Classically Activated Macrophages Protect against Lipopolysaccharide-induced Acute Lung Injury by Expressing Amphiregulin in Mice

ACUTE lung injury (ALI) is a serious respiratory condition with high mortality and is ascribed to the disrupted epithelial and endothelial barrier and infiltration of neutrophils and protein-rich fluid into the alveolar space.¹  Although neutrophils within the lung play an important part in the pathogenesis of ALI,^2,3  mounting evidence indicates that macrophages also contribute to the modulation of inflammatory responses, including the initiation and resolution of these responses.⁴  On the basis of the patterns of gene and protein expressions, in combination with the different roles they play in host defense, macrophages fall into two categories, i.e., classically activated (M1) and alternatively activated (M2) macrophages. Macrophages, when stimulated with lipopolysaccharide, can be activated into M1 phenotype, which is characterized by high level of proinflammatory cytokines and reactive oxygen species, thereby result in damage of adjacent tissues.⁵  Conversely, M2 macrophages, induced by interleukin-4 and/or interleukin-13, release high-level argninase-1 (Arg-1), Fizz1, and Ym1, thus promoting inflammation resolution and tissue repair.⁶  Previous studies exhibited that alveolar macrophages (AMs) are activated into M1 phenotype at the early phase of ALI,^4,7  and depletion of macrophages from lung tissues^8,9  or inhibition of M1 products, such as nitric oxide and interleukin-1β,^9,10  protect against pulmonary injury in some animal models. These findings suggest that M1 macrophages may be one of the initiators of the acute phase of ALI.

Amphiregulin, a ligand of epithelial growth factor receptor (EGFR), was reported to be up-regulated in a ventilator-associated ALI model.¹¹  Amphiregulin expression was also found to be increased in damaged lung tissues in patients with chronic obstructive pulmonary disease and asthma.^12–15  Even smoking and urban fine particulate materials can increase amphiregulin expression in the lung cells.^16–18  Interestingly, amphiregulin was shown to promote the proliferation of epithelial cells and modulate the gene expression of proinflammatory cytokines.^18,19  But little is known about the expression pattern and role of amphiregulin in lipopolysaccharide-induced ALI.

Given that lung epithelial damage plays an important role in the pathogenesis of ALI and that lipopolysaccharide can stimulate Kupffer cells (liver macrophages) secreting amphiregulin,²⁰  we were led to hypothesize that amphiregulin may be produced by AMs when challenged with lipopolysaccharide, thereby improving epithelial barrier function and further attenuating lipopolysaccharide-induced ALI. To test this hypothesis, in this study, we determined the level of amphiregulin in classically activated AMs in vitro and in lung tissues of lipopolysaccharide-induced ALI in vivo and observed protein exudation, infiltration of inflammatory cells, and other indicators of lung functions, with an attempt to understand the protecting role of amphiregulin in lipopolysaccharide-induced ALI.

Male C57BL/6 mice of 6 to 8 weeks old, weighing 20 to 25 g, were obtained from Wuhan University Laboratory Animal Center and were maintained under specific pathogen free conditions at a room temperature of 23° ± 1°C (50 ± 10% relative humidity) and on a 12:12-h light/dark cycle in the Experimental Animal Center, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China. All experimental protocols were carried out in strict accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23, revised 1996) and were approved by the Committee of Experimental Animals of Huazhong University of Science and Technology.

Mice were tagged and randomly allocated to each group before any treatment. Mice were anesthetized intraperitoneally with sodium pentobarbital (80 mg/kg; Sigma, USA) and then orotracheally intubated. To induce ALI, mice were instilled intratracheally with Escherichia coli serotype lipopolysaccharide 055:B5 (2 ml/kg; Sigma), and a bolus of 3 × 0.6 ml air was delivered to ensure that all instilled solutions reached the distal lung. Mice in the control group were treated intratracheally with phosphate-buffered saline (PBS; 2 ml/kg, Thermo Scientific, USA). To investigate the role of endogenous amphiregulin, mice were injected intraperitoneally with amphiregulin-blocking antibody (amphiregulin antibodies; 50 μg per mouse; R&D Systems, USA) 30 min before instillation of 1 mg/kg lipopolysaccharide. To determine the effect of exogenous amphiregulin on lipopolysaccharide-induced ALI, mice were administrated intraperitoneally with recombinant mouse amphiregulin (rmAreg; 5 μg per mouse; R&D Systems), and 30 min later, they were challenged with 3 mg/kg lipopolysaccharide as we previously reported.²¹  Mice were killed 6 and 24 h after lipopolysaccharide or PBS challenge. To examine amphiregulin expression in this model, mice were intratracheally instilled with lipopolysaccharide (3 mg/kg) and properly killed at 0, 3, 6, 12, and 24 h.

