All animal experiments were performed according to the guidelines of the Chinese Council on Animal Care and ethical approval was obtained for the use of animals prior to the commencement of the study from the Ethics Committee of the West China Second University Hospital, Sichuan University. Male BALB/c mice (weighing 18-22 g) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. The BALB/c mice were housed under standard conditions (12-h light/dark cycle, temperature of 25-27°C and a humidity of ~40%) with free access to food and water throughout the duration of the experiments. All mice were randomly divided into 4 groups (n=6/group) as follows: i) The control group; ii) the LPS group; iii) the LPS + agomir-93 group; and iv) the LPS + agomir-NC group. In the control group, mice were injected intravenously (via the tail vein) with an intra-tracheal instillation of 2 mg/kg normal saline (NS); in the LPS group, mice were injected intravenously (via the tail vein) with an intra-tracheal instillation of 2 mg/kg LPS; in the agomir-93 or agomir negative control (NC) groups, mice were injected intravenously (via the tail vein) with agomir-93 or agomir NC [8 mg/kg (22), RiboBio Co., Ltd.] at 24 h prior to the injection of 2 mg/kg LPS. Pentobarbital sodium (50 mg/kg, intraperitoneal injection) was used for anesthesia prior to each surgical procedure, and all efforts were made to minimize animal suffering. No mice were found dead during the process of anesthesia. If an animal reached the defined humane endpoints [a body weight loss of >15% in 1-2 days or an overall body weight loss of >20%, or displayed obvious signs of suffering (e.g., lethargy, squinted eyes, dehydration or hunched back)], it was humanely euthanized. Sacrifice was performed by an intraperitoneal injection of pentobarbital sodium (50 mg/kg) followed by cervical dislocation, and mortality was confirmed by observation (23) and the lung tissues and bronchoalveolar lavage fluid (BALF) were then collected for subsequent analysis.

In addition, the survival experiments were performed in 4 groups of mice (n=10/group) (LPS, Control, LPS + agomir-93 and LPS + agomir-NC). If an animal reached the defined humane endpoints mentioned above, it was humanely euthanized. The remaining mice were humanely euthanized 96 h later. The survival rate of the mice was observed from 0 to 96 h using the Kaplan Meier method.

Total RNA was isolated from the lung tissues or BALF using the miRNeasy mini kit (Qiagen, Inc.). The reverse transcription of miR-93 was performed using the miScript II RT kit and the reverse transcription kit (Invitrogen; Thermo Fisher Scientific, Inc.), respectively. miR-93 expression was measured using Exiqon SYBR-Green Master Mix (Exiqon) on a Light Cycler instrument (Bio-Rad Laboratories, Inc.). The reaction mixtures were denatured at 95°C for 3 min, followed by 40 two-step cycles of 95°C for 10 sec and 60°C for 30 sec. The primers for RT-qPCR analysis were as follows: miR-93 forward, 5′-AGG CCC AAA GTG CTG TTC GT-3′ and reverse, 5′-GTG CAG GGT CCG AGG T-3′; U6 forward, 5′-TGC GGG TGC TCG CTT CGC AGC-3′ and reverse, 5′-CCA GTG CAG GGT CC GAG GT-3′. miRNA relative expression levels were analyzed using the 2^−ΔΔcq method (24) and determined by normalization to U6.

At 24 h after the LPS challenge, the mice were sacrificed and the left lung tissues of mice were obtained and fixed in in 4% (v/v) paraformaldehyde, embedded in paraffin, sectioned at 4 µm thickness. Then, the fixed lung tissues were stained with hematoxylin and eosin (H&E) for the evaluation of the severity of lung injury. Images were captured using a microscope (Nikon Corp.), and the lung injury score was assessed by two pathologists as previously described (25). In brief, i) the thickness of alveolar walls; ii) infiltration of inflammatory cells; iii) hemorrhage; and iv) alveolar congestion were graded according to the following scale: 0, no damage; 1, mild damage; 2, moderate damage; 3, severe damage; 4, maximal damage.

Pulmonary capillary permeability was assessed using the Evans blue (EB) dye extravasation method, as previously described (26). Briefly, EB dye (20 mg/kg, Sigma-Aldrich; Merck KGaA) in normal saline was injected into the mice in each group (the Control group; the LPS group; the LPS + agomir-93 group; the LPS + agomir-NC group; n=6/group) through via the tail vein prior to the termination of the experiment. The mice were sacrificed after dye injection at 2 h, and EB dye was then extracted using formamide for 18 h at 60°C and the absorbance was measured at 620 nm. The dye concentration was reported as the ratio of absorbance relative to the amount of dry lung tissue (µg/100 mg).

The lung W/D ratio was used to calculate pulmonary edema. The right lungs of the mice were obtained and immediately weighted, then placed in an incubator at 80°C for 60 h to obtain the dry weight.

The concentrations of IL-6 (ab100712), IL-1β (ab197742) and TNF-α (ab100747) in BALF were analyzed using ELISA kits (Abcam) according to the kit instructions.

