Materials and methods

Reagent

Artificial seawater was prepared with reference to the table of major components of seawater along the southeast coast provided by the Third Research Institute of the Bureau of Oceanography of China [16], with the exact mass to volume concentrations of major salts as follows: NaCl 35.0 g/L, MgCl2 5.0 g/L, MgSO4 7.0 g/L, CaCl2 1.1 g/L, KCl 0.7 g/L, NaHCO3 0.2 g/L, NaBr 0.1 g/L. Sodium chloride was obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China); magnesium chloride, magnesium sulfate, calcium chloride, potassium chloride, sodium bicarbonate and sodium bromide were all purchased from Beijing Sunshine Yingrui Biotechnology Co., Ltd (Beijing, China); phosphate buffer was supplied by Gibco (New York, NY, USA); and the BCA assay kit and Trypan blue were sourced from Thermo Fisher Scientific (Massachusetts, USA).

Animal studies

Healthy male C57BL/6 mice (6–8 weeks old, body weight 20–25 g) were provided by the Experimental Animal Center of the Chinese PLA General Hospital. After 1 week of acclimatization under standard laboratory conditions (temperature 22–25 °C, 12 h light-dark cycle, free access to food and water), mice were randomized into experimental groups using a computer-generated random numbers (SPSS 27.0) by an independent experimenter not involved in subsequent model preparation or outcome assessment; the randomization sequence was concealed in sealed envelopes until group allocation to ensure allocation concealment. The baseline body weight and general activity status of mice in each group were assessed and compared prior to modeling, with no significant intergroup differences observed (p>0.05).

The sample size for experiments was pre-determined via power analysis using G*Power 3.1 software, confirming that 5–7 mice per group were sufficient to detect significant intergroup differences for primary outcomes including survival rate, lung coefficient, and percentage of injured lung area on micro‑CT – a sample size consistent with our previously published SWD-ALI research with reproducible and statistically significant results [17]. Seven mice from each group were used for survival analysis, with death or 24 h survival set as the termination event, and survival monitored every 4 h over the 24‑hour observation period.

DD: direct drowning

Mice were individually placed in container filled with artificial seawater (6 cm depth, 25 °C). Mice were rapidly retrieved after 25 s and dried with filter paper. Vital signs, physical state, and survival status were closely observed and documented.

ND: nasal drip

The operator stabilized the mouse’s lower jaw with the left thumb and forefinger, supporting the torso with the remaining fingers and palm to maintain an upright position with head hyperextension. A pipette was used to instill 4 mL/kg artificial seawater into the nasal cavity, and the mouse was vertically rotated for 15 s following instillation to facilitate uniform pulmonary distribution of seawater. Seawater inhalation, physical state, and survival were recorded.

NEI: neck incision endotracheal injection

Mice were fixed in a supine position on a surgical board at a 60–70° angle to the ground. The anterior cervical skin and fascia were incised, muscle tissue was separated to expose the trachea, and a 1 mL syringe was inserted into the trachea to inject 4 mL/kg artificial seawater at a uniform rate over 2 min. The mouse was rotated vertically for 15 s after injection. Tracheal fluid changes, respiration, cough reflex, cyanosis, and survival were observed and recorded.

LEI: laryngoscopic endotracheal injection

Mice were secured in a supine position on a surgical board at a 60–70° angle to the ground. Isoflurane inhalation anesthesia was used for all invasive procedures (induction: 3–5 L/min for 1–2 min, maintenance: 1–2 L/min for ≤5 min) with strict control of anesthesia duration to avoid respiratory or cardiac dysfunction. The operator held a laryngoscope with the left hand, inserted it into the oral cavity via the right buccal commissure, and gently retracted the mouse’s tongue with the right hand. The laryngoscope was used to lift the epiglottis, revealing the inverted triangular white cartilage that opens and closes with respiration. 4 mL/kg artificial seawater was injected via a microsyringe during cartilage opening, and the mouse was vertically rotated for 15 s post-injection using the identical procedure described above. Respiration, cough reflex, cyanosis, and survival were observed and recorded.

