Electroacupuncture Enhance Therapeutic Efficacy of Mesenchymal Stem Cells Transplantation in Rats With Intracerebral Hemorrhage

Previous studies have showed the beneficial effects of mesenchymal stem cells (MSCs) on experimental intracerebral hemorrhage (ICH) animal. Enhancement of the treatment efficacy of MSCs in ICH is essential, considering the diseases association with high rates of disability and mortality. Some auxiliary methods to enhance the beneficial efficacy of MSCs have been introduced. However, the effect of electroacupuncture (EA) on the therapeutic efficacy of MSCs transplantation in hemorrhagic stroke and its potential mechanism is not explored. ICH rat models were established using collagenase and heparin. 48 h after ICH induction, the rats were randomly divided into model control (MC), MSCs transplantation (MSCs), EA stimulation (EA) and MSCs transplantation combined with EA stimulation (MSCs + EA) groups. We used mNSS test and gait analysis to assess neurological function of rats, and PET/CT to evaluate the volume of hemorrhage focus and level of glucose uptake. The concentrations of MDA, SOD, NSE, S100B and MBP in serum or plasma were examined with ELISA. Neural differentiation of MSCs, and the expressions of Bcl-2, Bax, Arg-1 and iNOS proteins around hematoma were detected by immunofluorescence and immunohistochemistry staining respectively. Western blot was carried out to analyze the expression levels of COX4, OGDH, PDH-E1α, Bcl-2 and Bax proteins. TUNEL staining was used to estimate cell apoptosis and transmission electron microscopy (TEM) was used to observe the ultrastructure and number of mitochondria. Our data showed that EA promoted neuron-like differentiation of transplanted MSCs and the expressions of BDNF and NGF proteins in ICH rats. The score of mNSS and the gait analysis showed that the recovery of the neurological function in the MSCs + EA group was better than that in the MSCs and EA groups. EA improved the structure of brain tissue, and alleviated brain injury further after MSCs transplantation in ICH rats. When compared with the MSCs and EA groups, the level of glucose uptake and numbers of mitochondria and Arg-1 positive cells in MSCs + EA group increased significantly, but the numbers of apoptotic cells and iNOS positive cells and volume of hemorrhage focus reduced. The expressional levels of COX4, OGDH, PDH-E1α and Bcl-2 proteins increased, while the expressional level of Bax protein decreased compared with those in the MSCs and EA groups. Our results reveal that EA improve therapeutic efficacy of MSCs transplantation in ICH rats.


Introduction
Stroke is a leading cause of death and related with long-term disability for survivors. Intracerebral hemorrhage (ICH) accounts for 10-15 % of all strokes worldwide [1]. The management of ICH patients is generally limited to supportive care or removal of the hematoma; the efficacy of surgical evacuation of hematoma is variable and controversial [2]. After ICH occurs, the most significant pathological characteristic is the necrosis and apoptosis of a large number of nerve cells in lesions area, and loss of neurocytes is an important cause of disability [3][4][5]. Taking measures to reduce cell death or promote neurogenesis or replace the lost cells to improve brain structure are the basis for the recovery of motor, sensation and cognitive impairment in patients.
One major experimental strategy is to explore stem cells in improving neurological outcomes following ICH. Previous investigations indicated that mesenchymal stem cells (MSCs) could differentiate into neurons across embryonic layer, and were regarded as an vital cell source for treatment of ICH [6,7]. In ICH animal model, MSCs transplantation promoted neurogenesis, functional recovery, and inhibit inflammation, and reduced hemorrhage volume and apoptosis [8][9][10]. However, single treatment is difficult to achieve the desired effect [11].
Electroacupuncture (EA), a novel and practical therapy based on acupuncture combining with modern electrotherapy can not only quantify stimulus parameters but also improve the curative effect and save manpower, has been clinically used to treat many nervous system diseases and is often used as a common complementary treatment for stroke [12,13]. EA stimulation can improve neurological deficit scores and brain injury which connected with a series of signal pathways and various other factors [14]. However, whether EA enhances the therapeutic efficacy of MSCs transplantation in ICH rats and its potential mechanism are not clear.
In this experiment, we observed the effects of EA stimulation and MSCs transplantation on behavioral dysfunction, brain structure, ultrastructure of mitochondria, brain injury, microglia, cell apoptosis, and apoptosis-related proteins in rats subjected to ICH, and investigated whether EA improve therapeutic efficacy of MSCs transplantation in ICH rats.

