm
3 pieces. After digestion with 0.25% trypsin and DNase for 10 min, the cell suspensions were passed through a 70-mm Nylon mesh, and the digestion was ended with DMEM supplemented with 10% FBS and 1% penicillin/streptomycin. These cells were collected by centrifugation and seeded into a culture flask. Microglia within the astrocyte monolayer were removed by shaking at 220 rpm for 40 min after 10 days and were then re-cultured in 6- or 24-well plates. The remaining mixed glial cells in the flask were shaken at 220 rpm for 18 h continuously to remove oligodendroglia, and then the remaining cells in the flask were astrocytes with over 90% purity. Astrocytes and microglia were cultured at 37 °C in a 5% CO
2 atmosphere for a subsequent experiment.
Clinical specimen collection and ethics statement
The present study was conducted in accordance with the Declaration of Helsinki and was approved by the Research Ethics Board of Tongji Hospital. Written informed consent was obtained from all individuals who were included in the study. TBI patients were diagnosed according to the World Health Organization criteria. The clinical specimens of damaged brain tissue were taken from 15 patients who were operated on in the neurosurgical emergency department of Tongji Hospital (additional table
1). These tissues were either necrotic brain tissue or severe oedema around the lesion that needed to be removed. The expression of miR-873a-5p in brain tissue was detected by quantitative real-time PCR.
Cell transfection of the miR-873a-5p mimic and intracerebroventricular infusion
The miR-873a-5p mimic, negative control mimic (NC mimic) and agomir were purchased from RiboBio (China). They were dissolved and diluted according to the instructions provided by the manufacturer. The microglia were transfected with 100 nM aliquots of either the miR-873a-5p mimic or NC mimic using RiboFECT™ CP (RiboBio, China) as per the manufacturer’s protocol. After dissolving and mixing the miR-873a-5p agomir, it was allowed to stand for 15 min at room temperature and was then used for lateral ventricle injection. A Hamilton syringe (Gaoge, China) was inserted at 0.5 mm posterior and 1.0 mm lateral to the bregma and 3.0 mm ventral to the skull under the guidance of a stereotaxic instrument (RWD Life Science). A single dose of miR-873a-5p agomir (5 nM) was infused into the lateral ventricle 20 min after CCI.
Brain extracts (Ext) were made as previously described [
15]. Briefly, TBI was induced in C57BL/6 mice, and after 1 day, the cortices were collected before they were surgically induced with chloral hydrate (400 mg/kg body weight) for anaesthesia. A total of 4 ml complete medium per cortex were added into glass tubes, and the supernatant was collected after each cortex was ground fully with a glass grinding rod and centrifuged at 1000 rpm for 10 min. Then, 1 ml of the supernatant was aliquoted in 1.5 ml EP tubes and kept at − 80 °C for usage.
Astrocyte-derived exosome isolation and transmission electron microscopy analysis
The astrocyte-derived exosome isolation procedures were performed at 4 °C as described in the literature [
16]. Briefly, supernatants collected from cultured astrocytes were first filtered through a 0.2-μm filter to remove the large debris and dead cells. Small cell debris was removed by centrifugation at 10,000
g for 30 min, and then the supernatants were recentrifuged at 100,000
g for 3 h. The supernatants generated at this step were stored at 4 °C for future use as exosome-free controls (the average storage time was no more than 1 week). The pellets were resuspended in 30–50 μl of PBS and stored at − 80 °C for another usage. For the transmission electron microscopy (TEM) morphology investigation, the pellets obtained as described above were subjected to uranyl acetate negative staining on For-mvar carbon-coated 400 mesh copper electron microscopy grids (FCF400-Cu, Electron Microscopy Sciences, Hatfield, PA). Twenty microliters of the sample was applied to the grid and incubated for 5 min at room temperature, and then the excess solution on the grid was wicked off and dried for 30 min with filter papers. An equal part of 10% uranyl acetate was added to the grid for negative staining for 5 min. The preparations obtained were examined at 70 kV with a Philips 208 electron microscope (Philips, Bothell, WA) with an AMT digital imaging system (Advanced Microscopy Techniques Corp., Woburn, MA). Protein concentrations of exosome preparations were determined using the BSA assay. For neural cell treatment with astrocyte exosomes, we diluted the collected exosome-enriched fractions of the stored supernatant as noted above, and the supernatant without exosomes was used as a control. These media were then added to the cultured neural cells.
