Animal use protocols were approved by People's Hospital of Zhengzhou University Committee on the Use and Care of Animals (Zhengzhou, China). Animals were provided by Henan Experimental Animal Center of Zhengzhou. A total of 130 newborn Sprague-Dawley (SD) rats at postnatal day 4 (weight, 5–10 g) were used in the present study and were reared with their dams. The dams and the newborn rats were housed at 18–25°C in an atmosphere of 50–70% relative humidity, with a 12-h light/dark cycle. Rats were randomly divided into five groups: IVH control group (n=30; Group 1); G-CSF treatment group (n=30; Group 2); lithium chloride treatment group (n=30; Group 3); combination treatment group (n=30; Group 4); and sham surgery group (n=10; Group 5). Rats in Group 1 received hypodermic injection of saline on days 2, 4, 6, and 8 and intraperitoneal injection after 2 days of IVH. Rats in Group 2 received a hypodermic injection of 50 µg/kg/day G-CSF (Sigma-Aldrich, St. Louis, MO, USA) on days 2, 4, 6 and 8, and intraperitoneal injection of saline with equal amounts with lithium chloride after 2 days of IVH. Rats in Group 3 were given a hypodermic injection of saline with equal amounts with G-CSF and intraperitoneal injection of 3 mmol/kg/day lithium chloride after 2 days of IVH. Group 4 rats received combination treatment with a hypodermic injection of 50 µg/kg/day G-CSF on days 2, 4, 6 and 8, as well as intraperitoneal injection of 3 mmol/kg/day lithium chloride after 2 days of IVH. Rats in Group 5 were subjected to sham surgery and received no treatment.

The IVH rat model was established as previously described by Ahn et al (20) and Lodhia et al (21). Briefly, SD rats were anesthetized using 1.5–2% isoflurane (Sigma-Aldrich) in an oxygen-enriched atmosphere (20,21). A total of 100 µl fresh whole blood from the pregnant rats was slowly infused into the right ventricles over 5 min at a rate of 10 µl/min, 1.7 mm lateral to the sagittal suture, 3.8 mm posterior to the coronal suture and at a depth of 3.5 mm. After 10 min, equal amount of fresh maternal whole blood was infused into the left ventricles. For rats in Group 1, the same amount of saline was infused into the right and left ventricles rather than fresh maternal whole blood, according to the aforementioned procedure. The animals were then allowed to recover from anesthesia. Rats received intraperitoneal injection of 10 mg/kg/day bromodeoxyuridine (BrdU; Sigma-Aldrich; one day after surgery) to mark neural stem cells under proliferation. Rats in group 5 underwent a sham operation without blood injection.

Rats in each group were divided into the day 11, 18 and 32 time points. At each time point, rats were fixed on the experimental platform under deep anesthesia with 400 mg/kg chloral hydrate (Sigma-Aldrich). Next, the hair on the chest and abdomen was removed and the heart was exposed. The rats were decapitated and the brain was placed in paraformaldehyde solution. Brain tissue was collected in the periventricular cerebral mass, corpus callosum region, subventricular zone (SVZ) and hippocampal dentate gyrus and cut into sections of 2 mm. Subsequently, the obtained tissue was moved into sucrose solution treated with diethyl pyrocarbonate (Sigma-Aldrich) for 48 h. The tissue was then sliced into 5 µm using a freezing microtome and loaded onto glass slides. The samples were wrapped up with aluminized paper and stored at −70°C.

Cell death of nerve cells in the periventricular cerebral mass, corpus callosum region, SVZ and hippocampal dentate gyrus brain tissue was assessed using the immunofluorescent terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling (TUNEL) in the ApopTag Fluorescein In Situ Apoptosis Detection kit (cat. no. S7110; Chemicon; Merck Millipore, Temecula, CA, USA) according to the manufacturer's instructions. The nucleus presenting yellow or light yellow bulk or appearing granular under the light microscope was considered as TUNEL-positive. A blinded evaluator counted the number of TUNEL-positive nuclei in the brain tissue samples from the periventricular cerebral mass, corpus callosum region, SVZ and hippocampal dentate gyrus. A total of 10 random non-overlapping fields of view from each brain were counted.