Animals
All SPF Sprague-Dawley (SD) rats were obtained from the Laboratory Animal Center of Southwest Medical University. The rats were housed at room temperature (23 ± 2°C) in a 12 h/12 h light/dark cycle with free access to food and water. Experiments abided by all relevant ethical regulations and were approved by the Animal Care and Use Committee at the Southwest Medical University. Animal studies were conducted on the adult SD rats with weighing 220-250 g and aged 8-10 weeks at experimental onset. The animals were separated into sham operation (SO) group and intracerebral hemorrhage (ICH) group, and the latter were randomly subdivided into model control (MC) group, MSCs transplantation (MSCs) group, EA stimulation(EA) group and MSCs transplantation combined with EA stimulation (MSCs + EA) group.

ICH rat Model
ICH rat model was established as previously described by injecting collagenase and heparin [15]. Briefly, SD rat was anesthetized with 1 % pentobarbital sodium (40 mg/kg) via intraperitoneal administration, and then placed on a stereotaxic apparatus (Benchmark, myNeurolab.com, USA). The scalp was incised longitudinally with 1 cm long in the midline, and a 1 mm burr hole was drilled on the right cranial bone at the point that was 0.2 mm anterior and 3 mm lateral to bregma. A sterile 5-µl Hamilton syringe was then inserted into a point that was 6 mm ventral to the hole, and collagenase I (2µl, 0.125U/µl, Sigma) and heparin (1µl, 2U/µl) were slowly injected into the right striatum of rat. Then the needle was removed, and the burr hole was sealed with bone wax and the scalp was sutured. After operation, all rats were allowed free access to food and water. For the SO group, the same procedure was carried out in the rats without injection of collagenase I and heparin.

MSCs Transplantation
MSCs labelled with green fluorescent protein (GFP) (No. RASMX-01101, Cyagen Biosciences, Inc. Suzhou, People's Republic of China; Inspection report seen in Supplementary  Fig. 1) were recovered from liquid nitrogen and cultured by using alpha-minimum essential medium (α-MEM) (Hyclone) supplemented with 10 % FBS (Hyclone), 100 U/ml penicillin and 100 mg/ml streptomycin (Gibco) in an incubator (37°C, 5 %CO 2 ). When the 3-5 generation of MSCs reached 80 % confluence, the medium was substituted with preinduction solution comprised of α-MEM with 10 % FBS and 1 mmol/L β-mercaptoethanol (β-ME) for 24 hours, then with neuronal induction solution composed of α-MEM with 2 % dimethylsulfoxide (DMSO), 1 mmol/L β-ME and 1µmol/L all-trans retinoic acid (RA) (Sigma) for 6 hours. The cells were collected and the cell concentration was adjusted to 2.5 × 10 7 cells/ml. At 2 days following ICH, rats of MSCs group and MSCs + EA group were anesthetized and placed on a stereotaxic apparatus again. The suture was disassembled and a sterile microsyringe with cell suspension was inserted into the point (coordinates: 0.2 mm anterior, 6 mm ventral, 3 mm lateral to the bregma) via the burr hole described above. Then, 5 × 10 5 cells in the total 20 µL saline were slowly injected into the right striatum. After cell transplantation, the hole was filled with bone wax and the skin incision was sutured again. The animals survived for 2 weeks after cell implantation.

EA Stimulation
The EA stimulation was performed according to the methodological standards as previously described [16] and the Baihui (GV20, located at the intersection of the median line of the head and the line connecting the apexes of the two auricles) and Dazhui (GV14, located at the posterior midline of the neck and between the spinous process of the seventh cervical vertebra and that of the first thoracic vertebra) acupoints were selected for EA stimulation. After cleaning the skin with alcohol swabs, the researcher inserted a sterile acupuncture needle into GV20 to the depths of 2 mm obliquely, and another needle into GV14 to the depths of 5mm vertically to the skin. Then both of the needles were connected to EA stimulator (HM6805, People's Republic of China), and the rats were stimulated with continuous wave, intensity of 1 mA, frequency of 3 Hz and stimulation duration of 10 min. Rats of EA group and MSCs + EA group received EA stimulation once a day at 2 days following ICH, and fourteen days in succession until the rats were sacrificed. Rats of MC group were not treated with EA and MSCs implantation.

