Endothelial cell-derived exosomes protect SH-SY5Y nerve cells against ischemia/reperfusion injury.

Cerebral ischemia is a leading cause of death and disability. A previous study indicated that remote ischemic postconditioning (RIP) in the treatment of cerebral ischemia reduces ischemia/reperfusion (I/R) injury. However, the underlying mechanism is not well understood. In the present study, the authors hypothesized that the protective effect of RIP on neurological damage is mediated by exosomes that are released by endothelial cells in femoral arteries. To test this, right middle cerebral artery occlusion/reperfusion with RIP was performed in rats. In addition, an I/R injury cell model was tested that included human umbilical vein endothelial cells (HUVECs) and SH-SY5Y cells. Both the in vivo and in vitro models were examined for injury. Markers of exosomes (CD63, HSP70 and TSG101) were assessed by immunohistochemistry, western blot analysis and flow cytometry. Exosomes were extracted from both animal serum and HUVEC culture medium and identified by electron microscopy. They investigated the role of endothelial cell-derived exosomes in the proliferation, apoptosis, cell cycle, migration and invasion of I/R-injured SH-SY5Y cells. In addition, apoptosis-related molecules caspase-3, Bax and Bcl-2 were detected. RIP was determined to increase the number of exosomes and the expression levels of CD63, HSP70 and TSG101 in plasma, but not in brain hippocampal tissue. The size of exosomes released after I/R in HUVECs was similar to the size of exosomes released in rats subjected to RIP. Endothelial cell-derived exosomes partly suppressed the I/R-induced cell cycle arrest and apoptosis, and inhibited cell proliferation, migration and invasion in SH-SY5Y nerve cells. Endothelial cell-derived exosomes directly protect nerve cells against I/R injury, and are responsible for the protective role of RIP in I/R.

Introduction

Cerebrovascular disease (CVD) is a serious human health hazard with a high incidence rate, often resulting in disability and death. Currently, it is the third leading cause of death in humans (1–3). In China, there has been a rapid increase in the incidence of CVD, with ~2 million people suffering from this disease (3,4). Most survivors of CVD experience permanent neurological damage, with 70% losing their ability to work and 40% left with severe disability that impacts the quality of life. Cerebral ischemia accounts for >75% of CVD patients (1). No effective therapy is currently available for treating cerebral ischemia (5). Evidence from clinical and in vivo studies suggest that apoptosis of nerve cells produce significant benefits for cerebral ischemia injury (6). A possible mechanism is thought to be associated with endothelial dysfunction in cerebral ischemia (7). Remote ischemic postconditioning (RIP) in the treatment of CVD relieves ischemia/reperfusion (I/R) injury (8–10). However, it is not known if RIP induces neuroprotection against cerebral ischemia and what the underlying mechanism is. In the present study, the authors hypothesized that the protective effect of RIP on neurological damage is mediated by exosomes derived from endothelial cells in femoral arteries.

Exosomes are secreted from cells, and contain proteins, DNA, mRNA and some non-protein coding RNAs. They carry material and transducer information, is the carrier between cells for material and information transduction (11). Exosomes play an important role in the cellular microenvironment and are well-studied multi-functional extracellular vesicles. In cancer cells, the exosomes of 5-FU-resistant CCL227-RH cells, are devoid of microRNA-200, and accelerate the formation of circular chemorepellent-induced defects in vascular endothelial cell monolayers as compared to exosomes from naïve CCL227 cells (12). The paracrine effects of human umbilical vein endothelial cells (HUVECs) improve the generation of endothelial cells from cord blood circulating endothelial progenitor cells and may include the role of exosomes (13). A recent study reported that exosomes extracted from adipose-derived mesenchymal stem cells play a protective role against nerve injury induced by glutamate (14). Endothelial cell-derived exosomes potently increase the proliferation, migration, secretion of matrix metalloproteinase (MMP)-1, MMP-3 and nuclear factor (NF)-κB activity in the mesenchymal stem cells, stimulating local trophic support (15). Mesenchymal stem cells promote nerve growth through the support of Schwann cells, secreted neurovascular factors and possibly trans-differentiation into Schwann-like cells (16). Condition medium from cells that were treated under hypoxic conditions increased the number of differentiating neurons in vitro (17). Exosomes isolated from different kinds of cells all express the characteristic proteins CD63, HSP70 and TSG101 (18).