The 2 mL/kg, 4 mL/kg, and 8 mL/kg artificial seawater doses were selected based on clinical translation and pre-experimental screening: pre-experiments confirmed that a 1 mL/kg dose induced no significant lung injury, while a 10 mL/kg dose resulted in 100 % mortality within 4 h. Thus, 2,4,8 mL/kg were set as low-medium-high doses to reflect a dose-dependent SWD-ALI response, with the 4 mL/kg dose consistent with our published SWD-ALI model for inducing stable moderate lung injury [17].

Lung coefficient

Mice body weights (B) were recorded before sampling by an investigator blinded to group allocation, and the wet weights (W) of the lungs were measured immediately after sampling. Lung coefficient was calculated as W/B ×100 %, a validated index for assessing the severity of pulmonary edema.

Mouse micro-CT imaging

Mice were anesthetized with isoflurane. Subsequently, the mice were positioned in the CT device and scanned using the Quantum FX microimaging system (PerkinElmer, USA). Each mouse was scanned at 70 kV, 88 μA and 36 mm field‑of‑view (FOV) for 4 min. All micro-CT image analyses were performed in a retrospective double-blinded manner by two independent radiologists with expertise in small animal imaging. Scan data were analyzed using Analyze 12.0 software with standardized horizontal settings. 3D-Slicer software was used to segment the lung and the damaged area, and Python was used to calculate the damaged area, percentage of damaged area, and average gray value. Based on the 95 % reference interval of the control group, early micro-CT diagnostic criteria for murine SWD-ALI were established accompanied by focal or diffuse alveolar exudation. Quantitative data of average gray value and damaged area ratio for all time points (0.5 h, 2 h, 3 h, 6 h, 24 h) were calculated and presented as a time-course to show the dynamic changes of lung injury.

H&E staining of lung tissues

The lung tissues were fixed with 4 % paraformaldehyde, embedded in paraffin, and cut into 5 μm sections. Subsequently, H&E staining was performed, and photographs for damage assessment by a pathologist blinded to group allocation. Lung injury severity was assessed using a semi-quantitative pathological scoring system, which evaluated alveolar edema, hemorrhage, inflammatory cell infiltration, and alveolar architectural disruption.

Bronchoalveolar lavage fluid (BALF) analysis

After cervical tracheostomy, tracheal intubation and fixation were performed in mice. Bronchoalveolar lavage was conducted three times using 0.9 mL ice-cold 0.9 % PBS each time, with two repeated aspirations for fluid collection. The total lavage volume was 2.7 mL, and the recovery rate exceeded 80 %. The centrifugal pellet was resuspended in PBS. Then, 10 μL of the cell suspension was mixed with 10 μL AOPI, loaded into a counting chamber, and the total cell number was determined with a cell counter. A 20 μL aliquot of each sample was used for H&E staining, and the percentage of neutrophils was calculated by an analyst blinded to group allocation.

Detection of protein concentration in BALF

The protein concentration in BALF supernatant was quantified using the Bicinchoninic Acid (BCA) assay, and the absorbance was measured at 562 nm with a microplate reader. The protein concentration of the sample was then calculated against the standard curve.

Statistical methods

SPSS 27.0 software was used for statistical analysis. A Blinding design was implemented for all experimental procedures: analysts of micro-CT images, pathological sections, and BALF cell counting were blinded to group allocation, with all samples coded with random numbers (decoded only after all data collection). Data were expressed as the mean ± standard error of the mean (SEM), and normality was assessed accordingly. One-way ANOVA was applied for comparisons among multiple groups with normal distribution and homogeneous variance, while the Kruskal-Wallis H test was used for non-normally distributed data. Survival curves were constructed by the Kaplan-Meier method and compared via the Log-rank test. Pearson correlation analysis was used to evaluate the correlation between micro-CT damaged area percentage and pathological total injury score. p<0.05 was considered statistically significant.

Ethical approval

All experimental protocols were approved by the Experimental Animal Welfare Ethics Committee of Chinese PLA General Hospital (Approval No. 2024-X20-69). All animal experiments were performed in strict accordance with the Guide for the Care and Use of Laboratory Animals (8th Edition, NIH) to minimize animal suffering and ensure humane animal care.