All animal experiments were conducted in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals of Laboratory Animals. All the experimental procedures involved in this study were approved by the Animal Ethics Committee of Xuzhou Medical University (protocol: 202007A102).

Male Sprague-Dawley (SD) rats were provided by the Experimental Animal Center of Xuzhou Medical University. All rats were kept in pathogen-free conditions and housed under a light-dark cycle for 12 h (20 ± 1°C, and 55 ± 10% humidity), with access to food and water ad libitum.

1,25-D3 and VDR antagonist pyridoxal-5-phosphate (P5P) were purchased from Med chem Express. 2,3,5-triphenyltetrazolium chloride (TTC) was purchased from Sigma-Aldrich. The primary antibodies: anti-VDR antibody (14526-1-AP) and anti-β-actin antibody (20536-1-AP) were purchased from Proteintech. Anti-VEGF antibody (ab1316) and anti-platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31) antibody (ab222783) were purchased from Abcam. Anti- isolectin-B4 (IB4) antibody (DL-1207) was purchased from Vector Laboratories. The primary antibodies of anti-TGF-β (3711S), anti-Smad2/3 (8685T), anti-p-Smad2 (18338T) and anti-p-Smad3 (9250T) were purchased from Cell Signaling Technology. MCAO threads (2636-100, diameter of the head of nylon filament: 0.36 ± 0.02 mm) were purchased from Beijing Shadong Biotechnology. Reverse transcription kit, SYBRGreen^® PremixExTaq™ II kits were purchased from Takara.

A total of 120 male Sprague-Dawley rats weighing 250 to 280 g were used in this study. Rats were randomly assigned resulting in five groups, each consisting of 24 rats:sham group, middle cerebral artery ischemia-reperfusion group (MCAO), MCAO + dimethyl sulfoxide group (DMSO), MCAO + 1,25-D3 group (1,25-D3) and MCAO + 1,25-D3 + VDR antagonist (P5P, Pyridoxal-5-Phosphate) group (1,25-D3 + P5P). Rats from all the groups were subjected to a 90-min middle cerebral artery occlusion followed by reperfusion as previously described (10). The neurological function was scored according to the neurological deficit score established by Longa et al. (22). Grade 0: no neurological deficit; Grade I: mild focal neurological deficit (inability to fully extend the ischemic forelimb); Grade II: moderate focal neurological deficit (circling to the ischemic scar); Grade III: severe focal neurological deficit (dumping to the ischemic scar); Grade IV: lethargic and unable to walk spontaneously. Neurological severity scores of I to III were considered successful modeling, otherwise the corresponding number of rats was eliminated and supplemented. Rats of the 1,25-D3 group were administered intravenously with 1,25-D3 at a dose of 5 mg/kg (dissolved in 2% DMSO) 30 min before reperfusion, and the 1,25-D₃ + P5P group was administered intravenously with 1,25-D3 at a dose of 5 mg/kg (dissolved in 2% DMSO) and P5P at a dose of 0.4 mg/kg (dissolved in 0.9% NaCl). The same volume of DMSO was given in the DMSO group intravenously. The drug was given once a day for 3 consecutive days.

After anesthesia, the brains were refrigerated at −20°C for 20 min and then consecutively sliced into coronal slices, each slice being about 2 mm thick. Then they were placed into 2%TTC and incubated at 37°C for 20 min. After staining, the brain slices were fixed overnight in 10% paraformaldehyde and photographed by camera. Image J was used to calculate the percentage of cerebral infarction volume. The percentage of cerebral infarct volume in rats (%) = (volume of the normal cerebral hemisphere - volume of the non-infarct cerebral hemisphere on the infarct side)/volume of the normal cerebral hemisphere ^*100%.

Cortical cerebral blood flow (CBF) was monitored by a laser speckle flow imaging technique 3 days after reperfusion. All procedures were performed under double-blind conditions. Briefly, rats were anesthetized and disinfected using iodophor, and the skull was exposed. The fascia attached to the skull was removed as much as possible and 0.9% saline was added to maintain the liquid level. Images and quantification of cerebral blood flow in the penumbra can be accessed by the laser speckle flow imaging technique (RFLSI III, RWD, China).

The rats in each group were sacrificed 3 days after reperfusion and fixed for immunofluorescence of CD31 and IB4. Sections were incubated with 5% BSA for 1 h and then overnight with anti-CD31 antibody (1:200) at 4°C. After washing with PBST (0.1% Triton X-100 in 0.1 M PBS), fluorescent secondary antibody (1:500) and IB4 were incubated at room temperature for 1 h, then incubated with DAPI. Ultra-high resolution inverted fluorescence microscope (Stellaris5, Leica) was used for observation, and appropriate fluorescent light source was selected, and photos were taken. The vascular length and tip cell count was quantified by Image J.

The fresh tissue around the cerebral infarction was cryopreserved with liquid nitrogen, homogenized, centrifuged, and the total protein was extracted. The protein concentration was detected by BCA protein assay kit (Beyotime Biotechnology) and the 5 ug/ul system was prepared. Protein was boiled at 100°C for 5 min, then separated by SDS-PAGE and transferred to NC membrane (Cytiva). After incubated with 5% skimmed milk at room temperature for 2 h, incubated with the primary antibodies at 4°C overnight (1:1,000), incubated with the secondary antibodies at room temperature for 1 h (1: 10,000), and developed by ECL colorimetric method. The Image J was used to measure the gray value of the strip. The expression of the target protein band is expressed by the gray value of the target protein band/β-actin gray value.