Neurological Behavioral Score
At 2 h, 1 d, 2 d following the induction of ICH and 14 d after treatment, modified Neurological Severity Scores (mNSS) was used to evaluate the neurologic deficits of each rat by an investigator who was blind to the experiment design. The score is composed of sensory, motor, reflex, and balance tests. The degree of neurological deficits is graded on a scale of 0 to 18 (normal score, 0; maximal deficit score, 18). Rats with higher score showed more serious injury. Only rats with score of more than 8 at 2 h after ICH induction were used in the current study. The reduction of mNSS score of each rat is the difference in scores between 14 days after treatment and 1 day after ICH.

Gait Analysis
The TreadScan Gait System (Clever System Inc, USA), consists of a treadmill device and a high-speed digital video camera, was used to obtain and evaluate the gait of the ICH rats at 14 days after treatment following the provided protocol and previously described [16]. Rats were tested on the treadmill at the speed of 8 cm/s for 20 s sessions, and 2000 digital camera frames were captured. Gait performance was recorded and assessed using the TreadScan software. The gait parameters measured in this study are as follows: stride length, the run speed, print area, foot and pressure.

PET/CT
Positron-emission tomography/computed tomography (PET/ CT) imaging was performed by micro PET/CT (Siemense, German). 18 F-fluorodeoxyglucose ( 18 F-FDG) was prepared at the Department of Nuclear Medicine, Affiliated Hospital of Southwest Medical University. After overnight fast, the rats were injected with approximately 0.8 mCi/kg of 18 F-FDG through tail vein, then returned to their home cage to allow the uptake of 18 F-FDG. About 30 min later, the animals were anesthetized with 1 % pentobarbital sodium (40 mg/kg, i.p.) and placed in the micro PET/CT scanner. Subsequently, PET scan and CT scan were performed for head imaging (Scan duration 15 min). The hemorrhagic foci volume and its glucose uptake were detected by analyzing the PET and CT imaging of rats.

Pathologic Examination
After the neurologic evaluation and PET/CT scans at 14 days post-treatment, the rats were euthanized with an overdose anesthetic drug and perfused transcardially with 0.9 % saline and 4 % paraformaldehyde (PFA), consecutively. Brains were taken out and submerged in PFA for 1 day, and dehydrated in 15 %, and 30 % sucrose solution until the brain sank, then embedded in Tissue-Tek® O.C.T.Compound (Sakura Finetek). The brains were cut into 7-µm coronal sections on a freezing microtome (Leica CM1850) and stored at -20°C. Sections were selected from a region spanning from − 0.4 mm to + 1.8 mm with respect to the bregma, and stained with hematoxylin and eosin (HE) for pathomorphological observation.

Assessing the Morphological Change of Mitochondria Using TEM
Dissected caudoputamen samples (the peripheral area of hemorrhagic foci, taking this point located 1.7 mm lateral to the injection position as the center) were separated into 1 mm 3 pieces, then fixed with 3 % glutaraldehyde at 4°C. The pieces were post-fixed in 1 % osmium tetroxide for 2 h, dehydrated in a gradient series of acetone, infiltrated with propylene epoxide, and finally embedded in Epon 618. Ultra-thin sections (40 nm) were cut with ultramicrotome, collected on copper grids, stained with uranium acetate and lead citrate. Four sections were selected from each group, and seven visual fields in each section were examined with TEM (JEM-1400/1011, Japan). The number of mitochondria was counted in an area of 100 µm 2 , and the rate of damaged mitochondrial were calculated with the following formula: rate = (number of damaged mitochondria /total number of mitochondria) × 100 %.

Enzyme-linked Immunosorbent Assay (ELISA)
After anesthesia, blood of rats was collected for preparation of serum and plasma. Brain damage markers and oxidative stress factor contents in serum or plasma were determined by ELISA. According to the manufacturer's instructions, the expression levels of brain damage markers (MBP, NSE, S100B) and oxidative stress factors (MDA and SOD) in the rat serum/ plasma were detected by ELISA kits from Novus Biologicals (MBP) and Nanjing Jiancheng Corp. ( NSE, S100B, MDA, SOD). The experimental process of different ELISA kits is not exactly the same. Briefly (taking the MBP kit as an example), 100 µl standard working solution and l00 µl diluted sample were added to the corresponding plate wells, and incubated for 90 min at 37°C. Remove the liquid out of each well and immediately add 100 µL of biotinylated detection Ab working solution to each well, and incubate for 1 hour at 37°C. After the plate was washed, l00 µl HRP conjugate working solution was added to each well and incubated for 30 minutes at 37°C. After the plate was washed, 90 µl substrate reagent was added to each well and for 15 min at 37°C. The reaction was terminated by adding 50 µl stop solution. The absorbance values were detected at 450 nm on a Microplate Reader (Thermo, USA). The standard curve was drawn and the corresponding concentration of the sample was calculated according to the curve equation.