Immunohistochemistry (IHC) and IHC scoring systems were performed as described previously (27,28). 4-µm-thick formalin-fixed, paraffin-embedded serial sections of lung tissue were blocked and then incubated with a primary antibody specific for nuclear phosphorylated (p-)p65 (1:4,000 dilution; ab53489, Abcam) in 10% rabbit serum overnight at 4°C. The sections were then treated with a secondary antibody conjugated to horseradish peroxidase (1:10,000 dilution; ab7090, Abcam) overnight at 4°C. After the colorimetric reaction was completed, the sections were observed under a light microscope (Eclipse E100, Nikon; ×40 magnification). IHC scores are calculated as follows: IHC score=1× (number of weakly stained cells in the field) + 2× (number of moderately stained cells in the field) + 3× (number of intensely stained cells in the field) (29).

RAW 264.7 macrophages are widely used as a model of LPS-induced ALI, and play an essential role in the regulation of the inflammatory response (30,31). Therefore, RAW 264.7 cells were selected for use in an ex vivo experiment as the present study focused on the ALI-induced inflammatory response. The RAW264.7 cell line was obtained from ATCC. The cells were maintained in DMEM supplemented with 10% FBS (Sigma-Aldrich; Merck KGaA), 1% penicillin and streptomycin (Sigma-Aldrich; Merck KGaA) at 37°C and a 5% CO₂ incubator.

When the RAW264.7 cells in a 6-well plate grown to approximately 80% confluency, miR-93 mimics (20 nmol/l) and miR-93 inhibitor (20 nmol/l) were transfected into the cells at 37°C for 24 h, using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). miR-93 mimics, miR-93 inhibitor and the corresponding control vectors were purchased from RiboBio Co., Ltd.

miRNA target prediction tools, including miRanda (https://miranda.org.uk/) and TargetScan Release 7.0 (http://targetscan.org/) were used to search for the putative targets of miR-93. The dual-luciferase reporter assay was performed as described previously (32). Briefly, partial sequences of TLR4 3'-UTR containing miR-93 binding sites was amplified by PCR and cloned into the dual-luciferase reporter vector (pmirGLO; Promega Corporation) (wild-type pmirGLO-TLR4-3'-UTR, wt). Mutation of the predicted binding site (mutation pmirGLO-TLR4-mut-3'-UTR) was generated using the QuikChange Site-Directed Mutagenesis kit (Stratagene; Agilent Technologies, Inc.) according to the manufacturer's instructions. RAW264.7 cells were co-transfected with miR-93 mimics, miR-93 inhibitor and the luciferase reporter plasmids using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). At 48 h post-transfection, luciferase activities were detected using the dual luciferase reporter kit (Beyotime Institute of Biotechnology) with Renilla luciferase activity as an internal control.

Western blot analysis was performed as previously described (32,33). Total protein was extracted using RIPA buffer (Beyotime Institute of Biotechnology). The extraction and isolation of nuclear and cytoplasmic proteins were performed using the Nuclear and Cytoplasmic Protein Extraction kit (Beyotime Institute of Biotechnology). The concentrations of total cellular protein were determined using a BCA assay kit (Pierce; Thermo Fisher Scientific, Inc.). Briefly, 40 µg extracted protein samples were transferred onto polyvinylidene difluoride (PVDF, Millipore) membranes and then blocked with 5% skim milk at 4°C overnight. Each membrane was then probed with primary antibodies against TLR4 (1:2,000, ab9870), nuclear p-p65 (1:1,500, ab53489), p65 (1:2,000, ab86299), p-IκBα (1:2,000, ab133462), IκBα (1:2,000, ab32518), MyD88 (1:1,000, ab199247), Histone 3 (1:1,000, ab1791) and β-actin (1:1,000, ab8226) (all from Abcam) at 4°C overnight, followed by HRP-conjugated goat anti-rabbit IgG (1:10,000; cat. no. ab205718; Abcam). The protein bands were developed using an ECL kit (GE Healthcare) and blot bands were quantified using ImageJ software version 1.46 (Rawak Software, Inc.).

GraphPad Prism 5.0 (GraphPad Software, Inc.) was used to perform the statistical analyses. Each experiment was performed at least 3 times. Numerical data are presented as the means ± SD. Statistical significance between 2 groups was determined using a Student's t-test. Comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) with Tukey's post hoc test. The Kaplan-Meier method was performed to calculate the survival rates, and multiple comparisons were made using the Bonferroni method. P<0.05 was considered to indicate a statistically significant difference.

As is known, the mouse model of LPS-induced ALI is widely used to simulate the pathological conditions of lung injury in vivo (34). Following the injection of mice with LPS (2 mg/kg) for 24 h, lung histopathological changes were evaluated. As shown in Fig. 1A, the lungs of the mice in the ALI group displayed significant pathological changes, including obvious inflammatory cell infiltration and widespread alveolar wall thickness, compared with that in the control group. Moreover, lung injury was scored using a semi-quantitative histopathological score system, and the lung injury score was significantly increased in the ALI group, compared with that in the control group. The above-mentioned data indicated that the mouse model of ALI was successfully established. Subsequently, the levels of miR-93 in the lung tissues and in BALF from the mice with ALI were examined by RT-qPCR. It was observed that the expression levels of miR-93 were significantly downregulated in the lung tissues and in BALF, compared with those in the control group (Fig. 1B and C). These data suggest that miR-93 is involved in the pathogenesis of ALI.