Total RNA was extracted from peri-infarct tissues by Trizol. Complementary deoxyribonucleic acid (cDNA) was reverse transcribed and used as a template for the reaction by fluorescence quantitative real-time PCR apparatus. The target gene fragment was amplified according to the steps of the Power SYBR Green PCR Master Mix kit (Takara). Real-time PCR reaction conditions were as follows: pre-denaturation at 95°C for 10 s, denaturation at 95°C for 5 s, annealing and extension at 60°C for 20 s, a total of 40 cycles. After the reaction was complete, the dissolution curve program was run to detect the specificity of the amplified products. The following primers were used: TGF-β1_fwd 5'-3': TGAGTGGCTGTCTTTTGACG, TGF-β1_rev 5'-3': GGTTCATGTCATGGATGGTG. TGF-β2_fwd 5'-3': GTGATTTCCATCTACAACAGTACC, TGF-β2_rev 5'-3': TATAAACCTCCTTGGCGTAGTAC; TGF-β3_fwd 5'-3': CCCAACCCCAGCTCCAAGCG; TGF-β3_rev 5'-3': AGCCACTCGCGCACAGTGTC; VDR_fwd 5'-3': CCACCGGCAGAAACGTGTAT; VDR_rev 5'-3': TGCCTTGTGAGAGGCTCTAGGA; VEGF_fwd 5'-3': CCGTCCTGTGTGCCCCTAATG; VEGF_rev 5'-3': CGCATGATCTGCATAGTGACGTTG; GAPDH_fwd 5'-3': GCATCTTCTTGTGCAGTGCC and GAPDH_rev 5'-3': TACGGCCAAAT CCGTTCACA.

GraphPad Prism 5 (Graph Pad Software Inc, La Jolla, CA) was used to generate graphs and perform statistical analysis. Normal distribution was determined by the Kolmogorov-Smirnov test. One-way ANOVA was used for comparison among multiple groups, and Turkey′s post hoc test was used for further pairwise comparison. Data were expressed as Mean ± SEM. Differences were considered statistically significant at p < 0.05.

In order to confirm the effect of 1,25-D3 on stroke, we analyzed the volume of cerebral infarction 3 days after MCAO. Compared with sham group, the volume of cerebral infarction in MCAO group was significantly increased. There was no significant difference in cerebral infarction volume in DMSO and MCAO group. The cerebral infarct volume was decreased in 1,25-D3 group compared with the DMSO group. While it was more in 1,25-D3 + P5P group than 1,25-D3 group (Figure 1). It is suggested that 1,25-D3 treatment can significantly reduce the cerebral infarct volume and reduce the brain injury in stroke rats. VDR antagonist P5P partially reversed the neuroprotective effect of 1,25-D3.

We analyzed the CBF of peri-infarcted cortex in different groups of rats 3 days after ischemia-reperfusion using a laser speckle imaging system. The results showed that compared with the sham group, the CBF around the infarct cortex was significantly decreased in the MCAO group. Compared with the DMSO group, the CBF at the infarct margin was increased in the 1,25-D3 group, which was decreased in the 1,25-D3 + P5P group (Figure 2). Namely, 1,25-D3 increased CBF recovery in ischemia-reperfusion rats, while VDR antagonist P5P partially reversed the neuroprotective effect of 1,25-D3.

The results of CD31 staining suggested that the length of vessel at various positions after ischemia-reperfusion differ. The microvascular length was longer than that in the infarct core area (Figures 3A,B,D). Comparison of different groups showed that the microvascular length was lower in the MCAO group than in the sham group, higher in the 1,25-D3 group than in the DMSO group. In addition, the microvascular length in 1,25-D3+P5P group was lower than in the 1,25-D3 group, suggesting that P5P can partially reduce the microvascular increase induced by 1,25-D3 (Figures 3C,E).

To further understand the role of 1,25-D3 in neovascular development, we used IB4 fluorescence staining to look for neovascular sprouting as it is a marker of angiogenesis after we have confirmed that 1,25-D3 could increase CD31-labeled microvascular density after ischemia-reperfusion. IB4 staining showed that 1,25-D3 increased IB4⁺ tip cells in the peri-infarct area while P5P reversed the improvement of vascular development induced by 1,25-D3 (Figure 4).

In order to explore the angiogenic effects of 1,25-D3 on TGF-β/Smad2/3 signaling pathway, we analyzed the expression of VEGF, TGF-β/Smad2/3 and VDR at different time points: 6 h, 12 h, 24 h, 3 d, 5 d, 7 d after reperfusion. We found that compared with the sham group, the expression of VDR decreased gradually at 6 h after ischemia-reperfusion. VEGF increased gradually at 12 h after stroke, reaching the peak at 3 d and then decreased slowly. The expression of TGF-β increased gradually at 12 h after stroke and reached the peak at 5 d. The phosphorylation levels of Smad2 and Smad3 increased at 12 h after reperfusion, reached the peak at day 3 after reperfusion, and then decreased gradually (Figure 5). We chose 3 days after reperfusion as the time point to measure the expression of angiogenesis-related proteins induced by 1,25-D3. Results showed that the expression of VDR, TGF-β, p-Smad2/Smad2, p-Smad3/Smad3, and VEGF in the periinfarcted cortex was significantly higher than DMSO group 3 days after stroke. The effect can be partially reversed by P5P. These results were confirmed by western blotting (Figure 6).

RT-qPCR analysis reveals the expression of VDR, TGF-β1, TGF-β2, TGF-β3, and VEGF mRNA levels in the peri-infarct cortex after 1,25-D3 treatment. The results showed that compared with the sham group, the expression of VDR mRNA decreased in MCAO rats. 1,25-D3 activated the VDR, TGF-β1, TGF-β2, TGF-β3 and VEGF mRNA level induced by ischemia, while the increase was partially reversed by P5P treatment (Figure 7). Therefore, 1,25-D3 promotes angiogenesis after stroke by activating VDR, thus promotes the activation of TGF-β subtypes and the expression of VEGF mRNA.