Tongfu Xingshen capsules (TXC; 0.4 g/granule) were provided by the Pharmacy Department of Guangdong Hospital of Traditional Chinese Medicine. Evans Blue (EB), BrdU, formamide and sodium pentobarbital were purchased from Guangzhou Weijia Technology Co. Ltd. All chemicals and solvents were of an analytical grade.

Male SPF grade SD rats (age, 6 weeks) weighing between 200 and 250 g were purchased from the Institute of Experimental Animals, Chinese Academy of Medical Sciences (Beijing, China), and acclimatized in standard cages under controlled temperature (23±3°C) and a regular 12 h:12 h light-dark cycle for 7 days before the procedure. All experimental animal protocols were approved and performed following the Institutional Animal Care Guidelines (approval no. 2016040; Experimental Animal Ethics Committee, Guangdong Provincial Hospital of Chinese Medicine, Guangzhou, China).

The animals were divided into the sham-operated, untreated model, low dose TXC-treated, and high dose TXC-treated groups, and ICH was induced through autologous intracerebral blood infusion. In brief, the animals were all anesthetized using pentobarbital (50 mg/kg i.p.), and 50 µl blood was drawn from the tail artery. The rats were positioned in a stereotaxic frame (Stoelting Instruments), and a cranial burr hole (1 mm) was drilled near the right coronal suture 3.0 mm lateral to the midline. A microsyringe was stereotaxically inserted into the right basal ganglia (coordinates: 0.2 mm anterior, 6.0 mm ventral, and 3.0 mm lateral to the bregma), and autologous whole blood was injected at the rate of 10 µl/min (26–28). In the sham-operated control, the blood was also drawn from the tail artery before the needle was inserted without injecting any blood. The rats were intragastrically administered 1 ml/100 g normal saline (sham-operated and untreated controls), 1 g TXC/kg/day (high dose; 0.1 g/ml with normal saline), or 0.5 g TXC/kg/day (low dose; 0.05 g/ml with normal saline), as appropriate. For testing neurological function score, brain water content and vascular permeability test, 96 SD rats were divided as aforementioned (n=24 per group). Each group was further subdivided into the 1, 7, 14 and 28-day groups (n=6 per group). For immunofluorescence and reverse transcription-quantitative PCR (RT-qPCR), 72 animals were similarly divided into 7, 14 and 28-day subgroups in each treatment group.

BrdU (50 mg/kg) was intraperitoneally injected at a concentration of 5 mg/ml in 0.9% NaCl, once a day. Except for the 1-day group, the other groups received daily injections for 5 days after the model was established.

From each rat, 6-µm thick serial coronal sections of the brain were cut to include the entire SVZ (80–90 slides per animal, and 3 sections per slide). Every tenth slide was stained using hematoxylin and eosin at room temperature following standard protocols. Briefly, brain sections were stained with hematoxylin for 5 min and differentiated with 1% hydrochloric acid alcohol for 5 sec at room temperature. Following immersion in ammonia-water for 10 sec, sections were dyed with eosin for 3 min, followed by mounting with neutral balsam. After sealing, the SVZ structures were observed at ×40 magnification under an Olympus BX61 light microscope, and adjacent sections with similar SVZ structures were used for double immunofluorescence staining.

Neurological tests were performed 24 h post-stroke and were scored according to the modified Bederson scale as follows: 0, no deficits; 1, flexed forepaw; 2, an inability to resist the lateral push; 3, circling; 4, agitated circling; and 5, unresponsive to stimulation/stupor (29). The neurobehavioral score of 1–5 was considered as a successful model.

Evan's Blue dye extravasation test was performed, as previously described (30). The harvested brains were quickly separated into the left anterior, right anterior, left posterior and right posterior segments, weighed and digested using 3 ml formamide in a 60°C water bath for 24 h. The homogenates were centrifuged at 10,000 × g at room temperature for 20 min, and the absorbance of the supernatant at 630 nm was measured using a spectrophotometer. The Evans blue content was calculated as mg/g against a standard curve.

Rats were decapitated under deep anesthesia (pentobarbital; 50 mg/kg i.p.), which was validated based on the fact that there was no pain reaction, no stimulation reaction, general muscle relaxation, and smooth breathing, and the brains were immediately removed. The posterior brains were weighed using an analytical microbalance to obtain the wet weight (WW). The samples were then dried at 75°C for 5 days, and the dry weight (DW) was determined. Brain water content was calculated as (WW-DW)/WW ×100%.