Immunohistochemistry and Immunofluorescence
Immunohistochemistry was performed as follows. Frozen sections were soaked in phosphate buffer saline (PBS) for 10 min and then incubated in 0.3 % Triton-X 100 for 10 min. Next, 3 % hydrogen superoxide and 10 % normal goat serum were used to quench endogenous peroxidase activity and block non-specific binding respectively. Following removal of blocking buffer, the sections were incubated with monoclonal anti-rat Bax (1:200, Santa cruz), Bcl-2 (1:200, Santa cruz), Arg-1 (1:100, CST) and iNOS (1:100, Bioworld) at 4˚C overnight.Then corresponding biotinylated secondary antibody solution and avidin-biotin-peroxidase reagent were added to the sections, and diaminobenzidine was used for colorization. Finally, the samples were counterstained with hematoxylin. Negative controls were identically processed, except that the primary antibody was omitted.
For immunofluorescence, brain slices were washed with PBS for 10 minutes and permeabilized in 0.1 % Triton X-100 solution for 20 min at room temperature, then blocked in 10 % normal goat serum for 30 min. The samples were incubated with monoclonal anti-rat MAP2 (1:200, Santa cruz) at 4˚C overnight and Alexa Fluor 594-conjugated goat antimouse IgG (1:500, Invitrogen) at room temperature for 60 min. Finally, slices were covered with fluorescence mounting medium (Dako).

TdT-mediated dUTP Nick-end Labeling (TUNEL) Staining
Frozen sections were rinsed three times in PBS and then permeabilized with 0.1 % Triton X-100. After washing two times in PBS, the sections were treated with TUNEL reaction mixture (Roche) for 60 min at 37˚C in the dark. After washing three times in PBS, the sections were incubated with 4',6-diamidino-2-phenylindole (DAPI) for 5 minutes at room temperature, then covered with fluorescence mounting medium. Negative controls were identically processed, except that the TUNEL reaction mixture was omitted.

Western Blotting
Rats were anesthetized with 1 % pentobarbital sodium and decapitated, then the ipsilateral striatum was immediately taken and quickly frozen in liquid nitrogen until homogenization. Total proteins were extracted from brain samples using radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime) added with phenylmethanesulfonyl fluoride (PMSF) (Beyotime). The protein concentration of the lysates was quantified by the bicinchoninic acid (BCA) protein assay kit (Beyotime). Equal amounts of total protein (10 µg) from different samples were resolved by 10-12 % sodium dodecyl sulfate polyacrylamide gel by electrophoresis (SDS-PAGE), then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore) by trans-blot system (Bio-Rad) for detection. The membranes were incubated in blocking buffer (tris-buffer saline-tween, 5 % dried milk) at room temperature for 30 min, following by primary antibodies (BDNF, NGF, COX4, OGDH, PDH-E1α, Bax, Bcl-2, GAPDH and HSP70) at 4˚C overnight, then secondary horseradish peroxidase (HRP)-labeled goat anti-rabbit or anti-mouse IgG (Bio-Rad) at room temperature for 60 min. The targeted protein bands were visualized with horseradish peroxidase chromogenic Kit (Thermo Fisher), and the density of each band was quantified with an image analysis software (Quantity one). HSP70 or GAPDH was the internal control.

Statistical Analysis
The statistical analyses were performed with GraphPad Prism 6.0 (San Diego, CA, USA). All data are presented as mean ± s.e.m. Statistical significance is determined using unpaired Student's t-test between two groups and one-way analysis of variance test for multiple comparisons when comparing multiple groups. P value < 0.05 was considered to be statistically significant.

