Melatonin receptor agonist protects against acute lung injury induced by ventilator through up-regulation of IL-10 production

Experiments were performed on 6- to 8-week-old male Sprague Dawley rats weighing 350 ± 20 g. The animal care and experimental protocol were approved by the Animal Review Committee at the National Defense Medical Center in Taipei, Taiwan.

The rats were anesthetized with an intraperitoneal pentobarbital (80 mg/kg) injection. Catheters were inserted into the femoral artery and vein, and a tracheostomy was performed. Anesthesia was maintained with intravenous pentobarbital at 18 mg/hr. The femoral artery catheter was inserted to collect biological signals to monitor the mean arterial pressure and heart rate continuously. The rats were placed in a supine position and connected to a ventilator (Model 7025, Ugo Basile, Varese, Italy). Then, the rats were ventilated with volume control for 5 h with an inspired oxygen fraction of 0.21, positive end-expiratory pressure of 2 cmH₂O, and respiratory rate of 70 breaths/min. The rats were randomly assigned to receive mechanical ventilation with either a tidal volume of 20 mL/kg (high tidal volume) or 10 mL/kg (low tidal volume). A positive end-expiratory pressure of 2 cmH₂O was used in both cases [13].

The rats were randomly allocated into 10 groups (n = 6 per group): group 1 (control with low tidal volume ventilation, 10 mL/kg); group 2 (control+ 20 mg/kg melatonin); group 3 (control+ 5 mg/kg ramelteon); group 4 (control+ 2 mg/kg luzindole); group 5 (VILI with high tidal volume ventilation, 20 mL/kg); group 6 (VILI + 2 mg/kg luzindole); group 7 (VILI + 20 mg/kg melatonin); group 8 (VILI + 20 mg/kg melatonin + 2 mg/kg luzindole); group 9 (VILI + 5 mg/kg ramelteon); and group 10 (VILI + 5 mg/kg ramelteon + 2 mg/kg luzindole). We performed another animal study to evaluate the role of IL-10 in melatonin- and ramelteon-mediated protection against VILI. The rats were randomly allocated into 8 groups: group 1 (control); group 2 (control + 10 μg anti-IL-10 antibody); group 3 (VILI); group 4 (VILI + 10 μg anti-IL-10 antibody); group 5 (VILI + 20 mg/kg melatonin); group 6 (VILI + 10 μg anti-IL-10 antibody + 20 mg/kg melatonin); group 7 (VILI + 5 mg/kg ramelteon); and group 8 (VILI + 10 μg anti-IL-10 antibody + 5 mg/kg ramelteon). Melatonin, ramelteon, luzindole, or anti-IL-10 antibody treatment was administered via femoral vein injection 30 min before VILI. Ramelteon (LKT laboratories Inc., USA) and melatonin (Sigma-Aldrich, USA) were dissolved in absolute ethanol. The resulting solution was diluted with saline until a concentration of 0.3% ethanol in saline was obtained. Luzindole (Tocris, England) was dissolved in dmethyl sulfoxide (DMSO). The resulting solution was diluted with saline until a concentration of 0.15% DMSO in saline was obtained. The concentrations of ethanol and DMSO in control group did not cause lung injury (supplement Figure 1). The anti-IL-10 antibody (Abcam, Cambridge, MA, USA) was dissolved in 0.9% saline. The doses of various agents in this study were chosen according to previous studies [14‐17].

Blood (0.5 mL) was drawn from the right femoral artery at the beginning and at hourly intervals in each experiment. Arterial blood gas levels were immediately measured with a blood gas analyzer (ABL 800 FLEX blood gas analyzer, Radiometer, Denmark).

At the end of the experiments, the right lung was removed from the hilar region, and the wet weight was obtained for the calculation of the LW/BW ratio. The middle lobe of the right lung was excised, and the wet weight was recorded. The lobe was then placed in an incubator at 60 °C for 48 h to obtain the dry weight. The W/D weight ratios were calculated by dividing the wet weight by the dry weight.