NeuN/BrdU and GFAP/BrdU double staining were performed on the 14-day and 28-day samples, while the 7-day samples were stained using Nestin and BrdU. In brief, the sections were incubated overnight with rat anti-BrdU (cat. no. ab6326) and mouse anti-Nestin (cat. no. ab6142), rabbit anti-NeuN (cat. no. ab177487) or rabbit anti-GFAP (all 1:200; cat. no. ab33922; all Abcam), as appropriate, at 4°C. After probing with the relevant secondary antibody (all 1:500; Alexa Fluor^®488 donkey anti-rat, cat. no. A21208; anti-mouse, cat. no. A32744 or anti-rabbit IgG, cat. no. A32754; Thermo Fisher Scientific, Inc.) at room temperature for 1 h in the dark, the slides were washed and observed at ×400 magnification under a Nikon Ti2-E light microscope. ImageJ software (v2.1.4.7; National Institutes of Health) was used to measure the number of positive cells. The number of cells in the SVZ region of 3 non-overlapping fields were counted in each section, and the mean was calculated.

Total RNA was extracted from the brain tissues using the TRIzol reagent, following the manufacturer's instructions (Takara Biotechnology Co., Ltd.). The following primers were used to perform the RT-qPCR: Rat-BDNF forward primer, GATTAGGTGGCTTCATAGGAGAC; rat-BDNF reverse primer, CGAACAGAAACAGAGGAGAGATT; rat-TrκB forward primer, GATGTTCCAGCCACTGTGAACC; rat-TrκB reverse primer, TCCACCACCCTGTTGCTGTA. Rat-GADPH (forward, 5′-GATGTTCCAGCCACTGTGAACC-3′ and reverse, 5′-TCCACCACCCTGTTGCTGTA-3′) was used as the internal control. The purity of RNA samples was assessed via NanoDrop, and samples with ratios between 1.80 and 2.01 were used for cDNA synthesis using the GoScript™ Reverse Transcription system, according to the manufacturer's protocol (Promega Corporation). The qPCR reaction was performed in 20 µl with 5 µl template DNA and 500 nM primers. The PCR condition was as follows: Initial denaturation, 95°C for 10 min, 40 cycles of amplification (95°C for 10 sec and 60°C for 30 sec) and cooling period, 50°C for 5 sec. Each sample was analyzed in triplicate. iTaq™ Universal SYBR^® Green Supermix (Bio-Rad Laboratories, Inc.) was used according to the manufacturer's instruction. The PCR results were analyzed in a Real-Time PCR system (Bio-Rad Laboratories, Inc.; cat. no. CFX96). The specificity of the primers was tested by analyzing the dissolution curve. The mRNA of each target gene was homogenized, and the phase of mRNA of each target gene was calculated by Cq value for the relative expression level (2^−ΔΔCq method: ΔΔ Cq=(Cq _{target gene} -Cq _GAPDH) _{treatment group}- (Cq _{target gene} -Cq _GAPDH) _{control group}) (31).

The data were analyzed using SPSS 21 software (IBM Corp.) and are expressed as the mean ± SD. Differences were analyzed using one-way analysis of variance ANOVA with post hoc Bonferroni test. P<0.05 was considered to indicate a statistically significant difference.

ICH resulted in left hemiplegia, severe hemiplegia in the left forelimb and clockwise rear-ended rotation within 7 days of induction, and the symptoms were observed over 14 and 28 days. Compared with the untreated model group, both low and high dosages of TXC improved gait symptoms and decreased hemiplegia (Fig. 1A). Furthermore, TCX significantly decreased the extravasation of EB from brain tissues after 7, 14 and 28 days of stroke, compared with the untreated model group (Fig. 1B). Finally, the brain water content in the ipsilateral hemicerebrum substantially increased in the model group, compared with the sham-operated animals, and was significantly decreased by low and high dose TXC treatment at all time points (Fig. 1C).

ICH injury significantly increased the number of proliferating BrdU-positive cells in the SVZ region, increased by low or high dose TXC-treated (Fig. 2). As shown in Fig. 3, the total number of proliferating NSCs (Nestin- and BrdU-positive cells) was markedly increased after TXC treatment within 7 days post-stroke. However, the proportion of actively cycling neuron (NeuN- and BrdU-positive) cells was similar in both the untreated and treated groups, both at 14 and 28 days post-stroke (Fig. 4). Furthermore, TXC treatment also increased the number of proliferating glial cells (GFAP- and BrdU-positive) in the SVZ region after 14 and 28 days post-stroke (Fig. 5).

The BDNF mRNA levels were significantly higher in the brain tissues of the TXC-treated animals compared with the untreated controls at 7 and 14 days post-stroke. The upregulation on day 14 was particularly notable in the high dose TXC group. However, there was no significant difference in BDNF expression between the untreated and treated groups after 28 days (Fig. 6A). TrκB mRNA was also upregulated in the model group compared with the sham-operated group after 7 days (P<0.05), and this effect was augmented further by high-dose TXC treatment on day 7 and day 28 (Fig. 6B).