Immunological effects of the intraparenchymal administration of allogeneic and autologous adipose-derived mesenchymal stem cells after the acute phase of middle cerebral artery occlusion in rats

Experimental groups

Ninety pathogen-free adult male Lewis rats and 10 adult male Brown Norway (BN) rats (Beijing Vital River Laboratory Animal Technology Co., animal license number SCXK, Beijing 2012-0001) with a body weight range of 260–280 g were housed at 23 ± 2 °C, under a relative humidity of 45 ± 15%, and a light/dark cycle of 12 h (lights on: 07:00 to 19:00), with free access to food and water. All Lewis rats underwent surgery to remove adipose tissue (see below) for the preparation of auto-ADMSCs and to induce middle cerebral artery occlusion (MCAO). All BN rats underwent surgery to remove adipose tissue (see below) for preparing allo-ADMSCs; MCAO was not induced. After establishing MCAO, the Lewis rats were randomly assigned to three experimental groups with 30 animals in each study group: an auto-ADMSC group, receiving autologous ADMSC transplantation; an allo-ADMSC group, receiving allogeneic ADMSC transplantation; and a control group receiving only sterile saline. All experiments were designed to minimize animal suffering in compliance with Beijing’s Navy General Hospital’s Ethical Committee guidelines for the Care and Use of Animals in Research. The researchers responsible for functional evaluation and molecular and histological studies were blinded to the treatment groups.

Isolation of ADMSCs from rats

All 90 Lewis rats and 10 BN rats were anesthetized via an intraperitoneal injection of 6% chloral hydrate at 6 mL/kg body weight. Inguinal fat pads were carefully excised under sterile conditions, cut into < 1-mm³ pieces, and incubated with 0.125% type I collagenase solution (Sigma-Aldrich, St. Louis, MO, USA) at 37 °C for 80 min under constant agitation. After three washes with sterile saline, cell suspensions were prepared by gentle homogenization through a 100-µm cell strainer. Cells were collected via centrifugation, and a red blood cell lysing reagent was subsequently added (three times the volume of cells) (Noble Ryder Technology Co. Ltd., Beijing, China, Beijing, China). After pyrolysis for 5 min at room temperature, two sterilized saline washes were performed. Then, the cells were cultured in a serum-free medium (Sigma-Aldrich) for ADMSCs. Third-passage ADMSCs were used in subsequent experiments.

Cells were incubated with the appropriate antibodies in the dark at 4 °C for 30 min to confirm the presence or absence of MSC surface markers and major histocompatibility complex (MHC) classes I and II by flow cytometry before transplantation. Antibodies used to detect MSC surface markers were PE-conjugated anti-rat CD29 (Biolegend, Santiago, CA, USA) and CD90 (Biolegend). Antibodies used to detect MHC class I and II were the anti-MHC class I RT Ia antibody [OX-18] (Biolegend) and the anti-MHC class II antibody [MRC OX-6] (Biolegend). The antibody PE-IgG1 (Biolegend) was used as an isotype control.

Coculture of ADMSCs and spleen lymphocytes in vitro

ADMSC immunogenicity was analyzed using co-culture with T lymphocytes. T lymphocytes were extracted and purified from the spleens of the Lewis rats and were cocultured for 3 days with auto- or allo-ADMSCs at a proportion of 1:10 in serum-free medium for ADMSCs. Then, the expression levels of IL-2 (eBioscience, San Diego, CA, USA) and IFN-γ (R&D Systems, Minneapolis, MN, USA) were detected using ELISA kits according to the manufacturers’ instructions.

ADMSC labeling before autologous transplantation

In order to track the transplanted ADMSCs in vivo, third-passage ADMSCs were plated at a density of 1 × 10⁶ cells per flask (T25) and then transduced with lentivirus expressing enhanced green fluorescence protein (EGFP) (Shanghai Genechem Co. Ltd., Shanghai, China) 3 days prior to transplantation. The optimal multiplication of infection (MOI) in transduction medium containing 8 mg/L polybrene was 20. The medium was replaced 24 h after transduction. The efficiency of transduction was assessed by monitoring EGFP gene expression by fluorescence microscopy. Cell viability was determined prior to transplantation by trypan blue staining. Labeled cells were used for transplantation at a concentration of 1 × 10⁵ cells/µL.