In order to perform immunoblotting analysis, brain tissue obtained from rats in each group was lysed by radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology, Inc., Shanghai, China) and homogenates were centrifuged at 9,200 × g for 15 min at 4°C. Then the supernatant was collected and used for protein concentration determination using a Bradford protein assay kit (Beyotime Institute of Biotechnology). A total of 15 mg protein was loaded onto each lane of an 8% sodium dodecyl sulfate-polyacrylamide gel containing 4 mol/l urea, and electrophoresis was performed under standard conditions. Next, the protein sample was electrophoretically transferred to Immobilon-P polyvinylidene difluoride membranes (Merck Millipore) by semidry blotting. Subsequent to blocking in 3% nonfat dry milk in Tris-buffered saline/Tween 20 (TBST; containing 20 mmol/l Tris base, pH 7.6, 137 mmol/l NaCl and 0.05% Tween 20) for 1 h at room temperature, the membranes were incubated with antibodies against rabbit polyclonal anti-B cell lymphoma-2 (Bcl-2; 1:500; cat. no. ab59348; Abcam, Cambridge, UK), rabbit polyclonal anti-Bcl-2-associated X protein (Bax; 1:500; cat. no. ab69643; Abcam), or rabbit polyclonal anti-GAPDH (1:2,500; cat. no. ab9485; Abcam) overnight at 4°C. The membranes were washed in TBST and then incubated for 1 h at room temperature with horseradish peroxidase-conjugated goat anti-rabbit IgG H&L (1:10,000; cat. no. ab97051; Abcam). Immunoreactive bands were visualized in the linear range with enhanced chemoluminescence (Sigma-Aldrich). The scanned images were semi-quantitated using Quantity One software (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The relative integrated density values were calculated using FluorChem 2.0 software (Alpha Innotech Corporation, San Leandro, CA, USA) and normalized with GAPDH.

For the double immunofluorescence detection of BrdU along with glial fibrillary acidic protein (GFAP), neuronal nuclei (NeuN) or polysialylated-neural cell adhesion molecule (PSA-NCAM). Brain sections were washed in phosphate-buffered saline (PBS) and pre-incubated for 60 min in blocking buffer (5% normal donkey serum in PBS with 0.3% Triton X-100; both obtained from Sigma-Aldrich). Next, the brain tissue sections were incubated with the sheep polyclonal anti-BrdU antibody (1:1,000; cat. no. M20105S; Biodesign Inc., Saco, ME, USA), in addition to the goat polyclonal anti-GFAP (1:400; cat. no. sc-9065; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA), the mouse monoclonal anti-NeuN (1:500; cat. no. MAB377; Chemicon; Merck Millipore) or the mouse monoclonal anti-PSA-NCAM antibody (1:1,000; cat. no. MAB5324; Chemicon; Merck Millipore) for 12 h at 4°C. Subsequently, the sections were incubated with the fluorescein isothiocyanate (FITC)-labeled donkey anti-sheep IgG (1:500; Molecular Probes; Thermo Fisher Scientific Inc., Eugene, OR, USA) and rhodamine-labeled horse anti-mouse IgG (1:500; Chemicon; Merck Millipore) antibodies for 1 h. The sections were then rinsed in the dark, mounted on polylysine-coated slides and coverslipped using anti-fadent solution (Biomart, Shanghai, China). The tissue sections were viewed and photographed using epifluorescence on an Axiophot epifluorescence microscope with appropriate filter sets for fluorescein (excitation, 450–490 nm; emission, 514–565 nm) and Texas red (excitation, 530–585 nm; emission, >615 nm).

Data are presented in the present study as the mean ± standard deviation. Statistical differences among groups were determined using a Student's t-test. Analysis of variance with a Newman-Keuls post-hoc test was used for multiple comparisons. SPSS version 13.0 software (SPSS, Inc., Chicago, IL, USA) was used to analyze the data. A value of P<0.05 was considered to indicate statistically significant differences.