In the present study, the authors established an animal model of I/R injury with RIP in rats, and a cell model of I/R injury in HUVECs and SH-SY5Y cells. The levels of protein markers of exosomes were analyzed and measured and then exosomes were extracted from both rats and HUVECs. The role of endothelial cell-derived exosomes in proliferation, apoptosis, cell cycle, migration and invasion of SH-SY5Y cells undergoing I/R was evaluated. In addition, the authors detected the apoptosis-related molecules caspase-3, Bax and Bcl-2. These findings help to understand the mechanism underlying the protective role of remote ischemia in I/R injury.

Materials and methods

Animals

A total of 30 adult Sprague-Dawley (SD) rats (15 male and 15 female) at (10 weeks old) were used, ranging in weight from 220 to 250 g that were provided by the Laboratory Animal Center, Nanchang University (Nanchang, China). Animals were randomly divided into three groups that included the sham-operated (sham) group, the middle cerebral artery occlusion and reperfusion (MCAO/R) group and the RIP group, with 10 rats in each group. All the rats received humane care, according to the criteria outlined in the Guide for the Care and Use of Laboratory Animals, published by the National Institute of Health (NIH publication 86-23 revised 1985). The animal protocol was approved by the Animal Ethics Committee of the Second Affiliated Hospital of Nanchang University (Nanchang, China).

Establishment of the MCAO/R model

Transient cerebral I/R (MCAO/R) was induced, as previously described (19,20). Rats were anesthetized with 7% chloral hydrate (0.5 ml/100 g) and subjected to the operation. Bilateral femoral arteries were exposed before the occlusion of middle cerebral artery. In the RIP group, the middle cerebral artery was subject to a RIP protocol, which comprised an occlusion for 2 h followed by clamping of the femoral artery, repeated for 3 cycles. Rats in the sham group were subjected to the same surgical procedure as rats in the MCAO/R (model) group without occlusion of the artery. Rats in the RIP group were subjected to the same surgical procedure as the rats in the MCAO/R group without the use of remote ischemic conditioning. Following reperfusion for 24 h, the infarct size, number of apoptotic cells in the rat hippocampal tissue, expression of exosome markers and release of exosomes were determined.

Cell culture

The human SH-SY5Y (ATCC, CRL-2266) nerve cells were cultured in F12 + Dulbecco's modified Eagle's medium (DMEM; 1:1, v/v; Invitrogen, Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen, Thermo Fisher Scientific, Inc.), L-glutamine and antibiotics and incubated in a humidified atmosphere of 5% CO2 at 37°C. HUVECs (American Type Culture Collection, Manassas, VA, USA; PCS-100-010) were cultured in DMEM supplemented with 10% FBS (Invitrogen, Thermo Fisher Scientific, Inc.) and 10% growth factor containing epidermal growth factor, fibroblast growth factor-2, cAMP, heparin, hydrocortisone, penicillin, streptomycin and amphotericin-B (21).

Establishment of the model of I/R injury

To test the appropriate I/R time for subsequent experiments, oxygen glucose deprivation and reoxygenation of SH-SY5Y and HUVEC cells was performed. Cells were placed in a hypoxia chamber (1% O2, 5% CO2 at 37°C, cat. no. 27310; StemCell Technologies, Vancouver, BC, Canada) at different time-points followed by 24 or 21 h of reoxygenation, respectively (22).

Evaluation of infarct volume and brain injury

To evaluate the infarct in brain tissue, rat brains were sectioned, fixed and stained by using 2,3,5-triphenyltetrazolium chloride (TTC, cat. no. G3005; Beijing Solarbio Science & Technology, Co., Ltd., Beijing, China). The TTC-stained brains were separated and the images scanned. The infarct volume was analyzed by using ImageJ (version: v1.48u; National Institutes of Health, Bethesda, MD, USA) using the following formula: Infarct volume (%) = (contralateral hemisphere volume − surgery side without infarct volume)/contralateral hemisphere volume × 100% (20). Pathological changes in brain hippocampal tissues were evaluated by hematoxylin and eosin (H&E) staining. The apoptotic cells in the rat hippocampal tissues were also measured using the TUNEL assay kit (206386; Abcam, Cambridge, MA, USA) according to the manufacturer's instructions. Five fields of each group were observed under an optical microscope.