The left lungs were fixed in 4% paraformaldehyde and embedded in paraffin. Lung sections (5 μm) were stained with hematoxylin and eosin (HE) and examined under a light microscope. Lung injury was evaluated and scored by an investigator blinded to experimental grouping against a recently released criterion, and the lung damage was assessed on a two-point scale, ranging from 0 to 1.²² 

To collect bronchial alveolar lavage fluid (BALF), 1 ml PBS was instilled into mouse lung and slowly withdrawn. The procedure was repeated three times for each mouse. Cells were collected after centrifugation at 400g for 10 min and counted on a hemocytometer. For differential cell counting, cells were stained with phycoerythrin-conjugated anti-mouse Gr-1 (eBioscience, USA) for 45 min and flow cytometrically analyzed by using a Fluorescence Activated Cell Sorter machine (BD Biosciences, USA). Neutrophils were calculated as the total cell number in BALF multiplied by the percentage of Gr-1⁺F4/80⁻ cells. The supernatants of centrifuged BALF were used to measure the protein concentration by using a Pierce bicinchonininc acid protein assay kit (Thermo Scientific) or stored at −80°C for later cytokine and IgM analysis.

To differentiate cells in BALF, cells in BALF were first removed by centrifugation (400g, 4°C, 6 min), counted on a hemocytometer, stained with Giemsa fluids after centrifugation at 400g for 6 min and smeared onto a glass slide.

To distinguish the attached cells and unattached cells in BALF of mice challenged with lipopolysaccharide, cells in BALF were collected and cultured in a 24-well plate. Forty minutes later, the unattached cells in the supernatant were smeared onto a glass slide after centrifugation for 6 min at 400g. Both attached and unattached cells were stained with Giemsa fluid for morphologic identification (n = 5 for each group).

Ten milligrams of lung tissues from mice in different groups were harvested, and total RNA was extracted by using TRIzol (Invitrogen, USA) according to the manufacturer’s instructions. Five hundred nanograms of total RNA was isolated to generate cDNA in a reverse transcription reaction by using PrimeScript RT Master Mix (TaKaRa, China). Quantitative reverse transcription polymerase chain reaction (PCR) for amphiregulin (specific primers: forward 5′-AATGAGAACTCCGCTGCTAC-3′, reverse 5′-CCCCTGTGGAGAGTTCACTG-3′) and β-actin (forward 5′-TGTCCACCTTCCAGCAGATGT-3′, reverse 5′-AGCTCAGTAACAGTCCGCCTAGA-3′) was conducted using a QuantiFast SYBR Green PCR kit (Qiagen, Germany) on a StepOne Real-time PCR System (Applied Biosystems, USA). Relative gene expression was calculated by the 2^−ΔΔCT method, with β-actin serving as an internal control.

After deparaffinization, rehydration, and antigen retrieval, lung sections were blocked with 2% donkey serum for 2 h and then incubated with rabbit monoclonal anti-Iba1 (1:100, Wako, Japan) and goat polyclonal antiamphiregulin (10 μg/ml, R&D Systems) overnight at 4°C. The sections were washed and incubated with Dylight 488-conjugated donkey anti-rabbit (1:100; Abbkine, USA) and Dylight 549-conjugated donkey anti-goat (1:300, Abbkine) antibody for 1 h and then stained with dihydrochloride (DAPI) (1μg/ml) for 10 min for nuclei labeling. Sections were observed under an Olympus fluorescence microscope (Olympus, Germany) using the Cellsense software package (Olympus).

To evaluate which type of cells (neutrophils or macrophages) in the BALF secret amphiregulin, the cells in the supernatant centrifuged on a glass slide and the attached cells were stained with rabbit polyclonal antiamphiregulin (1:50, Santa Cruz Biotechnology, USA) overnight at 4°C and then incubated with Dylight 549-conjugated goat anti-rabbit secondary antibody (1:200, Abbkine) for 1 h, followed by stained with DAPI (1 μg/ml) for 10 min for nuclei labeling.

