We randomized the rats into four groups immediately after middle cerebral artery occlusion (MCAO) (Fig. 1). Rats were anesthetized via intraperitoneal injection with 10% chloral hydrate (0.35 ml/100 g) 24 h after stroke. The control group was treated with 0.3 ml saline as vehicle during the subacute phase. Animals in the BDNF group were treated with 0.3 ml BDNF (50 µg/ml) injection via the tail vein. Rats in the DPSCs group were injected with 0.3 ml DPSCs suspension (1×10⁷/ml) via the tail vein very slowly. Rats in the DPSCs+BDNF group were injected with 0.3 ml DPSCs (1×10⁷/ml) and BDNF (50 µg/ml) via the tail vein.

DPSCs were obtained from the incisors of 4-week-old male Sprague-Dawley (SD) rats under sterile conditions as previously described (17). In brief, dental pulp was treated with 3 mg/ml collagenase type I at 37°C for 1 h with gentle shaking, followed by centrifugation (1,500 rpm for 5 min). The supernatant and enzymes were discarded, and the residue was cultured in α-modified Eagle's medium (α-MEM) containing penicillin (1%, 100 U/ml), streptomycin (1%, 100 µg/ml) and fetal bovine serum (FBS 10%; Gibco, Grand Island, NY, USA) at 37°C and 5% CO₂. The medium was replaced every 3 days. Cells at 80% confluency were passaged. DPSCs were identified by flow cytometry based on the surface markers on stem cells. In brief, ~1×10⁶ cells were incubated with 2% FBS in phosphate-buffered saline (PBS) at 4°C for 30 min and 1 µl of rat monoclonal antibody specific for Sca-1, CD34, CD44 (18) and CD45 (19). Cells in the control experiment were cultured in buffer solution without primary antibodies. DPSCs were pre-labeled with green fluorescent dye PKH67 up to 90%. The final density of the labeled cell suspension was ~1×10⁷ cells/ml.

To test the pluripotency of the isolated DPSCs, passage 3 cells were seeded in 6-well culture plates up to 20,000 cells/cm². The proliferation medium was replaced by the inductive medium. For mineralized nodules, the differentiation-inducing medium was supplemented with ascorbic acid, dexamethasone, and an excess of phosphate. Osteogenesis was promoted by supplementing 50 g/ml ascorbic acid 2-phosphate, 10 nm dexamethasone and 10 µM β-glycerolphosphate. Adipogenesis was induced with 100 nm dexamethasone and 50 µg/ml indometacin. The culture was maintained for 3 weeks, during which the medium was changed every 2–3 days. Finally, the cells were stained with Alizarin Red and Oil Red O, and visualized under light microscopy.

The animal model of focal cerebral ischemia was generated by transient occlusion of the right middle cerebral artery (MCA) (20). Adult male SD rats weighing 250±10 g each, were purchased from the Animal Center of the Second Affiliated Hospital of Harbin Medical University. All the procedures were approved by the Ethics Committee of the Harbin Medical University, and were compliant with the NIH guidelines for use and care of laboratory animals. In brief, rats were anesthetized with 10% chloral hydrate (0.35 ml/100 g). A midline cervical incision was used to dissect the right common carotid artery (CCA), external carotid artery (ECA) and internal carotid artery (ICA). A 4-0 monofilament nylon thread with a silicone-coated tip (diameter 0.37 mm, length 3–4 mm) was extended from the ECA into the ICA to mask the origin of the right MCA. The nylon thread was secured and the incision was closed. The rat body temperature was maintained at 37±1°C using an incandescent lamp. Two hours later, the thread was withdrawn to perform reperfusion. Sham-operated animals were subjected to similar procedures without occlusion of the right MCA. MCAO was confirmed by suspending the rats by the tail for contralateral forelimb flexion. Further, circling behavior on the floor was detected. The weight of rats after ischemia was recorded at 0, 1, 7, 14 and 28 days post-MCAO.

TTC staining was used to verify ischemic injury at 24 h post-MCAO. The brain wa rapidly removed on ice under anesthesia. Fresh brains were frozen for 25 min at −20°C, sliced into five consecutive coronal sections, incubated in 2% TTC (Sigma-Aldrich, St. Louis, MO, USA) at 37°C for 20 min in the dark. Healthy zones appeared red and infarcted tissues were white in color. Brain sections were photographed using a digital camera. The infarct volume was determined as a percentage of dead cerebral tissue.

The brain tissues from rats were embedded in paraffin. Serial coronal sections were produced and mounted on slides. The specimens were stained with hematoxylin and eosin (H&E) according to standardized protocols, and visualized by light microscopy.