Exosome preparation

The release of exosomes was evaluated by detecting markers of exosomes that included CD63, HSP70 and TSG101 in brain tissue sections using immunostaining, and in rat plasma using western blot analysis and flow cytometry. Exosomes were obtained from plasma and cell culture media (FBS without exosomes). Exosomes were prepared by standard differential centrifugation as follows: centrifugation at 3,000 × g for 20 min at 4°C to obtain plasma, then 10,000 × g for 20 min at 4°C to remove cells and platelets, then twice at 100,000 × g for 70 min at 4°C with a SW-41 rotor, followed by washes with phosphate-buffered saline (PBS). The vehicle control consisted of PBS alone. A total of 5 µg of exosomes were incubated with 1.25 µl aldehyde/sulfate latex beads, 4% w/v (4 µm, A37304; Invitrogen, Thermo Fisher Scientific, Inc.) for 15 min, and then incubated with anti-CD63 (5 µl/100 µl sample, cat. no. ab18235; Abcam) anti-Hsp70 (5 µl/100 µl sample, cat. no. ab183435; Abcam) and anti-TSG101 (5 µl/100 µl sample, cat. no. ab207663; Abcam), respectively. After fixing with 1% PFA, flow cytometry was performed with a FACSCalibur flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA). The results were analyzed by the FlowJo 7.0.

Electron microscopy

Electron microscopy was performed on a JEOL 1010 transmission electron microscope (JEOL, Ltd., Tokyo, Japan) at 120 kV after a standard staining procedure with 1% uranyl acetate. The exosomes were imaged at 49,000 × g and their shapes and sizes were determined.

Cell cycle detection

Cell cycle status was determined with a FACSCalibur flow cytometer (BD Biosciences). Following trypsinization, cells were adjusted to a concentration of 1×106 cells/ml and treated using Cycletest Plus DNA reagent kit (BD Biosciences), according to the manufacturer's instructions. The cell cycle status was analyzed by flow cytometry using propidium iodide (PI).

Cell apoptosis detection

Cell apoptosis was determined by the Annexin V assay using the FACScan (BD Biosciences) flow cytometer and the apoptotic rate was calculated. After trypsinization and PBS washes, cells were suspended in Annexin V binding buffer, and were incubated with fluorescein isothiocyanate (FITC)-conjugated Annexin V followed by PI (Nanjing KeyGen Biotech, Co., Ltd., Nanjing, China). In addition, apoptosis was evaluated using Hoechst 33258 staining (Invitrogen, Thermo Fisher Scientific, Inc.) and by measuring the expressions of Bax, Bcl-2 and caspase-3.

Proliferation detection

Cell proliferation was determined by using a Cell Counting Kit-8 (CCK-8; Beyotime Institute of Biotechnology, Haimen, China) and EdU staining assay (Invitrogen, Thermo Fisher Scientific, Inc.). For the CCK-8 assay, 1×104 cells were seeded in 96-well plates and then incubated for 24, 48 and 72 h. Then, the absorbance of the solution was measured at 450 nm using the microplate reader. For further confirmation of cell proliferation of exosomes, treated SH-SY5Y cells were fixed and stained with EdU.

Scratch wound assay

For the scratch wound assay (23,24), prior to the endothelial exosome treatments, I/R-injured SH-SY5Y nerve cell monolayers were scratched by a pipette tip after cells reached confluence in 6-well plates, followed by two washes with PBS to remove cell debris. Following treatment for the indicated times, images were taken with an invert contrast microscope and digitized using a digital camera. The wound areas were calculated to evaluate the cell migration capacity by using ImageJ (National Institutes of Health).

Invasion assay

Invasion assays were performed, as previously described (25,26). Complete medium was added to the lower chamber. Matrigel was used to pre-coat the wells and then 1×105 cells were seeded. After 48 h, cells were fixed in methanol and stained with crystal violet. Invaded cells were quantified in at ×10 magnification by an optical microscope. Results were presented as mean ± standard deviation of three independent experiments repeated in triplicate.