Exosome labelling and uptake
Exosomes were fluorescence-labelled with PKH26 (Sigma-Aldrich), a lipophilic dye, according to the manufacturer’s protocol. Exosomes were incubated in diluent C and PKH26 for 5 min at room temperature. PKH26-labelled exosomes were diluted with PBS and ultracentrifuged at 150,000g 4 °C for 70 min to enable removal of unincorporated dye contamination from exosome labelling reactions. Subsequently, the PKH26-labelled or denatured exosomes were incubated with primary microglia for 24 h. After the cells were fixed, we stained the cells with 4′,6-diamidino-2-phenylindole (DAPI) and phalloidin (Sigma-Aldrich). The images were obtained under a confocal microscope.
miRNA microarray analysis
The miRNA microarray analysis was performed by GeneChem (Shanghai, China). The samples of exosomes derived from astrocytes were divided into 2 groups: the Ext group and the CON group, which were with or without exposure to brain extracts, respectively. The quality and integrity of the RNA extracted from the exosomes were evaluated first. Next, 200 ng of total RNA was labelled with the GeneChip 39 IVT Express Kit (Thermo Fisher Scientific) and hybridized to the GeneChip miRNA 3.0 Array (Thermo Fisher Scientific), which covered 1188 mature mouse miRNAs and 889 pre-miRNAs. RNA molecules were then polyadenylated, followed by a ligation step with a biotin-labelled DNA molecule attached. The labelled RNA was finally washed and stained in the GeneChip Fluidics Station 450 and scanned in the GeneChip Scanner 3000 (Thermo Fisher Scientific).
Western blot analysis
Protein concentrations of medium supernatant, cells and animal tissue were determined by using a BSA kit, and then the protein samples were diluted with 5× sample buffer solution, separated by electrophoresis in a 12% separation gel for 90 min and blocked with 1× PBS containing 5% (w/v) non-fat dried milk (PM) for 1 h at room temperature. Then, the cells were incubated with primary antibodies (iNOS, 1:1000; HMGB1, 1:1000; IL-1β, 1:1000; Arg1, 1:1000; CD9, 1:1000; CD63, 1:1000; p-NK-κB, 1:1000; NK-κB, 1:1000; IL-4, 1:1000; and IL-10, 1:1000—Abcam, USA; MyD88, 1:1000; p-ERK, 1:1000; and ERK, 1:1000—Cell Signaling Technology, Beverly, MA, USA) at 4 °C overnight. The membranes were then washed and incubated with HRP-conjugated anti-rabbit or mouse antibody (1:1000, Earth-Ox Life Sciences, Millbrae, CA) for 1 h at room temperature and then exposed and photographed on a Gene Gnome exposure instrument. Finally, the expression of the proteins was standardized for densitometric analysis to β-actin levels.
Modified neurological severity score test
To evaluate the neurological functional outcomes, the modified neurological severity score (mNSS) test was performed. The tests were carried out before CCI and at days 1, 3, 7 and 14 after CCI. The scale was graded from 0 to 18 (normal score, 0; maximal deficit score, 18). The mNSS comprises the motor (muscle status and abnormal movement), sensory (visual, tactile and proprioceptive) and reflex reactions and balance tests. One point is awarded if the mice are unable to perform the test or lack an expected reaction; thus, the higher the score, the more severe the injury.
Measurement of the brain water content
Brain oedema was evaluated by measuring the brain tissue water content using the wet-dry weight method as described previously [
17]. Animals were sacrificed 7 days after TBI, and the left brain cortical tissue was collected. The brains were harvested, and the pons and olfactory bulbs were removed. The tissue was positioned directly over the injury site, covering the contusion and the penumbra. The fresh tissue was weighed to record the wet weight, dried for 72 h at 80 °C and then weighed again to record the dry weight. The brain water content was calculated using the following formula: (wet weight − dry weight)/wet weight] × 100%.
Quantitative real-time PCR
Total RNA was isolated from the microglia or tissue using TRIzol (Invitrogen, Carlsbad, CA, USA) before being washed with PBS and reverse-transcribed to cDNA with the PrimeScriptTM RT Reagent Kit (Thermo, USA) according to the datasheet from the manufacturer. Gene products were then amplified by quantitative real-time PCR on an ABI-Prism 7500 Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA) using SYBR Premix Ex Taq TM II (Takara). The relative level of miR-873a-5p (MQP-0101, RiboBio, China) was normalized to the expression of control U6 snRNA (MQP-0201, RiboBio, China). Other mRNAs were normalized to the internal standard GAPDH. The primers are as follows. Data were analysed using the 2−ΔΔCt method.
