The present study was approved by the Ethics Committee of Zigong First People's Hospital. A total of 18 patients (including 8 females and 10 males, ages 55–75) with ARDS caused by sepsis were enrolled between June 2010 and October 2013. All peripheral blood samples were collected with written informed consent. The clinical characteristics of ARDS were summarized and disease severity was determined using the Acute Physiology and Chronic Health Evaluation (APACHE) II, the Murray Lung Injury Score (LIS) and the Simplified Acute Physiology Score (SAPS) II (15). The characteristics of patients with ARDS were as follows: i) PaO₂/FiO₂ ≤300; ii) presence of bilateral pulmonary infiltrates on frontal chest radiograph; iii) no clinical evidence of left atrial hypertension; iv) requirement for positive pressure ventilation via an endotracheal tube; v) composite of oxygenation, compliance, positive end expiratory pressure and the appearance of chest radiograph (16). Blood was collected using an indwelling arterial catheter at the time of ICU admission (baseline) into sterilized, silicone-coated glass tubes, and at 3 and 7 days subsequently. Healthy blood donors were used as controls (also including 8 females and 10 males, ages from 55–75). Serum samples (4 ml) were obtained by centrifugation at 2,500 × g for 10 min at room temperature and frozen at −70°C until use. The buffy coat cell layer was carefully aspirated and buffy coat cells were suspended. Concentrations of TGF-β1 and PPAR-α were determined using commercial, standardized enzyme-linked immunosorbent assay (ELISA) kits. The TGF-β1 ELISA kit (cat. no. ADI-900-155) was from Enzo Life Sciences, Inc. (Farmingdale, NY, USA) and the PPAR-α ELISA kit (cat. no. 40,196) was from Active Motif, Inc. (Carlsbad, CA, USA). The lower detection limit for TGF-β1 was 0.10 pg/ml and for PPAR-α was 0.30 pg/ml.

Human lung fibroblast (HLFs; IMR-90 cells) and Mus musculus monocyte/macrophage RAW 264.7 cells were purchased from American Type Culture Collection (Manassas, VA, USA) and cultured in Eagle's Minimum Essential Medium with 10% fetal bovine serum (FBS; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The RAW 264.7 macrophages were cultured in Dulbecco's modified Eagle's medium containing 10% FBS. The cells were cultured in a humidified incubator at 37°C in 5% CO₂.

LPS (isolated from Escherichia coli) and TGF-β1 (used at a dose of 10 µg/ml) were purchased from Sigma-Aldrich; Merck Millipore (Darmstadt, Germany). Mouse monoclonal antibodies against TGF-β1 were from Abcam (cat. no. PB190503; Cambridge, MA, USA) and rabbit monoclonal antibody against PPARα was from Cell Signaling Technology, Inc. (cat. no. 2443S; Beverly, MA, USA). HRP-conjugated anti-mouse and anti-rabbit secondary antibodies were from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Specific small interfering short hairpin (sh)RNA targeting TGF-β1 and PPAR-α and the negative control shRNA were all from Sigma-Aldrich; Merck Millipore. Following allowing cellular attachment to the plates, 2×10⁵ cells were treated with relative shRNAs using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) RNA and protein were collected 72 h following treatment.

All experiments involving mice were performed in strict accordance with Animal Care and Use guidelines from Beijing Charles River Laboratory Animal Centre institutional committee (Beijing, China). A total of 30 (15 male and 15 female) mice (8–10 weeks old; 19–22 g) were purchased from Beijing Charles River Laboratory Animal Centre, and kept under standard conditions at room temperature (24°C) with a 12 h day/night cycle under specific pathogen-free conditions. At 3 h following intratracheal instillation of LPS (4 mg/kg) (17), the mice were administered with an intravenous injection in the tail vein of 10% chloral hydrate (3.5 ml/kg) for anesthetization at 4, 12 and 24 h. The mice were divided into the following groups: Vector group, injected with the same volume of pyrogen-free PBS; LPS group, stimulated with LPS; PPARα group, pre-treated with PPAPα prior to stimulation with LPS. Following the successful induction of ALI, the mice were administered with a tail vein injection of adeno-associated viruses (AAV) carrying PPAR-α short hairpin RNA (shPPAR-α group) or with a scrambled (SCR) sequence as a control. Primary murine alveolar epithelial cells were isolated, as described previously (18).