Immunocytochemistry

The expression of caspase-3 was assessed by immunocytochemistry. Cells slides were fixed in 4% formaldehyde for 10 min and pre-incubated in 0.5% Triton X-100 and 1.5% bovine serum albumin (BSA; Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) for 15 min at room temperature. Immunostaining was performed by incubating overnight at 4°C with anti-caspase-3 (1:50, cat. no. 9662; Cell Signaling Technology, Danvers, MA, USA) followed by Alexa Fluor 594 secondary antibody (1:500, cat. no. R37117; Invitrogen, Thermo Fisher Scientific, Inc.) for 1 h at room temperature, washed twice with PBS. DAPI (5 g/l) was used for counterstaining for 5 min, and observed by a fluorescence microscope (Olympus Corp., Tokyo, Japan).

Reverse transcription-quantitative polymerase chain reaction

Total RNA was extracted using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Next, reverse transcription was performed using a cDNA synthesis kit (Takara Bio, Inc., Shiga, Japan). qPCR reactions were conducted using Takara SYBR Premix Ex Taq II (Tli RNaseH Plus; Takara Bio, Inc.) on a PCR amplifier (CFX-96 Content Real-time System; Bio-Rad Laboratories, Hercules, CA, USA). The forward and reverse primer sequences are as follows: caspase-3 forward, 5′-AGCAAACCTCAGGGAAACATT-3′ and reverse, 5′-CTCAGAAGCACACAAACAAAACT-3′; BAX forward, 5′-CCCGAGAGGTCTTTTTCCG-3′ and reverse, 5′-GCCTTGAGCACCAGTTTGC-3′; of Bcl-2 forward, 5′-GTGTGGAGAGCGTCAACC-3′ and reverse, 5′-CTTCAGAGACAGCCAGGAGA-3′; GAPDH forward, 5′-TGTTCGTCATGGGTGTGAAC-3′ and reverse, 5′-TGGCATGGACTGTGGTCAT-3′. The data shown is representative of three independent experiments.

Western blot analysis

Tissues or cells were homogenized in ice-cold RIPA lysis buffer (cat. no. P0013B; Beyotime, Beijing, China). After determining the concentration of protein by Bio-Rad protein assay kit (cat. no. 5000002), 30–50 µg protein were loaded onto a 8–12% SDS-PAGE and separated by electrophoresis at 80 V for 30 min and then 120 V for 1.5 h. Proteins were transferred to nitrocellulose at 250 V for 1 h, blocked with 5% non-fat dry milk for 1 h, and incubated with anti-CD63 (1:1,000, cat. no. ab134045; Abcam), anti-HSP70 (1:1,000, cat. no. ab53496; Abcam), anti-TSG101 (1:1,000, cat. no. ab125011; Abcam), anti-cleaved caspase-3 (1:1,000, cat. no. sc-271759; Santa Cruz Biotechnology, Inc., Dallas, TX, USA), anti-Bax (1:1,000, cat. no. sc-4239; Santa Cruz Biotechnology, Inc.), anti-Bcl-2 (1:1,000, cat. no. sc-7382; Santa Cruz Biotechnology, Inc.) or anti-GAPDH (1:1,000, cat. no. sc-47724; Santa Cruz Biotechnology, Inc.) overnight. After three times washes, membranes were incubated with horseradish peroxidase-conjugated secondary antibody (1:10,000, cat. no. A4416 and A6154; Sigma-Aldrich) for 1 h at room temperature. Immunoblots were detected using enhanced chemiluminescence (Beyotime Institute of Biotechnology). The quantification of western blot analysis was measured by the Image-Pro Plus 6.0.

Statistical analysis

Data were obtained from at least three independent experiments and expressed as means ± standard deviations. Statistical analysis was performed by one-way analysis of variance (ANOVA), followed by the Student-Newman-Keuls multiple comparison post hoc test or by the Mann-Whitney test by using the SPSS software (SPSS 19.0; SPSS, Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Results

The changes in the MCAO/R (model) and the right middle cerebral artery occlusion/reperfusion (RIP) groups were assessed (Fig. 1). The authors used TTC staining and revealed that greater infarct growth occurred in rats in the model group compared with those in the RIP group (Fig. 1A and B). No change in infarct size was observed in rats in the sham group. A TUNEL assay was used to detect apoptosis in the rat hippocampal tissues (Fig. 1C and D). A significant increase in the apoptotic rate was observed in the model and RIP groups. The apoptotic rate in the RIP group was lower than that in the model group. H&E staining demonstrated that RIP significantly reduced the pathological changes in brain hippocampal tissues (Fig. 1E). These results suggested that RIP has a significant protective effect on I/R injury.

