A total of 30 specific pathogen-free (SPF) rabbits (license no. BD20174321) were purchased from the Laboratory Animal Center (Jiangsu, China). They were randomly divided into: Control (n=10), model (n=10) and experimental groups (n=10). Rabbits in the model group and experimental group were treated with femoral venous injection of 0.1 ml/kg oleic acid to establish the oleic acid-induced ARDS model, while those in control group were injected with the same volume of normal saline. Whereas, rabbits in the experimental group received femoral venous injection of 5 mg/ml/kg sulforaphane (Aokai, Guangzhou, China) dissolved by normal saline, while those in the model group were injected with the same volume of normal saline. Within 12 h after treatment, the reactions and deaths of rabbits in each group were observed and recorded.

At 12 h after treatment of rabbits in each group, they were anesthetized with chloral hydrate, and the body weight of rabbits was recorded. Then blood was drawn from the femoral artery until the death of rabbits. The blood was placed at 4°C for 1 h for stratification, followed by centrifugation at 2,380 × g for 10 min. The upper-layer serum was taken, and rabbit lung tissues were excised and weighed, and the lung index (LI) was calculated: LI = total lung mass (g)/body mass (kg). After weighing, lung tissues were immediately divided into two parts; one part was stored in liquid nitrogen for reverse transcription-polymerase chain reaction (RT-PCR), while the other part was fixed in 4% paraformaldehyde for hematoxylin and eosin (H&E) morphological examination and immunohistochemistry (IHC). The study was approved by the Ethics Committee of Cangzhou Central Hospital (Cangzhou, China).

The alveolar damage coefficient [index of quantitative assessment (IQA)] was calculated following the steps below: Six H&E staining sections of lung tissues were taken from rabbits in each group, and 10 images were taken in the same field of view, and the number of damaged alveoli (containing more than 2 erythrocytes or neutrophils in alveoli) was calculated. The ratio of the number of damaged alveoli to the total number of alveoli was used as the IQA to evaluate the degree of lung injury.

Total ribonucleic acid (RNA) was extracted by using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) from lung tissues, and purified by using the extraction kit (Qiagen, Valencia, CA, USA) according to instructions provided by the manufacturer. The content of Nrf2 gene was detected by using the qRT-PCR kit, followed by quantification using the fluorescence quantitative detection system (Applied Biosystems, Foster City, CA, USA), with β-actin as the internal reference. Primer sequences of gene amplification are as follows: Nrf2 forward, 5′-CCCACACAAGGTTCGGCATCAC-3′ and reverse, 5′-TGGCGATTCCTCTGGCGTCT-3′; β-actin forward, 5′-CGCGCCATCAAGGAGAAGCTG-3′ and reverse, 5′-ATTGCCAATGGGTGATACCTG-3′.

Commercially-available ELISA kits were used to detect the serum Nrf2 in rabbits in each group. According to instructions provided by the manufacturer, Nrf2 was labeled by using the double-antibody sandwich method, and the optical density (OD) values were detected by using a microplate reader at dual wavelengths of 450 and 600 nm, and the sample concentration was calculated.

The nuclei and cytoplasm of lung tissues were stained with H&E (Google Biological Co., Ltd., Wuhan, China), and sections were dehydrated with gradient ethanol and then sealed with neutral gum. The Nrf2 protein (1:500; Cell Signaling Technology, Danvers, MA, USA) expressed in the nuclei was specifically labeled by using the two-step method and IHC assay kits (Zhongshan Golden Bridge Biotechnology Co., Ltd., Guangzhou, China). H&E and IHC sections were observed under an inverted microscope (DM-5000B; Leica Store Wetzlar, Wetzlar, Germany). At least 3 regions were photographed in each section, and brown yellow-stained cells in IHC staining were positive cells; the proportion of Nrf2-positive cells in each field of view in the total cells ≤5% indicated negative, while that >5% indicated positive.

Blood was drawn from the femoral artery under anesthesia, and the blood gas indexes (PaO₂, PaCO₂ and SaO₂) were determined by using a blood-gas analyzer (Hunan Sanhe Apparatus, Changsha, China).

Experimental results were analyzed by using GraphPad Prism statistical software 5.01 (GraphPad Software, Inc., La Jolla, CA, USA). Measurement data are presented as mean ± SD, and one-way analysis of variance (ANOVA) was used for the comparison of differences among groups. P<0.05 was considered to indicate a statistically significant difference.

Within 12 h after modeling, rabbits in both model and experimental groups spit pink frothy sputum, including 5 rabbits (50%) in the model group, and only 2 rabbits (20%) in the experimental group. There were 6 deaths (60%) in the model group and 3 deaths (30%) in the experimental group (Table I).

Compared with the rabbit lung tissues in control group, obvious alveolar edema, alveolar interstitial serous exudate, aggregation of a large number of inflammatory cells, and partial pulmonary septal thickening could be seen in rabbits in model group. In experimental group, there was no aggregation of a number of inflammatory cells, but only a small amount of serous exudate (Fig. 1).

There was no statistically significant difference in the comparison of Nrf2 mRNA level in rabbit lung tissues between model group and control group (p>0.05), and Nrf2 mRNA level in rabbit lung tissues in experimental group was significantly increased compared with that in model group (p<0.05; Fig. 2).

ELISA showed that there was no statistically significant difference in the comparison of Nrf2 protein level in rabbit lung tissues between model and control groups (p>0.05), and Nrf2 protein level in rabbit lung tissues in experimental group was significantly increased compared with that in model group (p<0.05), which were consistent with results of RT-PCR (Fig. 3).

Nrf2 protein was mainly located in the cytoplasm. As shown in Fig. 4, the Nrf2 expression was negative in rabbit lung tissues in control group, which was negative in model group, but strongly positive in the experimental group.

LI was selected to show the degree of pulmonary edema (Table II).

Compared with those in control group, the blood gas indexes (PaO₂, PaCO₂ and SaO₂) in the model and experimental groups were obviously decreased (p<0.05); but the blood gas indexes in the experimental group were significantly increased compared with those in the model group (p<0.05; Table III).