Ursodeoxycholic acid protects against lung injury induced by fat embolism syndrome

Abstract Acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) is a life‐threatening disease with a high mortality rate, which was a common complication of fat embolism syndrome (FES). Ursodeoxycholic acid (UDCA) has been reported to exert potent anti‐inflammatory effects under various conditions. In vivo, perinephric fat was injected via tail vein to establish a rat FES model, the anti‐inflammatory effects of UDCA on FES‐induced lung injury were investigated through histological examination, ELISA, qRT‐PCR, Western blot and immunofluorescence. In vitro, human lung microvascular endothelial cells (HPMECs) were employed to understand the protective effects of UDCA. The extent of ALI/ARDS was evaluated and validated by reduced PaO2/FiO2 ratios, increased lung wet/dry (W/D) ratios and impaired alveolar‐capillary barrier, up‐regulation of ALI‐related proteins in lung tissues (including myeloperoxidase [MPO], vascular cell adhesion molecule 1 [VCAM‐1], intercellular cell adhesion molecule‐1 [ICAM‐1]), elevated protein concentration and increased proinflammatory cytokines levels (TNF‐α and IL‐1β) in bronchoalveolar lavage fluid (BALF). Pre‐treatment with UDCA remarkably alleviated these pathologic and biochemical changes of FES‐induced ALI/ARDS; our data demonstrated that pre‐treatment with UDCA attenuated the pathologic and biochemical changes of FES‐induced ARDS, which provided a possible preventive therapy for lung injury caused by FES.


| Perinephric fat preparation
Allogeneic rats were anesthetized with pentobarbital sodium (80 mg/ kg, i.p.), the adipose tissues were isolated and the fat homogenate was prepared using an ultrasonic cell breaker, then the homogenate was centrifuged at 4°C (3000 g × 10 minutes). The whole process was carried out under pathogen-free condition, and the supernatant was stored at −20°C for further studies.

| Perinephric fat injection
The collected fat solution stored at −20°C was put into a electricheated thermostatic water bath to be rewarmed. Repeat centrifugation if the solution was not clear and transparent. Then, the rats were drived into a rat retainer to expose the tail (Yuyan Instruments).
We sterilized the injection area and injected the fat via tail vein using a microsyringe. After injection, we applied pressure with sterile cotton balls to stop bleeding. Finally, we returned the rats to their cages and monitored their vital signs closely.

| Oil red O staining
The lung tissues were collected after all treatments. After fixation and dehydration, the tissues were cut into slices about 8~10 µm thick. Fresh or frozen lung tissue sections were prepared with propylene glycol for 2 mins and then incubated using oil red O solution for 6 mins. Next, a mixture of 85% propylene glycol were applied for differentiation; then, the sections were rinsed for 2 times. Finally, sections were incubated in haematoxylin for 1-2 mins and rinsed for 2 times.

| Immunofluorescence
Confluent HPMEC monolayers grown on gelatin-coated coverslips were subjected to immunofluorescent staining. After treatment, cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. Rhodamine-labelled phalloidin (Sigma) and VE-cadherin antibody were used to visualize F-actin filaments and adherens junctions. Cell nuclei were stained with DAPI (Beyotime Institute of Biotechnology, Haimen, China) at room temperature.
After labelling, cells were rinsed to remove excessive label and examined using a confocal laser-scanning microscope (Leica).

| Quantitative real-time reverse transcription PCR (qRT-PCR)
Total RNA was extracted from lung tissues using trizol reagent (Invitrogen) following the manufacturer's instructions. Total RNA was reversely transcribed into cDNA, and then, equal amounts of the cDNA products were applied for PCR amplification using the qRT-PCR thermal cycler (steponeplus, ABI). The expression values of MPO and P-selectin were normalized with β-actin mRNA levels.

| Enzyme-Linked ImmunoSorbent Assay (ELISA)
TNF-α and IL-1β levels were assayed by Quantikine ELISA kits according to the manufacturer's instructions.

| Analysis of arterial blood gas
To analyse arterial blood gas, rats were randomly allocated into four groups (n = 8) and, respectively, injected with saline, UDCA (20 mg/ kg), perinephric fat (500 µL/kg) and perinephric fat + UDCA. 12 h after administration, rats were anaesthetized and 2 mL arterial blood was drawn through the left carotid artery, blood samples were immediately analysed using a blood gas analyzer (Radiometer). Levels of PaO 2 and PaO 2 / FiO 2 were all recorded.
BALF was obtained at 12 h after treatments by placing a catheter into the trachea, through which 1 mL of cold phosphate buffer saline (PBS) was flushed back for 3 times. Samples were determined for cell counting and centrifuged for 300 g × 10 minutes at 4°C with the supernatant being analysed for inflammatory cytokines. Protein concentration in the cell-free BALF was determined using a BCA protein assay kit (Thermo Fisher, Scientific). The BALF neutrophils were calculated using Wright staining. TNF-α and IL-1β levels were assayed by Quantikine ELISA kits according to the manufacturer's instructions.