Alveolar macrophage was depleted by using liposome-encapsulated clodronate (Department of Molecular Cell Biology, Free University Medical Center, Amsterdam, the Netherlands).^9,23  Briefly, mice were instilled intratracheally with 100 μl liposome-encapsulated clodronate (5 mg/ml). Mice in control group were administrated PBS-containing liposomes. Forty-eight hours after liposome treatment, mice were stimulated intratracheally with lipopolysaccharide (3 mg/kg) for 6 h. Cell-free BALF was collected for determination of amphiregulin level.

Alveolar macrophages were obtained as described previously.²⁴  Briefly, cells from the lavage fluid for each mouse were pooled, centrifuged at 400g for 10 min at 4°C, and resuspended in Roswell Park Memorial Institute 1640 medium (Hyclone) containing 10% fetal bovine serum (FBS; Life Technologies, USA), 100 U/ml penicillin and 100 μg/ml streptomycin (Hyclone). After incubation at 37°C and in 5% CO₂ overnight, cells were washed with PBS to remove unattached cells. For M1 and M2 polarization, AMs were seeded into 24-well plates (2 × 10⁵ cells/well) and incubated for 24 h with lipopolysaccharide (1 μg/ml) and interleukin-4 (20 ng/ml; PeproTech, USA), respectively. AMs treated with PBS were designated as M0. The supernatants were collected for amphiregulin detection. To obtain the conditioned medium of M0 and M1, AMs were incubated with PBS and lipopolysaccharide (1 μg/ml), respectively, in a 6-well plate (1 × 10⁶ cells/well) for 24 h. The supernatants were collected and designated as M0-CM and M1-CM. M1-CM was pretreated with amphiregulin antibodies (3 μg/ml) or IgG (3 μg/ml) at 37°C for 3 h.

Peritoneal macrophages (PMs) were isolated as described elsewhere.²⁵  Briefly, the abdominal cavity of the animals was lavaged with 5 ml PBS supplemented with 5% FBS, and then the recovered buffer was centrifuged at 400g for 10 min. Cells were resuspended in Roswell Park Memorial Institute 1640 medium (Hyclone) supplemented with 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin (Hyclone) and were seeded into 24-well plates (2 × 10⁵ cells/well). Cells were incubated at 37°C and in 5% CO₂ overnight and then washed with PBS to remove unattached cells. PMs were treated with PBS or lipopolysaccharide (1 μg/ml) and interleukin-4 (20 ng/ml) separately for 4 or 24 h.

The murine lung epithelial cell line (MLE12 cells) was purchased from American Type Culture Collection (USA). MLE12 cells were cultured in Dulbecco’s Modified Eagle’s Medium:Nutrient Mixture F-12 (Dulbecco’s Modified Eagle’s Medium/F12, Hyclone) with 10% FBS at 37°C and in 5% CO₂. For western blotting, MLE12 cells were seeded into 6-well plates (1 × 10⁶ cells/well) and stimulated with rmAreg (50 ng/ml) at indicated time points or with M0-CM and M1-CM containing IgG or amphiregulin antibodies for 10 min.

Enzyme-linked immunosorbent assays were carried out using commercially available kits to measure concentrations of amphiregulin (R&D systems), IgM (Plymouth Meeting, USA), tumor necrosis factor (TNF)-α, and interleukin-6 (Neobioscience Technology, China) by following the manufacturer’s instructions.

MLE12 cells were seeded into 12-well plates at 2 × 10⁵ cells/well and allowed to grow to 70% confluence. Programmed cell death was induced by incubation of the cells with lipopolysaccharide (100 μg/ml) for 24 h. To examine the effect of amphiregulin on apoptosis, MLE12 cells were pretreated with indicated dosage of amphiregulin before lipopolysaccharide stimulation. To examine the effect of amphiregulin secreted by M1 AMs, MLE12 cells were preincubated with M1-CM containing IgG or amphiregulin antibodies for 30 min and then stimulated with lipopolysaccharide (100 μg/ml) for 24 h. The challenged MLE12 cells were harvested, stained with annexin-V-fluorescein isothiocyanate and PI (BD Biosciences), and analyzed by the Fluorescence Activated Cell Sorter flow cytometry (BD Biosciences). In this experiment, a nonstained control tube and a single color tube were used to control the gate according to the manufacturer’s constructions.