The lung tissues were fixed, sectioned, stained with hematoxylin and eosin, and viewed under a light microscope. The scoring was performed by two pathologists who were blinded to the experimental conditions. A minimum of 10 random areas were examined at a magnification of × 400. Each histological characteristic was graded according to the following four-point scale: 0, no damage; 1, mild damage; 2, moderate damage; 3, severe damage according to the following morphological characteristics: (1) infiltration or aggregation of neutrophils, and (2) the thickness of the alveolar walls [18, 19]. The resulting scores were summed to represent the lung injury score. The numbers of polymorphonuclear neutrophils and lung injury score in the lung tissue were also analyzed.

The oxidative damage to proteins was assessed by determining carbonyl groups based on a reaction with dinitrophenylhydrazine, as previously described [18]. The carbonyl content was determined from the absorbance at 370 nm while assuming a molar absorption coefficient of 220,000 M^− 1. The result was expressed as the concentration of carbonyl derivatives in the protein (nmol carbonyl/mg protein).

Immunohistochemical staining was performed to identify myeloperoxidase (MPO) in the lung tissue. Paraffin-embedded lung tissue sections fixed with formalin were deparaffinized and pretreated for antigen retrieval. Endogenous peroxidase was blocked by incubating in 3% H₂O₂ in methanol for 15 min. Immunostaining of the lung sections was performed using an anti-MPO-specific antibody (1:1000; Cell Signaling Technology, Danvers, MA, USA). After being washed, the slides were sequentially incubated with rat tissue-specific a horseradish peroxidase-conjugated anti-rabbit antibody (Nichirei Corporation Tokyo, Japan) for 30 min, and the lung tissue sections were counterstained with hematoxylin [18, 19].

The levels of protein,tumor necrosis factor-α (TNF-α), IL-1β, IL-6, CXCL-1, and IL-10 in BALF were measured after the experiment. The protein concentration in the supernatant was determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL, USA). The levels of TNF-α, IL-1β, IL-6, CXCL-1, and IL-10 were determined using an ELISA kit (R&D Systems Inc., Minneapolis, MN, USA) [18, 19].

Immunoblotting was performed as described previously [18]. Immunoblotting was carried out with antibodies against STAT3, B-cell lymphoma (Bcl)-2, poly (ADP-ribose) polymerase (PARP), caspase-3, NF-κB p65, inhibitor of NF-κB (IκB)-α (Cell Signaling Technology, USA), iNOS (BD Biosciences, USA), heat shock protein 70 (Hsp70; Santa Cruz Biotechnology, USA), and β-actin (for cytoplasmic proteins, diluted 1:10,000; Sigma-Aldrich, USA).

The PaO₂ concentration at 5 h after VILI was significantly higher in the VILI + melatonin and VILI + ramelteon groups than the VILI group. However, the increased oxygenation concentration in the VILI + melatonin and VILI + ramelteon groups was reduced by the administration of luzindole (Table 1).

As shown in Fig. 1a-c, the LW/BW and W/D weight ratios and the protein levels in BALF were significantly increased in the VILI group compared to the control group. Moreover, melatonin or ramelteon treatment significantly reduced these increases. However, these effects were abolished by the administration of luzindole.

In comparison to the control group, the VILI group had significantly increased serum MDA levels, carbonyl content in the lung tissue, and numbers of MPO-positive cells in the lung tissue. Treatment with melatonin or ramelteon significantly diminished these increases. In contrast, the reduction in serum MDA levels, carbonyl content, and numbers of MPO-positive cells in the lung tissue by melatonin or ramelteon was abolished by treatment with luzindole (Fig. 2a-c).

As shown in Fig. 3a-d, the BALF levels of TNF-α, IL-1β, IL-6, and CXCL-1 were significantly increased in the VILI group compared with the control group. Melatonin or ramelteon treatment significantly inhibited the VILI-induced increases in TNF-α, IL-1β, IL-6, and CXCL-1 levels in BALF. However, the protective effect of melatonin or ramelteon was abolished by treatment with luzindole.

Compared with the control group, the VILI group had significantly lower IL-10 levels in BALF and greater STAT3 phosphorylation levels in the lung tissue. However, the decreased IL-10 level in the VILI group was significantly elevated by melatonin or ramelteon treatment. The increased STAT3 phosphorylation in the VILI group was significantly mitigated by melatonin or ramelteon treatment. Furthermore, the administration of luzindole blocked the effects of melatonin and ramelteon (Fig. 4a-b).