Electroacupuncture Promotes Neural Differentiation of MSCs in Rat with ICH
MSCs labeled with green fluorescent protein (GFP) (the quality control report of the MSCs derived from SD rat showed in supplementary Fig. 1) grew well after resuscitation (Fig. 1a), and took on neuron-like morphological change with elongated processes and inflated cells bodies after neuronal induction (Fig. 1b). ICH models of rats were established by injecting collagenase and heparin into caudoputamens, and treated with MSCs transplantation or/and electroacupuncture at Baihui and Dazhui after 48 hours (Fig. 1c). Immunofluorescence staining showed that the induced exogenous MSCs in rat brains could express the maker (MAP2) of neurons, and the rate of MAP2 positive cells in MSCs + EA group was significantly higher than that in MSCs group (9.34 ± 0.76 vs. 6.05 ± 0.51 %, p < 0.05; Fig. 1d,e). Furthermore, compared with MSCs group, the expression levels of BDNF and NGF proteins were obviously increased in MSCs + EA group (BDNF: 0.87 ± 0.02 vs. 0.72 ± 0.03, p < 0.05; NGF: 0.96 ± 0.01 vs. 0.74 ± 0.03, p < 0.05; Fig. 1f-h).

Combination Therapy Improves Neurological Function
The mNSS scores of all ICH groups were similar before treatment (the data hasn't shown). The reduced scores in the MSCs (4.60 ± 0.39), EA (4.95 ± 0.22) and MSCs + EA (6.55 ± 0.50) groups were significantly more than that in the MC group (3.25 ± 0.15, P < 0.05), and that of MSCs + EA group was most among all groups (P < 0.05, Fig. 2a).
Gait analysis revealed significant changes in gait functional parameters after ICH (Fig. 2b-k). When compared with MC group, MSCs transplantation, or EA stimulation or combination therapy could improve stride length (four paws), run speed, print area and foot pressure (P < 0.05). Furthermore, MSCs + EA groups had better therapeutic effect than MSCs group and EA group on stride length (four paws) and run speed (P < 0.05, Fig. 2b-k).

Combined Therapy Improves the Structure of Brain Tissue and Alleviates Brain Injury
The coronal sections of brains and hematoxylin-eosin (HE) staining showed that the hemorrhagic regions were located mainly to the striatum, and hemorrhage foci were visible in all ICH groups, but which in the different treatment groups were alleviated to varying degrees when compared with MC group (Fig. 5a, b).

Combined therapy alleviates the inflammatory reaction after cerebral hemorrhage
Microglia are the crucial immune cells in the central nervous system (CNS). Generally, they include two phenotypes of polarization: M1 type and M2 type. Immunohistochemistry showed that the number of microglial cells (including M1 and Fig. 3 Combined therapy reduce the volume of hemorrhagic foci and increased the level of glucose uptake. a PET-CT images of the rat brains in each group. First, third and fourth rows show axial, coronal and right sagittal images respectively, and the shadow represents the hemorrhagic focus. Second row shows higher magnification views of the boxed areas in first row, and pluses represent the checkpoints (3.4 mm 3 per point) of glucose uptake. b-c The quantitative analysis of the volume of hemorrhage focus (b) and glucose uptake quantitative analysis in the peripheral area of hemorrhagic foci (c) via PET-CT scan (n = 5 per group; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001) M2 types) increased in rats after ICH. The number of Arg-1 (M2 marker) positive cells was more in MSCs (69.20 ± 1.39), EA (63.40 ± 1.33) and MSCs + EA (101.00 ± 3.96) groups than in MC group (35.60 ± 1.21) (P < 0.05, Fig. 8a, c). However, the number of iNOS (M1 marker) positive cells was less in MSCs (54.00 ± 1.41), EA (56.60 ± 2.42) and MSCs + EA (45.60 ± 1.57) groups than in MC group (90.20 ± 1.56) (P < 0.05, Fig. 8b, d). Compared with MSCs and EA groups, the number of Arg-1 positive cells increased but the number of iNOS positive cells decreased in MSCs + EA group (P < 0.05, Fig. 8a-d).