Total RNA was extracted from the cultured cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). For analysis of the expression of messenger RNA (mRNA), cDNA synthesis was performed by reserve transcription using the Transcriptor First Stand cDNA synthesis kit (Roche Diagnostics, Basel, Switzerland). The qPCR was performed in duplicate with a QuantiTect SYBR Green PCR kit (Qiagen, Inc., Valencia, CA, USA) on an Applied Biosystems 7500 Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.), according to manufacturer's protocol. The primers were as follows: TGF-β1, forward 5′-CCACCTGCAAGACCATCGAC-3′ and reverse: 5′-CTGGCGAGCCTTAGTTTGGAC-3; PPARγ, forward 5′-GAGATCATCTACACGATGCTGGC-3′ and reverse 5′-CGCAGGCTTTTGAGGAACTC-3′; and GAPDH, forward 5′-GAGAAGTATGACAACAGCCTC-3′ and reverse 5′-ATGGACTGTGGTCATGAGTC-3′. All primers were purchased from Thermo Fisher Scientific, Inc. The PCR amplification was performed at 95°C for 1 min, followed by 35 cycles of 95°C for 15 sec, 60°C for 15 sec, and 72°C for 30 sec. The expression levels of genes were normalized against that of GAPDH and relative fold changes in mRNA expression were calculated using the formula 2^−ΔΔCq (19).

Whole cell lysates were collected and protein from the cultured cells was extracted using RIPA lysis buffer on ice, following centrifugation at 4°C, 10,000 × g for 15 min, the supernatants were collected. The BCA protein assay kit (Thermo, USA) was used to determine the protein concentration, and 30 µg from each sample was mixed with 4X SDS loading buffer and heated at 100°C for 10 min. Protein samples were then separated by 10% SDS-polyacrylamide gel electrophoresis and transferred onto a PVDF membrane (EMD Millipore, Billerica, MA, USA). A mixture of 5% non-fat milk in Tris-buffered saline Tween-20 (TBST) was used to block the nonspecific proteins for 1 h at room temperature. The membrane blots were then probed with primary antibodies at 4°C overnight, as follows: PPAR-α (1:500), TGF-β1 (1:1,000), TNF-α (1:1,000), and β-actin (1:2,000). The membranes were washed five times with TBST for 5 min, followed by incubation with horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology, Inc.) for 1 h at room temperature. An enhanced chemiluminescent system (GE Healthcare Life Sciences, NJ, USA) was used to visualize the protein antigen. The signals were recorded using X-ray film (Kodak, Rochester, NY, USA). Images were captured, representative of three repeats.

The levels of PPAR-α and TGF-β1 in the bronchoalveolar lavage fluid (BALF), serum and cell culture supernatants were determined using sandwich ELISA, according to the manufacturer's protocol.

All statistical analyses were performed using SPSS 18.0 statistical software (SPSS, Inc., Chicago, IL, USA). Data were analyzed by comparing the mean ± standard deviation from three experiments using Student's t-test P<0.05 was considered to indicate a statistically significant difference. One-way analysis of variance with a Bonferoni correction was used for statistical analysis, followed by a Fisher's exact test, as necessary.

In order to investigate whether PPAR-α was involved in TGF-β signaling, the present study treated HLFs with PPAR-α and, 48 h following treatment, the cells were harvested. The mRNA and nuclear proteins were isolated, followed by the examination of TGF-β1 using RT-qPCR and western blot analyses. It was found that PPAR-α inhibited the expression of TGF-β1 in HLF cells at the mRNA level (Fig. 1A) and protein level (Fig. 1B); the suppression of TNF-α by PPAR-α was used as a positive control. To further confirm these findings, RAW 264.7 cells were used and a similar experiment was performed. As shown in Fig. 1C and D, the expression of TGF-β1 was suppressed by the enforced expression of PPAR-α.