To investigate the effect of combination treatment of G-CSF and lithium chloride on hydrocephalus following IVH, rats were divided into five groups (Group 1, IVH control; Group 2, G-CSF treatment; Group 3, lithium chloride treatment; Group 4, combination treatment; and Group 5, sham surgery). G-CSF was used to mobilize endogenous neural stem cells (10). In addition, lithium chloride was observed to have neuroprotective effects by promoting neural stem cell differentiation (11). As shown in Fig. 1A, severe hydrocephalus occurred in the majority of rats in Group 1 after IVH, compared with the rats only subjected to sham surgery (Group 5; P<0.001). In addition, the hydrocephalus incidence after IVH in Group 2 and Group 3 rats was significantly decreased compared with that in Group 1 rats (both P<0.001). Furthermore, a significant reduced hydrocephalus incidence was observed in Group 4 when compared with that in Groups 2 and 3 (P=0.028; P=0.015). Images of brain tissue obtained from rats in the different groups were captured, and are presented in Fig. 1B. The brain of rats in Group 1 appeared to be larger in size when compared with that in other groups. The brains of rats in Groups 2 and 3 were smaller than those in Group 1, however they were still larger in size compared with those in Group 4, which appeared to be closer to the normal brain size in Group 5 (Fig. 1B). These data indicated that combination treatment with G-CSF and lithium chloride was more effective against hydrocephalus after IVH compared with single drug administration, and significantly attenuated the development of hydrocephalus.

The current study further detected the nerve cell apoptosis in brain tissue samples obtained from the five groups using TUNEL assay. It was observed that the apoptotic cell number in Group 1 was significantly higher than that in Group 5 (P=0.002). As shown in Fig. 2A, the amount of apoptotic cells in Groups 2, 3 and 4 were significantly reduced compared with that in Group 1. Furthermore, apoptosis of nerve cells was significantly reduced in Group 4 when compared with that in Groups 2 and 3 (Fig. 2A; both P=0.001). Images of the apoptosis assay in brain tissue samples from the five groups are shown in Fig. 2B, further verifying these results. These findings suggested that combination treatment with G-CSF and lithium chloride significantly reduced the nerve cell apoptosis induced by hydrocephalus subsequent to IVH.

As Bcl-2 and Bax are known to be involved in cell apoptosis, their protein expression levels were examined in nerve cells of brain tissue samples obtained from the various rat groups. As showed in Fig. 3, the protein expression of Bcl-2 in Groups 2 and 3 was significantly higher compared with that in Group 1; however, the expression in Groups 2 and 3 was significantly lower compared with that in Group 4 (P=0.0015; P=0.001). By contrast, the expression of Bax was significantly upregulated in Group 1, while a lower protein level was observed in Group 4, when compared with Bax expression in Groups 2 and 3. In addition, the expression levels of Bcl-2 and Bax were significantly decreased or increased in Group 1 compared with that in Group 5, respectively (P=0.003; P=0.0025).

In order to detect the differentiation of nerve cells in brain tissue samples obtained from the five rat groups, the expression levels of GFAP, NeuN and PSA-NCAM were examined using a double immunofluorescence labeling method with BrdU to evaluate cell proliferation intensity. As shown in Fig. 4, the results demonstrated that the expression levels of BrdU/GFAP, BrdU/NeuN and BrdU/PSA-NCAM were significantly upregulated in Group 1 compared with those in Group 5 (P=0.03). In addition, the expression levels of BrdU/GFAP, BrdU/NeuN and BrdU/PSA-NCAM in Groups 2 and 3 were significantly increased when compared with that in Group 1 (Fig. 4A). In addition, highest expression levels of BrdU/GFAP, BrdU/NeuN and BrdU/PSA-NCAM were observed in Group 4. The fluorescence distribution images shown in Fig. 4B further verify this differential expression. These data indicated that combination treatment with G-CSF and lithium chloride was more effective at promoting nerve cell differentiation compared with single drug administration.