Discussion
The present study investigated the hypothesis that EA stimulation could improve therapeutic effects of MSCs transplantation in rats subjected to ICH. To do this, we evaluated neurobehavioral function, differentiation of transplanted cells, ultrastructure of mitochondrion, volume of hemorrhagic foci and glucose uptake, cell apoptosis, inflammatory reaction, structural changes of brain tissue, etc. Results showed that EA stimulation further alleviates brain injury and improves neurologic deficit symptoms through a variety of pathways in ICH rats after MSCs transplantation.
Although successful neuroregeneration in the adult brain offers an attractive therapeutic prospect for brain injuries such as ICH, only finite number of mature cells developed from neuronal progenitor cells have been found to provide cellular renewing [17,18]. Thus, transplantation of exogenous cells can be an effective strategy for cell renewing in the damaged brain, and stem cell therapy has exhibited good neuroprotective and neurorestorative effect in ICH and is a promising Fig. 4 Combined therapy improve mitochondrial quality in rat brain with ICH. a Electron photomicrographs of the peripheral area of hemorrhagic foci in each group. White arrow: intact mitochondria; Black arrow: part of mitochondrial cristae fracture; White triangle: part of mitochondrial vacuolization (Scale bar, 1µm). b-c The rates of damaged mitochondria (b) and the number of mitochondria (c) in different groups (n = 4 per group; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001). d The expressions of COX4, OGDH and PDH-E1α in peripheral region of hematoma detected by Western blotting. e-g The relative expressions of COX4, OGDH and PDH-E1α proteins in different groups (n = 4 per group; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001) therapeutic method [19]. A variety of stem cells, including MSCs, neural stem cells (NSCs), embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs), have been used to treatment for experimental ICH [19]. MSCs can be derived from various adult tissues. The characteristics of presenting no major ethical concerns, having multidirectional differentiation potential and low immunogenicity, possessing self-renewal ability and immune modulation functions, and improving microenvironment make MSCs promising candidates for stem cell therapy [11,19,20].
However, strategies to ameliorate the low neural differentiation of transplanted MSCs are still under investigation. EA is one of the main healing arts in Oriental medicine. Previous studies have shown that EA can increase the synthesis and secretion of neurotrophic factors (NTFs), such as brainderived neurotrophic factor (BDNF), neurotrophin (NT) and glial-derived neurotrophic factor (GDNF), which have protective effects on neural tissues and are helpful for differentiation of stem cells in brain [21,22]. Ahn SM et al. reported that EA treatment could promote differentiation of grafted MSCs into Fig. 5 Brain tissue structure and the contents of MDA, SOD, NSE, MBP and S100B in serums/plasma. a Representative images of hematoma in ICH rats show reduced degree of brain tissue injury after treatment. b Representative brain sections stained with H&E obtained in ICH rats after treatment. The lesion area was mainly present in the striatum (Scale bars, 500 µm). c-g The content of MDA (c), SOD (d), NSE (e), MBP (f), S100B (g) in serums/plasma in different groups (n = 5 per group; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001) astrocyte-or neuron-like cells in mice with ischemic stroke [22]. Here, we also present evidence that EA stimulation increase the synthesis of BDNF and NGF, and promote neuronal differentiation of transplanted MSCs.
There is accumulating evidence that both MSCs transplantation and electroacupuncture stimulation can improve recovery after hemorrhagic brain injury via various pathways [9,[23][24][25]. Several studies have shown that combined MSCs with EA treatment may lead to a better therapeutic effect in restoring the motor function of both paws and alleviating atrophic changes of the striatum in mice with MCAO, and improving remyelination and function in demyelinated spinal cord of rats [22,26,27]. The results from our behavioral, hematological and histological studies showed that both MSCs transplantation and EA stimulation improved neuronal deficit function, increased the concentration of SOD and decreased the levels of MDA, MBP, NSE and S100B in ICH rat, what counts is that a better synergistic effect of combined MSCs and EA to restore the motor function and alleviate oxidative stress and brain damage was observed.