Figure 1

RIP protects rats against ischemia-reperfusion injury. The MCAO/R (model) and the right middle cerebral artery occlusion/reperfusion (RIP) were established. The infarct size and pathological changes in brain hippocampal tissue were observed. (A) Image of infarct and (B) infarct volume in the sham, MCAO/R model and RIP groups. The white area indicates cerebral infarction and the red area indicates healthy tissue. (C) TTC staining of the rat hippocampal tissue and (D) the apoptotic rates. (E) Hematoxylin and eosin staining of brain hippocampal tissue (original magnification, ×400). *P<0.01, **P<0.01 vs. sham. RIP, remote ischemic postconditioning; MCAO/R, middle cerebral artery occlusion and reperfusion.

MCAO/R increases exosomes in plasma, but not in brain hippocampal tissue

The exosome makers and exosomes in both brain hippocampal tissue and plasma were analyzed (Fig. 2). No difference was observed in the expressions of exosome makers CD63, HSP70 and TSG101 by immunohistochemical staining, in brain hippocampal tissue among the sham, model and RIP groups (Fig. 2A). However, in rat plasma, the expressions of CD63, HSP70 and TSG101 were upregulated in the model group, and were further increased in the RIP group by both western blot analysis (Fig. 2B) and flow cytometry (Fig. 2C). Transmission electron microscopy revealed a significant increase in particles in the RIP group, compared with the model group (Fig. 2D). The authors concluded that RIP promotes the release of exosomes (40–100 nm) in rats subjected to I/R.

Figure 2

The release of exosomes in ischemia/reperfusion rats. Brain hippocampal tissue and plasma were collected from the sham, model and remote ischemic postconditioning groups. To analyze the release of exosomes, exosome makers were detected by IHC, western blotting, flow cytometry, and the size of the exosomes were evaluated by electron microscope. (A) IHC staining of exosome makers CD63, HSP70 and TSG101 in brain hippocampal tissue (original magnification, ×400). Expression levels of CD63, HSP70 and TSG101 in plasma detected by (B) western blotting (C) and flow cytometry. (D) TEM particle size calculated by transmission electron microscope. IHC, immunohistochemistry.

Isolation of exosomes from the endothelial cell model of ischemia and reperfusion

An endothelial cell model of ischemia and reperfusion was established (Fig. 3A). The authors observed a significant increase in the rate of apoptosis with ischemia time as shown in Fig. 3A and B. Following 6 or 8 h of ischemia followed by 24 h of reperfusion, an increase in the rate of apoptosis was observed. The exosomes (40–100 nm) from the endothelial cells in ischemia (6 h)-reperfusion (24 h) models in vitro were collected and observed by transmission electron microscope (Fig. 3C).

Figure 3

Exosomes derived from the endothelial cell model of ischemia and reperfusion. The endothelial cell model of ischemia and reperfusion was established. (A) Apoptosis of endothelial cells in the ischemia-reperfusion model and (B) apoptosis rates. Exosomes were collected from the cells and observed by transmission electron microscope. (C) Observation of exosomes by transmission electron microscope. *P<0.01, **P<0.01 vs. control. FITC, fluorescein isothiocyanate.

I/R-induced cell injury in SH-SY5Y nerve cells

The authors established the SH-SY5Y nerve cell model of ischemia and reperfusion (Fig. 4). A significant increase was observed in the rate of apoptosis with ischemia time. Following 6 or 8 h of ischemia followed by 24 h of reperfusion, an increase in the rate of apoptosis was observed. Finally, the SH-SY5Y nerve cells in the ischemia (6 h)-reperfusion (24 h) models in vitro were used for the following tests.

Figure 4

Establishment of the SH-SY5Y nerve cell model of ischemia and reperfusion. Cells were exposed to ischemia at different time-points followed by 24 h of reperfusion, apoptosis was detected by flow cytometry and the apoptotic rate was calculated. *P<0.01 vs. control.