Animal model of acute ischemic stroke

After ADMSCs were cultured for 2–3 weeks, acute stroke was induced in the Lewis rats that had undergone adipose-removal surgery. A transverse neck incision was made to expose the right common carotid artery, where a small incision was made to permit the insertion of a 0.28-mm diameter nylon filament. The filament was advanced into the distal right internal carotid artery to occlude the right middle cerebral artery, inducing brain infarction at the blood supply region. Two hours after occlusion, the nylon filament was removed, followed by closure of the muscle and skin layers. Neural function was evaluated using the Zea Longa 5-grade scale [10]. Seventy-five animals with a score of 1–3 points were used in subsequent transplantation experiments.

Transplantation of ADMSCs

Eight days after the induction of ischemia, the animals were anesthetized and fixed to a stereotactic apparatus. The skull was exposed, and a burr hole was made 3 mm posterior and 1.9 mm lateral from the bregma using a small dental drill. A microinjector was inserted 2.9 mm into the brain parenchyma from the surface of the dura mater, and 10 µL of cell suspension (approximately 1 × 10⁶ ADMSCs) or an equal volume of physiological saline was injected into the brain parenchyma over a period of 20 min.

Apoptosis of transplanted ADMSCs

Apoptosis within populations of transplanted cells was detected using the TdT-mediated dUTP nick-end labeling (TUNEL) assay. One day after transplantation, the apoptosis of transplanted ADMSCs was detected in the peri-infarct zone using a TUNEL assay kit according to the manufacturer’s instructions (Beyotime, Shanghai, China). Cells were counted in one brain tissue section of each animal (n = 5 per group). The number of double-staining-positive (red and green fluorescence) cells was counted in a minimum of 10 microscopic fields based on their nuclear morphology, and dark color was quantified using a 40× objective and Image-Pro image analysis software.

Viability, migration, differentiation, and immunological effects

The viability, migration, and differentiation of transplanted ADMSCs, and the in vivo local accumulation of CD4+ T lymphocytes, CD8+ T lymphocytes, and microglial cells induced by rat ADMSCs, were detected using immunohistochemistry. The determination of concentrations of IL-2 and IFN-γ in the brain induced by transplanted ADMSCs was determined using an ELISA assay.

Five rats from each group were anesthetized at 1, 7, and 28 days after transplantation. After perfusion with normal saline and 4% paraformaldehyde and dehydration with 15 and 30% glucose, brain tissues were harvested to produce cryosections. The localization of transplanted cells was observed and photographed under an inverted fluorescence microscope (Olympus IX51, Olympus Corporation, Tokyo, Japan). After EGFP markers were assessed quantitatively with the image analysis software Image-Pro Plus 6.0 (Media Cybernetics, Rockville, MD, USA), the survival rate of the transplanted cells in each group was calculated.

To detect the differentiation potential of transplanted ADMSCs at 8 days post infarction, twenty 10-µm-thick cryosections of the infarcted tissue from each group (n = 5 per group) were examined using immunofluorescent antibodies to mark astrocytes with glial fibrillary acid protein (GFAP; monoclonal antibody diluted 1:1000, incubation at 4 °C overnight, Abcam) to investigate lymphocyte infiltration and microglial proliferation and activation. At 28 days post infarction, additional cryosections were similarly examined using neuronal nuclear antigens (NeuN; monoclonal antibody diluted 1:300, at 4 °C overnight, Abcam) to analyze the differentiation of transplanted ADMSCs.

The distribution of CD4+ and CD8+ T lymphocytes and CD68+ and Iba-1+ microglias around transplanted ADMSCs was also investigated using immunohistochemistry. To this end, these cells were labeled using anti-CD4 antibody (1:200 dilution, incubation at 4 °C overnight, Abcam), anti-CD8 antibody (1:200 dilution, incubation at 4 °C overnight, AbD Serotec, Kidlington, UK), anti-CD68 antibody (1:100 dilution, incubation at 4 °C overnight, Abcam), and the anti-Iba1 antibody (1:100 dilution, incubation at 4 °C overnight, Abcam), respectively. Red fluorescence-labeled (Alexa Fluor 594) secondary antibodies (Jackson ImmunoResearch, Hamburg, Germany) were diluted to 1:100 prior to use. Double-staining-positive (red and green fluorescence) cells were counted as previously described.

To determine the concentrations of IL-2 and IFN-γ in the brain as induced by transplanted ADMSCs, ELISA assays were performed 7 days after transplantation. Five rats from each group were euthanized. Brain tissues from the hippocampus of the graft side were separated and ground on ice and then centrifuged at 5000×g for 5 min at 4 °C. Total protein in the supernatant was quantified using a BCA assay. Concentrations of IL-2 and IFN-γ were then determined using ELISA kits according to the manufacturer’s instructions. The protein expression levels of these factors in the brain tissue were expressed as nanogram per milligram total protein.