Two small pieces of adhesive-backed paper dots, measuring 113.1 mm² in size, were used as bilateral tactile stimuli to the distal-radial region of the wrist in each forelimb. Rats were pre-trained for a week before MCA occlusion. They were monitored on days 1, 7, 14, 21 and 28 after stroke. The mean duration of the three trials for the removal of adhesive papers was recorded (21). The mNSS encompass a series of motor, sensory, balance and reflex tests graded on a scale of 0–18 (normal 0; maximal deficit score 18). Scores reflect the degree of inability to perform a specific test: higher scores suggest severe neurological deficit (22).

The rats were anesthetized with 10% chloral hydrate 0.35 ml/100 g and their chests were split to expose the hearts. The animals were rapidly perfused with 400 ml saline, followed by 300 ml 4% paraformaldehyde (pH 7.4) through the left ventricle. The brain tissues of the rats were post-fixed in 4% paraformaldehyde (pH 7.4) at 4°C for 24 h. Brains were dehydrated in a graded series of sucrose at 20 and 30% daily. Frozen sections of the brain were blocked with normal goat serum for 20 min at room temperature, and incubated with anti-nestin (1:100), anti-doublecortin (DCX) (1:100), and anti-neuronal specific filament (NF-H) (1:200) overnight at 4°C. They were washed with PBS three times (5 min each time) followed by 1 h reaction with rhodamine-conjugated goat anti-rabbit and anti-mouse IgGs. Brain sections were counter-stained with DAPI for 10 min at 37°C. The survival and spatial distribution of PKH67-labeled DPSCs were observed using a laser scanning confocal microscope. Five randomly-selected fields were used to count the average number of positive cells in the regions. The antibodies are listed in Table I.

Paraffin sections were washed with xylene and ascending alcohol, to air-dry and hydration, and incubated in 1% H₂O₂ for 10 min. Antigen retrieval was performed by heating the sections at 95°C for 10 min. Paraffin sections were incubated with anti-nestin (1:200), anti-DCX (1:200), anti-NF-H (1:200) at 4°C overnight and biotinylated secondary antibodies for 20 min at room temperature. The sections were incubated with avidin-biotin complex (ABC) (1:400) for another 20 min, followed by washing with PBS three times, 5 min each time. The sections were counterstained with hematoxylin, followed by dehydration, and analyzed under light microscope. The antibodies are listed in Table I.

Animals were euthanized 4 weeks after MCAO using an overdose of anesthesia. Infarcted cerebral tissues were lysed with RIPA buffer containing protease/phosphatase inhibitors and phenylmethanesulfonyl fluoride (PMSF) on ice. After centrifugation at 15,000 rpm for 15 min, protein concentration was measured using a bicinchoninic acid (BCA) protein assay kit. Protein samples (30 µg each) were electrophoresed and transferred to polyvinylidene difluoride (PVDF) membrane, which was blocked with 5% nonfat dry milk in PBS containing 0.2% Tween-20 1.5 h at room temperature. After gentle shaking, and washing four times (10 min each) followed by incubation with primary antibodies: anti-nestin (1:1,000), anti-DCX (1:500), anti-NF-H (1:1,000) and anti-β-actin overnight at 4°C, the membrane was incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5,000) at room temperature for 1.5 h. An ECL advance western blotting detection kit was used to detect chemiluminescence in the dark. The antibodies are listed in Table I.

Total RNA was extracted from the right cerebral cortex with TRIzol (Invitrogen, Carlsbad, CA, USA). The cDNA synthesis was performed using the reverse transcription system with oligo(dT)₂₀ according to the manufacturer's instructions. A control reaction without the reverse transcriptase was also carried out. The amplifications were conducted on a real-time PCR system (Bio-Rad, Hercules, CA, USA) using Accupower^® 2X SYBR-Green I qPCR Master Mix (Bioneer, Daejeon, Korea) under the following conditions: pre-denaturation at 95°C for 15 min, followed by 40 cycles of 95°C for 10 sec, 53°C for 30 sec, and cooling at 4°C for 5 min. The melting temperatures (Tm) of the amplification products immediately after the last reaction cycle were used to determine the specificity. The 2^−ΔΔCT method was used to compare the gene expression of neural markers in each group. All the PCRs were performed in triplicate and validated by the presence of a single peak in the melting curve. The primers are shown in Table II.

All data are expressed as mean ± standard error of the mean (SEM). Means were compared using one-way ANOVA with Bonferroni's multiple comparison tests. P<0.05 was considered statistically significant. GraphPad Prism version 5 was used for all the statistical analyses.