Effects of exosomes extracted from endothelial cells on cell cycle, proliferation and apoptosis of SH-SY5Y nerve cells

The authors measured the cell cycle status, proliferation and apoptosis of SH-SY5Y nerve cells (Fig. 5). I/R induced the G0/G1 arrest in SH-SY5Y nerve cells, and this effect was diminished by exosomes released from endothelial cells (Fig. 5A). Endothelial exosomes promoted the increase in SH-SY5Y cells in S phase.

Figure 5

Exosomes from endothelial cells diminished the I/R-induced G0/G1 arrest and apoptosis, and inhibited proliferation in SH-SY5Y nerve cells. Effects of exosomes derived from the endothelial cell model of ischemia and reperfusion on cell cycle, apoptosis, and proliferation of SH-SY5Y nerve cells were detected. (A) Cell cycle status. The percentage of G0/G1 phase cells was higher in I/R-injured cells than control cells. *P<0.05. Exosomes released from endothelial cells reduced the percentage of G0/G1 phase cells significantly compared with cells exposed to I/R, *P<0.05. The percentage of S phase cells was lower in the I/R-injured cells than the control cells; ##P<0.01. Exosomes released from endothelial cells increased the percentage of S phase cells significantly compared with cells exposed to I/R, ##P<0.01. (B) CCK-8 assay. I/R inhibited the proliferation of SH-SY5Y nerve cells at 48 h (control vs. I/R, **P<0.01) and 72 h (control vs. I/R, *P<0.05). Exosomes released from endothelial cells reversed the inhibition of cell proliferation induced by I/R at 48 h (I/R vs. I/R+Exo, **P<0.01) and 72 h (I/R vs. I/R+Exo, *P<0.05). (C) EdU assay. (D) Immunofluorescence staining of caspase-3. (E) Cell apoptosis by flow cytometry. Both I/R and I/R+Exo induced cell apoptosis comparing with control (control vs. I/R, **P<0.01; control vs. I/R+Exo, *P<0.05). Exosomes released from endothelial cells reversed the induction of cell apoptosis induced by I/R (I/R vs. I/R+Exo, #P<0.05). (F) Hoechst 33258 staining.

Both the CCK-8 (Fig. 5B) and EdU (Fig. 5C) assays demonstrated that I/R inhibited the cell proliferation of SH-SY5Y nerve cells, and this inhibition decreased after exosomes were released from endothelial cells.

I/R induced expressions of caspase-3 in SH-SY5Y nerve cells (Fig. 5D). Exosomes released from endothelial cells significantly inhibited I/R-induced caspase-3. Consistent with changes in caspase-3 expression, endothelial exosomes reduced I/R-induced apoptosis by both Annexin V FITC-PI flow cytometry (Fig. 5E) and Hoechst 33258 assays (Fig. 5F).

Exosomes extracted from endothelial cells promote migration and invasion of SH-SY5Y nerve cells

I/R inhibited wound healing and invasion of SH-SY5Y cells (Fig. 6). Following treatment with exosomes released from endothelial cells, I/R-inhibited cell migration (Fig. 6A) and invasion (Fig. 6B) were partly recovered, suggesting that the exosomes from endothelial cells promote migration and invasion of SH-SY5Y nerve cells.

Figure 6

Exosomes from endothelial cells promote migration and invasion of SH-SY5Y nerve cells. Effects of exosomes derived from the endothelial cell model of ischemia and reperfusion on cell migration and invasion of SH-SY5Y nerve cells. To measure cell migration, a (A) wound assay and (B) wound area calculation was conducted (original magnification, ×40). To measure cell invasion, (C) invasion assay and (D) invasion of cells per field. *P<0.05, **P<0.01 vs. SH-SY5Y. #P<0.05 vs. SH-SY5Y I/R. I/R, ischemia/reperfusion; Exo, exosomes.

Exosomes extracted from endothelial cells inhibit expression of cleaved caspase-3 and Bax, and upregulate expression of Bcl-2 in SH-SY5Y nerve cells

I/R induced expression of Bax and caspase-3 and inhibited the expression of Bcl-2 at both the mRNA and protein levels in SH-SY5Y nerve cells (Fig. 7). After treatment with exosomes released from endothelial cells, expression levels of Bax and caspase-3 were downregulated, and expression levels of Bcl-2 were upregulated, suggesting that exosomes from endothelial cells promote the expression of Bcl-2 and inhibit the expression of Bax and caspase-3 in SH-SY5Y nerve cells.

