Fourteen days after model establishment, the vascular endothelium specimens were collected from executed normal and ICU mice (4 mice each group). The specimens were fixed in 4% paraformaldehyde for 24 h and then dehydrated with 80%, 90%, and 100% gradient ethanol and nbutyl alcohol before paraffin embedding at 60oC. Paraffin-embedded specimens were cut into 5 μmserial sections. Xylene dewaxing was performed after the specimens were paved at 45oC and later baked at 60oC for 1 h. The specimens were dehydrated with gradient ethanol and soaked in 3% H2O2 solution for 10 min, followed by 90-s high pressure antigen repair. Then, the specimens were cooled to room temperature. After that, the specimens were added with 100 μL of 5% bovine serum albumin (BSA) and received 30-min incubation at 37oC. The specimens were added with 100 μL primary rabbit anti mouse monoclonal antibodies (Abcam Inc., Cambridge, UK) to ICAM-1 (ab93952, 1 : 1000 - 1 : 4000) or CD31 (1 : 1000 - 1 : 4000, ab2836), and then overnight incubation was performed at 4oC. The 30-min specimen incubation was operated with biotin labeled goat anti rabbit secondary antibody working fluid (HY90046, 1 : 100; Shanghai Hengyuan Science & Technology Co., Ltd., Shanghai, China) at 37oC. Subsequently, the specimens were sliced, followed by 30-min culture with streptomycin-avidin-biotin-peroxidase complex (Beijing Zhongshan Biotechnology Co., Ltd., Beijing, China) at 37oC. The specimens were developed by 3,3′-diaminobenzidine (Beijing Bioss Biotechnology Co., Ltd., Beijing, China) at room temperature before 5-min staining with hematoxylin and 4-s soaking in 1% hydrochloric acid alcohol. The average OD value of the ICAM-1 positive staining was assessed using an Image-Proplus software (Media Cybernetics, Inc., Silver Spring, MD, USA) at high magnification. The ratio of ICAM-1 positive cells to total cells was calculated by randomly selecting 5 high-power visual fields (× 400) with 100 cells in each field from each section (Xiong et al., 2012).