Assessment of motor function

Motor function was assessed at 14 and 28 days post transplantation using the modified Neurological Severity Score (mNSS) and rotarod test by two investigators blinded to the experimental groups. The mNSS test is a composite of motor, sensory, balance, and reflex tests. Neurological function was graded on a scale of 0–18, where 0 is normal and 18 represents maximum deficit. In the rotarod test, the 10-cm diameter rotarod cylinder was accelerated from 5 to 40 rpm over 5 min. The time remaining on the rotarod was measured at 0, 14, and 28 days post transplantation, and the data were presented as the mean duration from three trials.

Assessment of relative cerebral infarction volume

In order to assess the relative cerebral infarction volume at 28 days post transplantation, 2,3,5-triphenyltetrazolium chloride (TTC; Sigma-Aldrich) staining was performed. Five animals from each group were anesthetized using 6% chloral hydrate. After the left ventricle was perfused with 200 mL physiological saline, the brain was removed, sliced into 2-mm sections, stained with 2% TTC for 30 min at 37 °C in the dark, and then fixed with 4% paraformaldehyde. Under these conditions, live tissues retain a red color, while infarcted areas appear white. The ischemic area of the brain was quantified using ImageJ software (NIH, Bethesda, MA, USA), and ischemic volume was expressed as the percentage relative to the ipsilateral hemisphere.

Statistical analysis

Statistical analysis was performed using SPSS 17 for Windows (IBM, Chicago, IL, USA). Quantitative data are shown as mean values ± SDs. The Kruskal–Wallis test followed by the Mann–Whitney U test were used to compare functional evaluation scores, survival rates of transplanted ADMSCs, numbers of apoptotic transplanted ADMSCs, cerebral infarction volumes, and the expression of IFN-γ and IL-2 in each group. Values of p < 0.05 were considered significant at 95% confidence.

Characterization of ADMSCs from Lewis and BN rats

The third-passage cultured ADMSCs waiting for transplantation showed typical, fibroblast-like cell morphology (Fig. 1). Cultured cells from Lewis (96.78%) and BN (92.28%) rats were positive for the CD90-PE surface marker. Cultured cells from Lewis (96.76%) and BN (94.45%) rats were also positive for the CD29-PE surface marker. Non-MSC surface markers (CD34 and CD45) were expressed in less than 6% of cultured cells from Lewis and BN rats (Fig. 2). The MHC class I marker was expressed in 47.86% of the cultured cells from Lewis rats and in 28.82% of the cultured cells from BN rats, while the MHC class II marker was expressed in 5.36% of the cultured cells from Lewis rats and 2.76% of the cultured cells from BN rats (Fig. 3).

Immunogenicity of ADMSCs in vitro

To characterize the in vitro immunogenicity of ADMSCs, we evaluated the levels of IL-2 and IFN-γ in mixed T-lymphocyte cultures prepared with either allo- or auto-ADMSCs. There were no significant differences in the production of IFN-γ and IL-2 between cultures with auto-ADMSCs and cultures without ADMSCs. However, allo-ADMSCs induced a 6.56-fold increase in the production of IFN-γ and a 3.16-fold increase in the production of IL-2 when compared with auto-ADMSCs (p < 0.01, Fig. 4).

Apoptosis and the survival rate of transplanted ADMSCs in ischemic brain tissues

TUNEL staining was performed 1 day after transplantation to detect the apoptosis of ADMSCs. ADMSCs were tracked by transfection with EGFP (green fluorescence), and the nuclei of the apoptotic cells were labeled by red fluorescence (TUNEL+) such that EGFP- and TUNEL-positive cells represented apoptotic-transplanted ADMSCs (Fig. 5a, b). TUNEL staining showed that the number of apoptotic transplanted allo-ADMSCs per cubic millimeter of brain tissue near the infarct area was 3.19 times higher than that of apoptotic transplanted auto-ADMSCs (1323.02 ± 278.71 vs. 415.06 ± 68.79, p < 0.01; Fig. 5c). One day after cell transplantation, the survival rate of auto-ADMSCs was significantly higher than that of allo-ADMSCs (9.45 ± 0.34% vs. 4.19 ± 0.11%, p < 0.01; Fig. 5d). The survival rate of auto-ADMSCs was also significantly higher than that of allo-ADMSCs 7 days after transplantation (5.44 ± 0.25% vs. 1.33 ± 0.16%, p < 0.01; Fig. 5d). Twenty-eight days after cell transplantation, the survival rate of auto-ADMSCs was 3.35 ± 0.16% (Fig. 5d); however allo-ADMSCs were not observed in ischemic brain tissues at this time point.