Total protein from lung tissues or cells was extracted by using a commercially available kit (KGP250; Nanjing Keygen Biotech Co. Ltd., China). Protein concentrations were determined by the bicinchonininc acid protein assay (Thermo Scientific). Equal amount of protein extract was separated on sodium dodecyl sulfate-polyacrylamide gels and then transferred onto a polyvinylidene difluoride membrane (Millipore, USA). The membranes were blocked with 5% nonfat milk in Tris-buffered saline and Tween 20 (0.1% Tween in Tris-buffered saline) for 1 h at room temperature and then incubated overnight at 4°C with rabbit monoclonal anti-EGFR (1:1,000; Abcam, UK), rabbit monoclonal anti-p-EGFR-Tyr 1068 (1:1,000; Cell Signaling Technology, USA), rabbit monoclonal anti-AKT (1:1,000, Cell Signaling Technology), rabbit monoclonal anti-p-AKT-Thr 308 (1:1000, Cell Signaling Technology) or rabbit polyclonal anti-β-actin (1:2,000; Santa Cruz Biotechnology), and then with horse radish peroxidase–conjugated goat anti-rabbit IgG antibody (1:5,000; Proteintech Group, USA). The immunoreactive bands were visualized and photographed by using an EC3 Imaging System (UVP Inc., USA).

Data were expressed as mean ± SD. Gaussian distribution of the data was assessed by using the d’Agostino and Pearson omnibus normality test, and the equalities of variances were examined by using the F test. Student’s t test was employed for pairwise comparisons. One-way ANOVA followed by Bonferroni post hoc comparison test was used for multiple comparisons. The lung injury scores were compared by utilizing nonparametric Mann-Whitney U test. The sample size was based on our previous research work.²¹  All statistical analyses were performed by using the GraphPad Prism software package (version 5 for Windows, GraphPad Software. Inc., USA), and statistical significance was accepted at two tailed P < 0.05. Four mice died after intratracheal intubation, and the data were lost.

In the mouse model of lipopolysaccharide-induced ALI, amphiregulin gene expression (fig. 1A) in lung tissues was significantly induced between 3 and 12 h after lipopolysaccharide challenge. The levels of amphiregulin protein in both lung tissues (fig. 1B) and BALF (fig. 1C) were significantly increased 3 h after intratracheal lipopolysaccharide challenge, reached a peak at 6 h, and declined thereafter, but staying at high level throughout the experiment.

To confirm whether macrophages express amphiregulin when stimulated with lipopolysaccharide, we first detected amphiregulin expression in macrophages by using immunofluorescent staining. In vivo results showed that only green immunofluorescence, an indicator of Iba-1 antibody, was observed, whereas no obvious amphiregulin protein expression was found in the control group (fig. 2A, 0 h). In contrast, immunofluorescence indicating amphiregulin protein expression was enhanced conspicuously in Iba⁺ cells at 3 h (fig. 2B) and stayed at a high level of 6 (fig. 2C) and 12 h (fig. 2D) after lipopolysaccharide stimulation.

To confirm that amphiregulin was expressed by macrophages rather than neutrophils in lipopolysaccharide-induced ALI, we, by using Giemsa staining, demonstrated that the attached cells in BALF were mainly macrophages (fig. 3A) and that the unattached cells in the supernatant were, for the most part, neutrophils (fig. 3B). Then we stained the attached cells and the unattached cells separately on glass slides with amphiregulin antibody and DAPI. Our result showed that the attached cells (macrophages) secreted amphiregulin (fig. 3C). Conversely, unattached cells (neutrophils) did not express amphiregulin (fig. 3D).

To further confirm that activated AMs produce amphiregulin, we determined amphiregulin expression in the mice depleted of AMs by intratracheal administration of liposome-encapsulated clodronate as reported previously.^9,23  We found that liposome-encapsulated clodronate challenge reduced the number of macrophages and increased the number of neutrophils in BALF irrespective of exposure to lipopolysaccharide (fig. 4, A–D). At the same time, AM depletion significantly lowered amphiregulin protein production in the BALF of lipopolysaccharide-challenged mice when compared with their liposome-PBS–treated counterpart (fig. 4E). But lipopolysaccharide challenge significantly increased the expression of amphiregulin mRNA compared with control group (fig. 4F).