The transient MCAO is an established animal model of stroke. In the present study, the infarct zone ratio for each brain was calculated as follows: (left hemisphere area-right uninfarcted area)/(left hemisphere area^*2), TTC staining showed that the ischemic areas accounted for an average 32.1±1.8% of the total volume at 24 h (Fig. 2). MCAO-induced injury mainly involves frontal, parietal, temporal, occipital and striatum regions of the right cerebral hemisphere. No visible lesions occurred in the sham-operated animals. H&E staining revealed normal brain tissues in the sham-operated group, with uniform neuronal distribution. Neurons were round or oval, with pink cytoplasm, abundant and with clear nucleus. No inflammatory cells were detected. In the MCAO group, infarcted brain tissues were found, with a limited number of disordered neurons containing pale cytoplasm, eosinophils and abundance of vacuoles. In addition, a large infiltration of necrotic neurons and inflammatory cells was observed. Widespread ischemic damage in the brain resulted in significant neurological deficit. The 18-point mNSS and the results of the adhesive removal tests increased significantly.

Cultured DPSCs exhibited plastic adherence, colony formation, fibroblast-like morphology, self-renewing capacity, expression of surface markers and multilineage differentiation (Fig. 3). Fluorescence-activated cell sorting (FACS) analyses were used to evaluate the expression of various surface markers on the isolated cells. In conclusion, 91.32% of the passage-3 cells expressed stem cell marker Sca-1; and 73.20% expressed CD44. In contrast, few cells expressed hematopoietic stem cell markers, 1.35% expressed CD34; 1.23% expressed CD45. In addition, randomly distributed mineralized nodules were detected in the culture dish. Osteogenesis and adipogenesis were characterized by the presence of calcium deposits and fat droplets. In conclusion, the cells derived from SD rat dental pulp appeared to be largely DPSCs.

In the present study, we observed PKH67 fluorescent signal 3 weeks after DPSCs transplantation (Fig. 4). PKH67-labeled cells were abundant in the DPSC group around the periphery of the infarcted area, but not in the contralateral nonischemic brain (data not shown) under laser scanning confocal microscope. Several PKH67-labeled cells were found in the DPSCs+BDNF treatment group than in the DPSC group, as BDNF prolonged the viability of transplanted DPSCs. No PKH67-labeled cells were found in the brains of control rats. These results suggest that DPSCs penetrated through the blood-brain barrier (BBB) (23), proliferated, and migrated toward the ischemic areas of stroke. Homing ability is the characteristic of stem cells. Grafted cells can migrate and localize in injured areas of the brain tissues. PKH67 represents an ideal cell-labeling agent to track transplanted cells in vivo.

PKH67-labeled cells emitted green fluorescence at 488 nm. Rhodamine-conjugated antibody emitted red fluorescence at 594 nm, and DAPI displayed blue fluorescence at 405 nm. We found that most of the PKH67-labeled DPSCs accumulated in the peripheral ischemic regions of MCAO models (Fig. 4). They displayed neuronal morphology and were identified with a panel of neural-specific markers including nestin (24), DCX (25), NF-H and DAPI. The arrow in Fig. 5B shows that DCX was localized mainly in the cytoplasma. We systematically analyzed and compared the expression of markers in each group. Immunofluorescence revealed increased number of positive cells in DPSCs+BDNF group compared with DPSCs group. The percentage of positive cells was significantly higher in the different treatment groups than in the saline control group (P<0.05). The number of positive cells in the DPSCs+BDNF group was significantly higher than in the BDNF and the DPSCs groups (P<0.05) (Fig. 5). Using western blot analysis and RT-PCR analysis, we further confirmed an increase in neuron-specific markers in the DPSCs+BDNF group compared with the other groups. Protein and gene markers in DPSCs or BDNF group were also higher than in the saline control group (P<0.05) (Figs. 6 and 7). Data suggest that BDNF promoted DPSC differentiation into neuron-like cells and triggered neurogenesis.

The body weight of MCAO models decreased significantly 24 h after surgery and recovered slowly eventually. The weight was higher in the experimental groups than in the saline control group. The recovery in the DPSCs+BDNF group was significantly higher than in the BDNF and DPSCs groups (P<0.05). The groups differed significantly from each other suggesting that BDNF and DPSCs prevented neuronal ischemia (Fig. 8).

The stroke models were functionally assessed using the adhesive-removal test and the mNSS on days 1, 7, 14, 21 and 28 after MCAO. All the animals showed functional impairment on day 1 postoperatively, followed by gradual recovery. Compared with the saline group, animals treated with BDNF or DPSCs exhibited improved neurological performance in the adhesive removal task and in mNSS at 1 week post-treatment (P<0.05). The combination treatment group showed greater functional recovery compared with other groups (P<0.05). The greatest improvement was observed in the DPSCs+BDNF-treated rats on day 28 post-MCAO (P<0.01). These data suggest that functional deficits following ischemia due to MCAO were successfully treated by intravenous transplantation of DPSCs or BDNF injection (Fig. 7). Further, DPSCs and BDNF showed synergistic neuroprotection against stroke.