The iNOS level of the lung tissue was significantly increased after VILI induction, but the increase was significantly suppressed by melatonin or ramelteon treatment. In contrast, treatment with luzindole counteracted the protective effect of melatonin and ramelteon (Fig. 5).

As shown in Fig. 6a, the control group exhibited normal lung tissue structures. In contrast, severe lung damage was observed in the VILI group, as indicated by extensive interstitial edema and leucocyte infiltration. Melatonin or ramelteon treatment significantly reduced the histological changes (Fig. 6a), neutrophil infiltration (Fig. 6b), and lung injury scores (Fig. 6c) in the VILI group. However, the protective effects were abolished by luzindole treatment.

The protein level of NF-κB p65 in the nucleolus was significantly increased, but the protein level of IκB-α in the cytoplasm was significantly decreased in the VILI group compared with the control group (Fig. 7a-b). Melatonin or ramelteon treatment restored the suppressed IκB-α expression and reduced nuclear NF-κB p65 expression. Treatment with luzindole counteracted the protective effect of melatonin and ramelteon (Fig. 7a-b).

The Hsp70 (Fig. 8a) and Bcl-2 (Fig. 8b) protein levels in the lung tissue were substantially lower in the VILI group than the control group. However, they were significantly increased upon melatonin or ramelteon treatment. The protective effect was abolished by treatment with luzindole. The expression of caspase-3 and cleaved PARP in the lung tissue was higher in the VILI group than the control group, but melatonin or ramelteon treatment attenuated the increased expression of caspase-3 and cleaved PARP in the VILI group. Similarly, the protective effects of melatonin and ramelteon were abolished by treatment with luzindole (Fig. 8c-d).

Compared with the control group, the control + anti-IL-10 antibody group had significantly lower IL-10 level in BALF. Moreover, VILI group had significantly lower IL-10 level in BALF. The decreased BALF IL-10 level in the VILI group was significantly elevated by melatonin or ramelteon treatment. However, the administration of anti-IL-10 antibody blocked the effects of melatonin and ramelteon (Fig. 9).

Compared with the control group, the VILI group had significantly increased LW/BW and W/D weight ratios and protein levels in BALF. Melatonin or ramelteon treatment significantly reduced these increases. However, anti-IL-10 antibody administration blocked the protective effect of melatonin and ramelteon against VILI induced lung edema (Fig. 10a-c).

As shown in (Fig. 11a-c), melatonin or ramelteon treatment significantly inhibited the VILI-induced increases in TNF-α, IL-1β, and IL-6 levels in BALF. However, the protective effect of melatonin or ramelteon was abolished by treatment with anti-IL-10 antibody.

As shown in Fig. 12, the control group exhibited normal lung tissue structures. In contrast, severe lung damage was observed in the VILI group, as indicated by extensive interstitial edema and leucocyte infiltration. Melatonin or ramelteon treatment significantly reduced the histological changes (Fig. 12a), neutrophil infiltration (Fig. 12b), and lung injury scores (Fig. 12c) in the VILI group. However, the protective effects were abolished by anti-IL-10 antibody treatment.

The Bcl-2 protein levels (Fig. 13a) in the lung tissue were substantially lower in the VILI group than the control group. However, they were significantly increased upon melatonin or ramelteon treatment. The protective effect was abolished by treatment with anti-IL-10 antibody. The expression of cleaved PARP in the lung tissue was higher in the VILI group than the control group, but melatonin or ramelteon treatment attenuated the increased expression of cleaved PARP in the VILI group. Similarly, the protective effects of melatonin and ramelteon were abolished by treatment with anti-IL-10 antibody (Fig. 13b). The protein level of NF-κB p65 in the nucleolus was significantly increased in the VILI group compared with the control group. Melatonin or ramelteon treatment reduced nuclear NF-κB p65 protein expression. Treatment with anti-IL-10 antibody counteracted the protective effect of melatonin and ramelteon (Fig. 13c).