Figure 7

Influence of endothelial exosomes on the expression of Bax, Bcl-2, caspase-3 in I/R SH-SY5Y cells. Effects of exosomes derived from the endothelial cell model of ischemia and reperfusion on expression of Bax, Bcl-2, caspase-3 in I/R SH-SY5Y nerve cells. The mRNA expressions of (A) Bax, (B) Bcl-2 and (C) cleaved caspase-3. (D) Western blot analysis of Bax, Bcl-2 and caspase-3. *P<0.05, **P<0.01 vs. SH-SY5Y. #P<0.05 vs. SH-SY5Y I/R. I/R, ischemia/reperfusion; Exo, exosomes.

Discussion

In the present study, the authors presented evidence that supports the hypothesis that exosomes released by endothelial cells in femoral arteries mediate the protective effect of RIP on neurological damage. MCAO/R is a widely used model for cerebral I/R injury (27). The authors established a right middle cerebral artery occlusion/reperfusion rat model with RIP. In rats that were subjected to RIP, all the infarct sizes, the rate of nerve cell apoptosis and the pathological changes in brain hippocampal tissue were significantly reduced. These results validated the idea that RIP serves a protective role in I/R injury (28–30). Moreover, RIP is neuroprotective in young male rodents after experimental stroke, and is also effective after embolic stroke in ovariectomized female mice (31). In clinical settings, RIP protects the aged liver against I/R injury, and is further associated with improvement in vascular function (32). Although the mechanism involved is not understood, here, the authors provided direct evidence that indicates that vascular endothelial cells secrete exosomes during cerebral I/R injury that protect neurobehavioral outcomes.

Exosomes are secreted by various cells, including endothelial cells; the condition medium of endothelial cells is typically rich with exosomes (33). Exosomes are microextracelluar vesicles with a diameter of ~30–100 nm. They play important roles in material and signaling transduction between cells in the microenvironment (11). The findings demonstrated that exosomes express CD63, HSP70 and TSG101. The expressions of CD63, HSP70 and TSG101 did not increase in brain hippocampal tissue in the sham, model and RIP groups, but increased in plasma in the RIP group, indicating that RIP promotes the release of exosomes. Furthermore, the authors detected exosomes that were 40–100 nm in size using transmission electron microscopy, in plasma of rats that underwent cerebral I/R injury. The paracrine effects of HUVECs potentially improve endothelial cell generation from cord blood circulating endothelial progenitor cells, and may contain the role in the production of exosomes (13). Condition medium from cells treated under hypoxic conditions increased the observed number of differentiating neurons in vitro (17). The exosomes were isolated from endothelial cells during ischemia and reperfusion injury and the size of extracted exosomes was comparable to those observed in rats that underwent cerebral I/R injury.

Exosomes comprise of proteins, DNA, mRNA and some non-protein coding RNA, and have multiple functions in different types of cells (11,34). For example, endothelial cell-derived exosomes potently increase the proliferation, migration, secretion of matrix metalloproteinase (MMP)-1, MMP-3 and nuclear factor-κB activity in mesenchymal stem cells, stimulating local trophic support (15). Mesenchymal stem cells support nerve growth in conjunction with Schwann cells, secreted neurovascular factors and possibly trans-differentiation into Schwann-like cells (16). A recent study reported that exosomes extracted from adipose-derived mesenchymal stem cells have a protective effect against nerve injury induced by glutamate, which may be mediated by activating the PI3/K-Akt signaling pathway (14). Consistent with previous studies (35,36), the current results suggested that cell cycle arrest and apoptotic rate of SH-SY5Y nerve cells are promoted, and proliferation, migration and invasion are inhibited during ischemia and reperfusion injury. Treatment with I/R endothelial cell-derived exosomes partly recovered the I/R-induced injury to SH-SY5Y nerve cells. The role of endothelial cell-derived exosomes on apoptosis of I/R-injured SH-SY5Y nerve cells was further validated by observing the expression levels of caspase-3, Bax and Bcl-2.

Taken together, the results strongly suggest that endothelial cell-derived exosomes directly protect nerve cells against I/R injury by promoting cell growth, migration and invasion, as well as by inhibiting cell apoptosis. Endothelial cell-derived exosomes could potentially be developed as a novel treatment strategy in treating neurological damage during I/R injury.