IL-1β | Forward, AGAACCAAGCAACGACAAAATAC |
Reverse, GTATTGCTTGGGATCCACACTC |
IL-6 | Forward, GGAGCCCACCAAGAACGATA |
Reverse, CAGGTCTGTTGGGAGTGGTA |
TNF-α | Forward, GGATTATGGCTCAGGGTCCA |
Reverse, ACATTCGAGGCTCCAGTGAA |
iNOS | Forward, CATTCAGATCCCGAAACGCT |
Reverse, TGTAGGACAATCCACAACTCGC |
IL-4 | Forward, GTAGGGCTTCCAAGGTGCTTC |
Reverse, CATGATGCTCTTTAGGCTTTCCAG |
IL-10 | Forward, ACCTGGTAGAAGTGATGCCC |
Reverse, ACACCTTGGTCTTGGAGCTT |
Arg1 | Forward, GCATATCTGCCAAAGACATCGT |
Reverse, TCTTCCATCACCTTGCCAATC |
CD32 | Forward, TGTCACTGGGATTGCTGTCG |
Reverse, CCCCAGAGGGCTGTCTGTAC |
CD206 | Forward, CGTTTCGGTGGACTGTGGA |
Reverse, GTTGTGGGCTCTGGTGGG |
GAPDH | Forward, TGAAGGGTGGAGCCAAAAG |
Reverse, AGTCTTCTGGGTGGCAGTGAT |
Immunofluorescence staining
Cells or tissues were incubated with GFAP, Iba1, CD9, Arg1 or iNOS antibodies (Abcam, USA) overnight at 4 °C and then incubated with conjugated secondary antibody for 1 h at room temperature in the dark. After several washes with PBS, the slides were incubated with DAPI for 3 min and then mounted in glycerol. After mounting, immunofluorescent signalling was observed with an Olympus Fluoview laser scanning confocal microscope (Olympus, Tokyo, Japan), and the percentages of positive cells were counted in a blinded manner using ImageJ.
Statistical analyses
All data are expressed as the mean ± standard deviation. The programmes GraphPad and InStat were used for statistical analyses. One-way ANOVA followed by Newman-Keul’s post hoc test was used for multiple comparisons. A non-paired t test was used when two groups were compared. Two-way ANOVA was used to compare the NDs between the three groups at different time points.
Discussion
In this study, we determined that activated astrocytes can regulate microglial polarization by releasing miRNAs that are transported by exosomes in mice TBI models in vitro and in vivo. Although the exosomes derived from BE-stimulated astrocytes require further exploration, our findings are still consistent with evidence that (1) BE-stimulated astrocyte-derived exosomes were enriched with microRNA-873a-5p, which can promote microglial M2 phenotype transformation in the early stage of TBI; (2) miR-873a-5p effectively suppressed pro-inflammatory factors and promoted anti-inflammatory factors released from the microglia via inhibiting the phosphorylation of ERK and the NF-κB signalling pathway; and (3) the anti-neuroinflammatory effect of miR-873a-5 can attenuate the mNSS and brain oedema in mice TBI model.
Although the microglia play a key role in the inflammatory and immune response following TBI, there are still no effective therapeutic targets to regulate their activation and function. Recent evidence showed that the microglia acted as a double-edged sword in many pathophysiological conditions based on their differential polarization: the pro-inflammatory M1 and anti-inflammatory M2 phenotypes. This M1/M2 phenotype transformation can also be observed in neuroinflammation following TBI. Our previous study indicated that promoting microglial transformation from the M1 to M2 phenotype is beneficial for attenuating the immune response and ameliorating neurological impairment after brain trauma [
14]. However, up to now, the regulation of the microglial phenotype is lacking as a specific effective drug treatment.