The process of ICH has always been accompanied by neuroinflammatory reaction characterized by microglial activation. Activation of microglia is considered to be a doubleedged sword in ICH disease [28]. Increasing evidence points out that microglia within the CNS milieu appear to be heterogeneous with opposite functional phenotypes that range from cytotoxic M1 phenotype to neuroprotective M2 phenotype [29]. The changes of microglial phenotypes rely on the disease stages and severity, and mastering the stages transformation of M1/M2 phenotypes within appropriate time windows may offer better therapeutic benefit [29]. Microglia produce M1 and M2 phenotypes to respond to brain damage in ICH pathology [30]. Our date showed that either proinflammatory M1 or anti-inflammatory M2 microglia increased in rats after ICH. Furthermore, MSCs or EA administration supressed microglia polarization to the M1 phenotype and instead  , 100µm). b The number of TUNEL-positive cells in different groups (n = 5 per group; *P < 0.05, ***P < 0.001, and ****P < 0.0001) increased a shift to the M2 phenotype. Alternatively activated M2 microglia might accelerate phagocytosis of tissue debris and erythrocytes, a major contribution to haematoma clearance [30]. In addition, EA stimulation can enhance the effects of MSCs transplantation in rat after ICH.
The pathologic processes after ICH are intricate. ICH can not only cause excitotoxicity, reactive oxygen species (ROS) generation, oxidative stress and inflammation response, but also lead to mitochondrial injury and dysfunction [31][32][33]. Neurons are hypersensitive to energy changes and need to keep up mitochondrial stability to provide enough energy, and mitochondria injury can lead to neuronal apoptosis [34][35][36][37]. It has been shown that hMSC transplantation in the treatment of cerebral hemorrhage can improve the accumulation of serial 18 F-FDG PET in the paralateral basal ganglia [38]. Zhang J et al. reported that EA reversed mitochondrial damage and improved mitochondrial functions in hippocampus in depressed rats [39]. Bao CL et al. indicated that EA could improve the energy metabolism, raise the activity of chondriosome, and decrease the accumulation of lactate acid in the perihemotoma tissue of ICH rats [40]. Our experiment showed that the treatments of both MSCs and EA increased the level of glucose uptake in focal area, meanwhile, increased the expressions of mitochondrial enzyme (COX4, OGDH and PDH-E1α) and improved ultrastructure of mitochondria. Interestingly, EA could enhance the effect of MSCs transplantation therapy on mitochondria and further reduced the hemorrhagic volume. These data suggest that the combined treatment of MSCs and EA plays an important role in maintaining the stability of mitochondrial structure and function which is helpful to promote cell survival.
Apoptosis was an important mechanism of early tissue injury in the region around the hematoma after ICH. Many factors could induce cell apoptosis after ICH, such as inflammation, free radical cascade reaction, cytokine stimulation, mitochondrial structural damage and dysfunction, the induction of thrombin and blood components [31]. In addition, various genes were also involved in regulating neuronal apoptosis, including Bcl-2 which inhibited apoptosis and Bax which promoted apoptosis [41]. MSCs transplantation could upregulate Bcl-2 expression and reduce apoptosis in ICH rat [42]. Our previous studies also showed that MSCs administration reduced cell apoptosis, and EA increased Bcl-2 expression but decreased Bax expression in rat after ICH [15,16]. In this study, we found that MSCs combinated with EA has the synergetic effects in reducing neuronal apoptosis and regulating apoptotic relative proteins in ICH rat. Fig. 7 The expressions of apoptotic-related proteins in different groups. a Images of Bax and Bcl-2 immunohistochemistry in ipsilateral striatums of rats (Scale bar, 100µm). b The protein expression levels of Bax and Bcl-2 were determined by western blotting. c-d Quantitative analyses of the relative expression levels of Bax (c) and Bcl-2 (d) proteins (n = 5 per group; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001) There are some shortcomings in the present study, and one of the limitations is that it only comprised a time point; thus, the short or long-term effects of combination therapy of MSCs and EA on ICH remain unclear. Moreover, only the effects of combination therapy on mitochondria, oxidative stress, apoptosis, neuroinflammation, and other brain injury were investigated, but the mechanisms were not characterized. Furthermore, the survival time of MSCs in the brain was not examined, although we observed that MSCs could survive in ICH rat brain.
In summary, notwithstanding limitations of this study, to the best of our knowledge we firstly reported that to some extent, EA enhance therapeutic efficacy of MSCs transplantation in rats with ICH. Of course, the major mechanisms and the key targets of combination therapy of MSCs and EA involved in offering neuroprotective effects for ICH showed in this study would be explored in our future work.

Conclusions
In the present study, we demonstrated that whether MSCs transplantation or EA stimulation, generates neuroprotective effects in hemorrhagic brain injury, and found that combination treatment of MSCs and EA produced better synergistic effects in improvement of neurological dysfunction and protecting brain tissues after ICH by impacting mitochondrial structure and function, neuroinflammation, oxidative stress, and apoptosis. Our findings suggest that the administration of MSCs combined with EA may be a novel therapeutic strategy against hemorrhagic injury.