To further investigate the roles of miR-93 in LPS-induced ALI, the mice were intravenously injected with agomir-93 or agomir NC prior to LPS administration. As shown in Fig. 2A, miR-93 expression in the lung tissues from mice with ALI was notably increased compared with that in the agomir NC group. Subsequently, H&E staining was carried out to detect the inflammatory status in the lungs. The results revealed that the agomir-93 injection markedly attenuated the histopathological lesions, compared with those in the ALI group (Fig. 2B). Lung micro-vascular permeability was assessed by EB extravasation. As shown in Fig. 2C, pulmonary permeability was significantly increased in the ALI group, compared with that in the control group, whereas the increased pulmonary permeability was markedly alleviated after the agomir-93 injection. The lung W/D ratio was evaluated to indicate pulmonary edema. It was found that the lung W/D ratio was significantly increased in the ALI group, compared with that in the control group, whereas the increased lung W/D ratio was markedly decreased after the agomir-93 injection (Fig. 2D). In addition, only 40% of the mice administered LPS survived up to 96 h, whereas 80% of the mice that were injected with agomiR-93 survived following exposure to LPS (Fig. 2E). Collectively, these results indicate that agomir-93 injection was able to improve LPS induced lung injury.

Currently, ALI is recognized as an unbalanced inflammatory response, and the inflammatory process is regulated by multiple miRNAs (35-37). In the present study, to further investigate whether miR-93 affects the inflammatory response induced by LPS in mice, ELISA was performed to examine the levels of pro-inflammatory cytokines, including IL-6, IL-1β, and TNF-α in BALF from mice with ALI. As shown in Fig. 3, exposure to LPS resulted in a significant increase in the production of IL-6, IL-1β and TNF-α. By contrast, these increased levels of pro-inflammatory cytokines were markedly attenuated by the agomir-93 injection. These data thus suggest that the agomir-93 injection alleviated lung injury by suppressing the inflammatory response in mice with ALI.

According to previous research, TLR4 is a common receptor of LPS, and plays important roles in the inflammatory response in ALI (38-42). Through bioinformatics prediction using TargetScan 7.0 and miRanda, the present study found that miR-93 directly targeted the TLR4 gene and that the target sequences were highly conserved among species (Fig. 4A and B). Notably, a previous study demonstrated that TLR4 was a direct target of miR-93 in chondrocytes (20). However, the association between miR-93 and TLR4 in ALI has not yet been clarified. To validate the possibility that TLR4 is directly targeted by miR-93, a luciferase reporter assay was performed. As shown in Fig. 4C, the expression levels of miR-93 were significantly increased following transfection with miR-93 mimics, while they were decreased by transfection with miR-93 inhibitor in the RAW264.7 cells, demonstrating the sufficient transfection efficacy. It was observed that transfection with miR-93 mimics markedly inhibited the luciferase activity in the TLR4-3'UTR wt reporter, and that the miR-93 inhibitor induced an increased luciferase activity; however, no changes were observed in the cells transfected with the TLR4 3'-UTR-mut plasmid (Fig. 4D), indicating a targeting interaction between miR-93 and TLR4. In addition, western blot analysis demonstrated that TLR4 protein expression was decreased by transfection with miR-93 mimics, whereas it was increased by transfection with miR-93 inhibitor in the RAW264.7 cells (Fig. 4E). The effect of agomir-93 on the protein expression of TLR4 protein in vivo was also assessed by western blot analysis. It was found that exposure to LPS significantly increased the protein expression of TLR4 compared with the control group, whereas this promoting effect of LPS on TLR4 expression was attenuated by agomir-93 (Fig. 4F). These results suggest that miR-93 targets TLR4 and suppresses its expression induced by LPS in vitro and in vivo.

Previous studies have reported that the NF-κB pathway, acting as the downstream signaling effector of TLR4, plays a crucial role in LPS-induced ALI (43,44). Thus, the present study sought to determine whether miR-93 influences the TLR4/MyD88/NF-κB pathway in a mouse model of ALI. The expression levels of key proteins of this pathway, including MyD88, IκB-α, p-IκB-α, p65 and nuclear p-p65 were measured by western blot analysis. As shown in Fig. 5A and B, exposure to LPS significantly increased the expression levels of MyD88, p-IκB-α and nuclear p-p65, compared with those in the control group. However, the promoting effects of LPS on the expression levels of these proteins were markedly attenuated by agomir-93 injection. Similar results were observed for the expression of p-p65, as determined by IHC staining (Fig. 5C). Collectively, these results suggest that miR-93 inhibited the TLR4-mediated activation of the NF-κB pathway, and thus suppressed the inflammatory response in LPS treated mice (Fig. 6).