To evaluate the amphiregulin expression among different phenotypes of macrophages, we then detected amphiregulin expression in both alveolar and PMs in M1 and M2 phenotypes. Macrophages were stimulated with lipopolysaccharide and interleukin-4 for M1 and M2 polarization, respectively. Cells in control group (M0) were treated with PBS. As shown in fig. 5, A and B, amphiregulin was highly expressed at both mRNA and protein levels exclusively in M1-polarized AMs, whereas amphiregulin expression was very low in M0 and M2 phenotype. Similar results were observed in the PMs (fig. 5, C and D).

To investigate the role of amphiregulin in ALI, mice were given amphiregulin Abs to neutralize amphiregulin 30 min before lipopolysaccharide (1 mg/kg) challenge. IgG was used as an isotype control. HE staining (fig. 6A) showed that treatment with amphiregulin antibodies alone did not cause damage to lung tissues in mice. However, when exposed to lipopolysaccharide, mice pretreated with amphiregulin antibodies suffered from severe lung injury compared with their IgG-pretreated counterparts, as evidenced by more severe alveolar destruction, thicker alveolar wall, higher lung injury score (fig. 6B), more neutrophil infiltration (fig. 6C), increased levels of total protein (fig. 6D), IgM (fig. 6E), and proinflammatory factors (TNF-α and interleukin-6, fig. 6, F and G) in BALF compared with the IgG-pretreated mice.

Then, we examined the protective effect of exogenous amphiregulin on lipopolysaccharide-induced ALI. We made an ALI model by intratracheal administration of 3 mg/kg of lipopolysaccharide. Briefly, mice were first injected intratracheally with rmAreg or PBS and then challenged with lipopolysaccharide 30 min later. HE staining (fig. 7A) exhibited that treatment with 3 mg/kg lipopolysaccharide resulted in severe lung injuries, and rmAreg pretreatment obviously ameliorated alveolar disarray, interstitial edema, and neutrophil infiltration compared with PBS pretreatment (fig. 7B). In contrast, rmAreg administration significantly lowered the lipopolysaccharide-induced increase of neutrophil number, (fig. 7C), TNF-α (fig. 7D), and interleukin-6 (fig. 7E), total protein (fig. 7F), and IgM (fig. 7G) expressions in BALF.

We then investigated whether amphiregulin has protective effect on pulmonary epithelial cells in vitro. MLE12 cells were challenged with lipopolysaccharide to mimic in vivo lung epithelial cells in sepsis, and cell survival was determined by flow cytometry. Lactate dehydrogenase assay found that the difference in cell death was no different among all the groups (shown in fig. 1, Supplemental Digital Content 1, https://links.lww.com/ALN/B246). Figure 8A shows that apoptosis ratio of MLE12 cell significantly increased from 5.41 ± 0.60% in control group to 16.60 ± 1.82% after lipopolysaccharide challenge, but this pro-apoptosis effect of lipopolysaccharide was significantly attenuated by administration of rmAreg in a dose-dependent manner with only 9.47 ± 1.67% cells suffering apoptosis in Areg(50) + lipopolysaccharide group, suggesting that amphiregulin can protect MLE12 cells against apoptosis induced by lipopolysaccharide challenge. To further confirm the effect of AM-released amphiregulin on epithelial apoptosis, MLE12 cells were preincubated with conditioned media of PBS-treated AMs (M0-CM), conditioned media of M1-polarized AMs (M1-CM) treated with IgG, and amphiregulin antibodies. Preincubation with M0-CM and IgG-treated M1-CM exerted no effect on lipopolysaccharide-induced apoptosis of MLE12 cells, but amphiregulin antibodies–treated M1-CM significantly exacerbated lipopolysaccharide-induced apoptosis of MLE12 cells, increasing the apoptosis ratio from 17.45 ± 1.13% to 21.67 ± 1.10% (fig. 8B).

Western blotting (fig. 9, A–C) showed that the levels of EGFR and AKT phosphorylation in lung tissues were very low in the control group but obviously increased in lipopolysaccharide-challenged lung. Amphiregulin antibodies alone did not affect phosphorylation of EGFR and AKT, but pretreatment with amphiregulin antibodies significantly inhibited lipopolysaccharide-induced EGFR and AKT activation. Conversely, rmAreg alone activated both EGFR and AKT in mouse lung tissues, and amphiregulin treatment further enhanced lipopolysaccharide-induced EGFR and AKT phosphorylation (fig. 9, D–F).