Migration and expression of astrocyte or neuron markers of ADMSCs in ischemic brain tissues

ADMSCs labeled with EGFP were transplanted into the CA1 region of the hippocampus. One day after cell transplantation, the migration of auto-ADMSCs was observed in the penumbra cortex and hippocampus, while the migration of allo-ADMSCs was observed only in the penumbra cortex (Fig. 6a). The migration of allo-ADMSCs was observed in the hippocampus 7 days after transplantation. Overall, the migratory scope of auto-ADMSCs was wider than that of allo-ADMSCs (Fig. 6b). Twenty-eight days after cell transplantation, auto-ADMSCs were distributed mainly in the hippocampus, and to a lesser extent in the cortical injury area (Fig. 6c). However, allo-ADMSCs were not observed in either the cortical injury area or the hippocampus (Fig. 6c).

Twenty-eight days after transplantation, cells positive for EGFP (green fluorescence) and GFAP (red fluorescence) were found in the brain sections of the auto-ADMSC treatment group, indicating that the EGFP-labeled auto-ADMSCs expressed astrocyte markers. The percentage of EGFP- and GFAP-positive cells among all surviving auto-ADMSCs was 8.73 ± 0.92% (Fig. 6d–f). We also observed cells positive for EGFP and NeuN (red fluorescence) in the brain sections of the auto-ADMSC treatment group. The percentage of these immature neuron-like cells with spindle or round shapes that exhibited the expression of the apophyse marker was 2.14 ± 0.69% among all surviving auto-ADMSCs (Fig. 6g–i). Because surviving allo-ADMSCs were not observed on day 28 post transplantation, we did not examine the expression of astrocyte or neuron markers for the administered allo-ADMSCs.

Immune response evoked by transplanted auto-and allo-ADMSCs in ischemic brain tissues

Immunohistofluorescence analysis demonstrated that allo-ADMSCs evoked a local immunological response, characterized by an accumulation of CD4+ and CD8+ T cells, CD68+ activated microglial cells, and IBA-1+ resting microglial cells in the brain tissue near the transplanted cells at 7 days post transplantation (Fig. 7a–d), as well as a significant upregulation of the pro-inflammatory factor IFN-γ (3.12-fold, Fig. 8a) and the inflammatory cytokine IL-2 (2.53-fold, Fig. 8b) in comparison to controls (p < 0.01). However, these local immune responses were not observed around transplanted auto-ADMSCs.

Auto- and allo-ADMSCs improve functional recovery after transplantation

Both auto- and allo-ADMSCs improved functional recovery after MCAO in animals at 14 and 28 days post transplantation (Fig. 9). There were no significant differences in mNSS scores between any groups prior to cell transplantation. Both transplantation groups showed significantly improved functional recovery compared with the controls, where allo-ADMSCs yielded scores of 6.6 ± 1.1 at 14 days (p < 0.05) and 5.4 ± 0.9 at 28 days (p < 0.01), and auto-ADMSCs yielded scores of 5.6 ± 0.9 at 14 days (p < 0.01) and 3.8 ± 0.4 at 28 days (p < 0.01). By comparison, the control group yielded scores of 8.2 ± 0.8 at 14 days and 7.6 ± 0.5 at 28 days. Moreover, the auto-ADMSC transplantation group showed significantly better functional recovery compared with the allo-ADMSC transplantation group 28 days after transplantation (p < 0.01; Fig. 9a). In the rotarod test, there were no significant differences in the mean holding time between any groups prior to cell transplantation. Allo-ADMSC (34.6 ± 2.3 s, p < 0.01; 46.6 ± 3.8 s, p < 0.01) and auto-ADMSC (44.2 ± 3.7 s, p < 0.01; 68 ± 4.3 s, p < 0.01) transplantation groups showed significantly improved functional recovery in comparison to the controls (21.6 ± 4.0 s; 26.6 ± 2.8 s) at 14 days and at 28 days, respectively. Moreover, auto-ADMSC transplantation yielded a significantly longer holding time in comparison to auto-ADMSC transplantation at 14 days (p < 0.01) and 28 days (p < 0.01) post transplantation (Fig. 9b).

Auto- and allo-ADMSC transplantation reduces infarct size

Staining with TTC 28 days after transplantation (Fig. 10a) showed that both allo-ADMSC (28.22 ± 2.38%, p < 0.01) and auto-ADMSC transplantation (24.35 ± 2.14%, p < 0.01) significantly reduced relative lesion volume with respect to the MCAO model group (34.69 ± 1.84%; Fig. 10b). Moreover, auto-ADMSC transplantation reduced lesion volume to a significantly greater extent than allo-ADMSC transplantation (p < 0.05; Fig. 10b).