| Survival analysis
Adult male SD rats were divided into four groups (n = 10) randomly and injected with saline, UDCA (20 mg/kg), perinephric fat (500 μL/kg) and perinephric fat + UDCA, respectively. Survival was monitored till 96 h. Efforts were made to minimize rat suffering and humane endpoints (weight loss ≥ 20% of pre-experimental bodyweight) were used.

| Pathological evaluation
We used separate sets of rats for pathological evaluation. Slices used for haematoxylin and eosin staining were prepared following standard methods. Severity of lung injury was evaluated according to the following criteria: interstitial inflammatory infiltration, oedema, haemorrhage, atelectasis and hyaline membrane. A semi-quantitative scoring system was employed to evaluate the lung injury, and the scoring criterions were as follows: 0: normal appearance, 1: mild interstitial hyperaemia, polymorphonuclear leucocyte infiltration, 2: perivascular oedema and moderate pulmonary structural damage, 3: massive cell infiltration and moderate alveolar structure destruction, 4: massive cell infiltration and severe lung structural damage. 7,8

| Western blot
Equal amount of protein samples were loaded and separated on 10% SDS-PAGE, and the separated bands were then electroblotted onto PVDF membranes. The membranes were blocked and then probed with related primary antibodies at 4°C overnight; the membranes were visualized with HRP-conjugated antibody. Protein signals were detected by Image Quant Ai600 (General Electric Co.) using an enhanced ECL substrate (Thermo Fisher, Scientific). The protein expression was quantified using ImageJ software (National Institutes of Health). The targeted proteins were normalized to β-actin for statistical analysis.
HPMECs were treated with free fatty acids (FFAs) according to our previous studies. 9 FFAs are the main products of fat decomposition; they are also the direct cause of injury. Thus, a mixed solution consisted of 25% sodium-oleic acid, 25% sodium-palmitic acid, 25% stearic acid and 25% arachidonic acid, which were dissolved in 0.01 mmol/L NaOH solution, was used to treat HPMECs with a final total FFAs concentration of 0.5 mmol/L.

| Statistical analysis
Data were presented as their mean ± SEM (n = 8) and were analysed with Graph-Pad Prism 7.0 (Graph Pad Software). Statistical differences among groups were determined via Gaussian distribution prior to one-way analysis of variance (ANOVA) followed by Tukey's post hoc test. In the case of F achieved P < .05, Tukey's post hoc tests were further applied for analysis. Non-parametric tests were used if data were not normally variated, P < .05 presented a statistical significance.

| RE SULTS AND D ISCUSS I ON
Inappropriate accumulated neutrophils, dysregulated inflammation and altered alveolar-capillary permeability are the major pathophysiological characteristics of ALI/ARDS. In this study, histopathological examination revealed that lung tissues of FES model rats suffered severe damages, being characterized by interstitial oedema, haemorrhage and inflammatory cells infiltration (Figure 1A,B). However, the results indicated that UDCA pre-treatment but not post-treatment improved pathological features of FES-induced lung injury.
This might due to faster occurrence and more severe symptoms of  13,14 In response to increased TNF-α/IL-1β, VCAM1 and ICAM1 were remarkably activated and up-regulated in pulmonary microvascular endothelium ( Figure 2C). Encouragingly, our experimental results indicated that UDCA pre-treatment significantly attenuated all these pathological alterations and improved rat survival Limitations also existed in this study. UDCA was proven to be prophylactically protective, but not therapeutically in FESassociated lung injury. It is difficult to rule out the effects of UDCA on fat absorption, solubilization and metabolism. We are making efforts to clarify this question in our future work.
In summary, the current study showed that the administration of UDCA significantly attenuated ALI/ARDS in a rat FES model, which might offer a potential novel and effective precautionary measurement for lung injury.

ACK N OWLED G EM ENTS
This work was sponsored by the National Natural Science Foundation of China (No. 81772062 and No. 81370187).

CO N FLI C T S O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data included in this study are available upon request by contact with the corresponding author.