To further investigate the roles of PPAR-α and TGF-β1 in ALI, LPS-induced ALI mice were used, and the expression levels of PPAR-α and TGF-β1 were examined using RT-qPCR and western blot analyses. Following LPS treatment, in the murine lung, a decrease in the mRNA expression of PPAR-α and an increase in the mRNA expression of TGF-β1 were observed using RT-qPCR analysis (Fig. 2A). Similarly, the western blot analysis showed that the protein expression PPAR-α was decreased and that of TGF-β1 was significantly increased by treatment with LPS (Fig. 2B). PPAR-α treatment prior to stimulation with LPS resulted in a decrease in the mRNA expression levels of TGF-β1 in the BALF, peripheral blood and splenocytes (Fig. 2C). Consistently, the present study found that the expression level of PPAR-α in the BALF was significantly elevated by treatment with the PPAR-α expression vector (Fig. 2D, left panel). The expression levels of PPAR-α in the peripheral blood and splenocytes were also upregulated (Fig. 2D, middle and right panels). These findings demonstrated that the enforced expression of PPAR-α downregulated the expression of TGF-β1 in LPS-induced ALI mice.

The question of whether PPAR-α was involved in the recovery of lung function following ALI remained; therefore, to evaluate the role of PPAR-α in the recovery of lung function, the present study developed a virus carrying the shPPAR-α or SCR sequence as a control. It was found that the RAW 264.7 cells, which were transduced with shPPAR-α, exhibited significantly decreased levels of PPAR-α (Fig. 3A). Following injecting of the virus expressing either shPPAR-α or SCR into the mouse following LPS treatment, the effects of PPAR-α on lung recovery were examined. Significantly decreased levels of PPAR-α in the mouse lung were detected using RT-qPCR analysis (Fig. 3B) and western blot analysis (Fig. 3C). Impaired lung function has been considered to cause a dose-dependent increase in lung resistance index (RI) and decrease in dynamic compliance (Cdyn) in response to a cholinergic stimulus (methacholine) and a significant increase in BALF protein (11). The analysis of pressure and flow waveforms indicated that the knockdown of PPAR-α resulted in a marked increase in RI (Fig. 3D), a notable decrease in Cdyn (Fig. 3E) and a significant increase in BALF protein (Fig. 3F). These data suggested that PPAR-α was essential for the recovery of lung function following ALI.

In order to further evaluate the association between TGF-β with PPAR-α in the recovery of lung function following ALI, TGF-β1 was inhibited in the mouse lung following LPS treatment by injecting AAV expressing shTGF-β1 (or SCR as a control). TGF-β1-knockdown efficiency was first confirmed in the mouse lung. It was found that the application of these viruses significantly decreased the levels of TGF-β1 (Fig. 4A), resulting in a significant decrease in RI (Fig. 4B), a significant increase in Cdyn (Fig. 4C) and a significant decrease in BALF protein (Fig. 4D). These data suggested that shTGF-β1 was essential for the recovery of lung function following ALI. The results showed that the knockdown of PPAR-α and knockdown of TGF-β1 in the mouse lung partially reduced the shPPAR-α-impaired recovery of lung function, as detected by RI (Fig. 4E), Cdyn (Fig. 4F) and BALF protein (Fig. 4G). Together, these experiments suggested that PPAR-α was essential for the recovery of lung function following ALI, possibly through the suppression of TGF-β.

To further investigate the clinical significance of the above-mentioned findings, the present study detected the serological expression levels of PPAR-α and TGF-β1, which revealed that PPAR-α was inversely correlated with TGF-β1 in patients with ARDS (Fig. 5A; P<0.05). In addition, the correlation with disease activity in patients with ARDS was analyzed. It was found that serum PPAR-α was positively associated with the ratio of PaO₂/FiO₂, whereas the serum level TGF-β1 was negatively associated with PaO₂/FiO₂ (Fig. 5B; P<0.05).

Using the LIS (Fig. 5C; P<0.05) and SAPS II (Fig. 5D; P<0.05) scoring methods to determine disease activity, it was revealed that the serum expression level of PPAR-α was inversely correlated with these indices, whereas the serum expression level of TGF-β1 was consistent with these indices. No significant correlation was observed between the APACHE II scoring method and PPAR-α or TGF-β1 serum levels (Fig. 5E; P>0.05). On the whole, these results suggested that PPAR-α was negatively correlated with TGF-β1 and involved in the pathogenesis of ARDS.