Some research declares that astrocytes have a powerful ability to regulate microglial activation. A number of studies indicate that astrocytes in the brain are active participants in both propagating and regulating neuroinflammation [
8]. Astrocytes become activated by inflammatory mediators and cytokines and produce many regulatory factors that may influence CNS immunity. It was reported that proinflammatory cytokines such as TNFα and IL-1β may stimulate astrocytes toward a predominantly harmful reactive state (A1). Meanwhile, stimulation with the anti-inflammatory cytokine IL-4 can lead to a more protective or restorative reactive state A2 [
18,
19]. The results showed that both A1 phenotype biomarker (iNOS and TNF-α) and A2 biomarker (TGF-β and Arg-1) were increased after TBI (additional figure
1). In our previous study (the manuscript under submission), IL-1β and IL-4 were upregulated in the “brain extracts”. This may explain why both the A1 and A2 astrocyte phenotypes existed. It has been reported that the addition of conditioned media from astrocytes to microglia cultures increased antioxidant and anti-inflammatory gene expression by providing negative feedback to activated microglia [
9]. However, the underlying mechanism has not been well defined. Some recent studies proposed that exosomes take part in cell-to-cell communication. For example, Hu et al. demonstrated that EVs derived from astrocytes exposed to morphine can be taken up by microglial endosomes, leading, in turn, to activation of Toll-like receptor 7 (TLR7) [
20]. Sobue et al. reported that the overexpression of MHCI in astrocytes affects microglial proliferation as well as neuronal numbers and spine densities, thereby leading to social and cognitive deficits in mice, possibly via exosomes created by astrocytes [
21]. However, there are few reports about the immunoregulatory effects of exosomes from astrocytes on microglial activation. The highlight of this study is that we used BE to stimulate primary cultured astrocytes to simulate traumatic brain injury in vivo. The results confirmed that stimulated astrocytes produce and secrete more exosomes that promote the microglia phenotype transformation from M1 to M2.
The cargo shuttled by exosomes is complicated, including proteins, miRNAs, and mRNAs. We focused on miRNAs whose potential effects have been increasingly appreciated. Recently, some astrocyte-derived miRNAs have been investigated. Xu et al. reported that miR-92b-3p from preconditioned astrocytes can be carried to neurons by exosomes and ameliorate OGD-induced cell death and apoptosis [
22]. miR-7 from astrocytes can cause the downregulation of neuronal neuroligin 2 (NLGN2) and ultimately lead to synaptic alterations after HIV-1 infection [
23]. Through high-throughput miRNA sequencing, we found that the expression of miR-873a-5p in exosomes was significantly changed after BE insult. This microRNA exhibited anti-inflammatory effects by suppressing LPS-induced microglial activation. This is the first report to demonstrate the function of miR-873a-5p in regulating inflammation during brain trauma.
As a tumour-associated miRNA, miR-873a-5p has been well recognized to regulate tumour proliferation and invasion by modulating certain molecules in colorectal cancer, endometrial cancer and hepatocellular carcinoma [
24‐
26]. Cui et al. reported that the expression level of miR-873a-5p in breast tumours was much lower than normal. Moreover, the overexpression of miR-873a-5p or its mimic inhibits breast cancer cell proliferation both in vitro and in an in vivo mouse model [
27]. It has been established that miR-873 increases lung adenocarcinoma cell proliferation and migration by targeting SRCIN1, and its expression is decreased in glioblastoma multiforme (GBM) tumour tissues and cell lines [
28]. Liu et al. first reported that miR-873 in astrocytes induced by IL-17 promotes inflammatory cytokine production and exacerbates demyelination in experimental autoimmune encephalomyelitis (EAE) through the A20/NF-κB pathway [
29]. However, the functional subtype of miR-873 and its correlation with microglia are unclear. Our results indicated that miR-873a-5p expression was increased, and release by activated astrocytes might be a negative feedback mechanism to the microglia-mediated immune response during TBI insult. This might be a potent treatment target for regulating neuroinflammation. This phenomenon is also confirmed by LPS-stimulated microglia in vitro. These data are inconsistent with the results reported by Liu’s group. The possible reasons are as follows: Liu et al. did not indicate which subtype of miR-873 plays a pro-inflammatory role. Also, miR-873a may have a pro-inflammatory effect on astrocytes and an anti-inflammatory effect on microglia, which may also explain the inflammatory balance in the central nervous system. These data together demonstrate a new role for miR-873a-5p in regulating microglial activation. The inflammatory response is the main cause of damage after TBI, and miR-873 can exert neuroprotective effects by inhibiting this inflammation.
NF-κB is a classical inflammatory signalling pathway that is associated with neurotoxicity [
30]. This study showed that the NF-κB signalling pathway was likely one of the important targets of miR-873a-5p. Suppression of the NF-κB signalling pathway with miR-873a-5p could transform microglial polarization from M1 to M2 under TBI conditions. This effect of miR-873a-5p can provide new strategies for future clinical drug development.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.