To understand the role of amphiregulin in EGFR and AKT activation in pulmonary epithelial cells, we stimulated MLE12 cells with rmAreg. It was found that EGFR phosphorylation rapidly reached a peak within 5 min and declined thereafter to almost baseline levels 30 min after the stimulation (fig. 10, A–C). In contrast, AKT was activated 5 min after challenged with rmAreg, but the phosphorylation level remained unchanged over a time period of 30 min. To examine the effects of amphiregulin secreted by AMs on phosphorylation of EGFR and AKT in MLE12 cells, MLE12 cells were exposed to M1-CM treated with IgG or amphiregulin antibodies and M0-CM, respectively (fig. 10, D–F). The M0-CM did not alter the phosphorylation of EGFR or AKT compared with control cells. Conversely, IgG-treated M1-CM significantly activated EGFR and AKT in MLE12 cells, but the activation could be dramatically attenuated when the MLE12 cells were exposed to amphiregulin antibodies-treated M1-CM.

Activated AMs are generally believed to exert a damaging effect on lung tissues when they work as a defense against infection. In this study, however, we found that M1 phenotype of AMs protect lung tissues by expressing high level of amphiregulin. Our study showed that amphiregulin reduced the influx of protein-rich fluid into alveolar space, inhibited infiltration of inflammatory cells and expression of inflammatory cytokines, and thereby protected the lung against lipopolysaccharide-induced ALI by strengthening epithelial barriers.

Amphiregulin expression in lung tissues of ALI models is poorly understood. Our data showed that lipopolysaccharide challenge caused a significant increase in amphiregulin expression in lung tissues of ALI mice. At the same time, we found that, although the gene expression increased at early stage after lipopolysaccharide challenge, the amphiregulin protein expression sustained for a long time, suggesting that amphiregulin effects may be prolonged during the recovery stage of inflammatory response to balance the proinflammatory response. This finding was consistent with previous results that amphiregulin expression was increased in lung tissues damaged by mechanical ventilation or bleomycin.^11,26  We also observed the concomitant appearance of amphiregulin and Iba-1 (a macrophage biomarker) fluorescence in lipopolysaccharide-challenged lung tissues and coexistence of fluorescence of both amphiregulin and attached cells (macrophages) in BALF isolated from ALI mice. Further investigation revealed that AM depletion resulted in dramatically reduced amphiregulin concentration in BALF of ALI mice. However, lipopolysaccharide challenge of macrophages significantly increased the expression of amphiregulin. Collectively, these findings suggested that AMs were the main sources of amphiregulin production in lipopolysaccharide-induced ALI. Because AMs are activated to M1 phenotype at early phage of ALI,^4,7  we further evaluated the amphiregulin expression in different phenotypes of AMs and PMs in vitro. We found that only M1 macrophages secreted high-level amphiregulin. Both in vivo and in vitro results suggested that amphiregulin was mainly expressed by classically activated macrophages.

To evaluate the biologic significance of amphiregulin up-regulation in lipopolysaccharide-induced ALI, we examined the effects of both exogenous recombinant amphiregulin and amphiregulin antibody on pulmonary pathology (including inflammation and changes in permeability). Our results consistently showed that exogenous recombinant amphiregulin reduced influx of protein-rich fluid into alveolar space, inhibited infiltration of inflammatory cells, down-regulated the expression of inflammatory cytokines, and resultantly protected the lung against lipopolysaccharide-induced ALI. Conversely, amphiregulin antibody abolished the protective effect of endogenous amphiregulin. These results indicated that amphiregulin produced by activated AMs had a protective effect on the lung in the ALI mice.

Previous studies reported that alveolar epithelial cell death in ALI is associated with severity of lung injury, and inhibition of epithelial apoptosis can improve lung function and animal survival in ALI models.^27–29  In view of these findings, we investigated the mechanism underlying the protective effect of amphiregulin by examining the effect of amphiregulin on lipopolysaccharide-induced apoptosis of MLE12 cells (lung epithelial cells). By mimicking events that epithelia experienced in lipopolysaccharide-induced ALI, we found that amphiregulin inhibited the apoptosis of lipopolysaccharide-treated epithelial cells in a concentration-dependent manner. Interestingly, we observed that the conditioned media of M1-polarized AMs did not directly protect MLE12 cells against lipopolysaccharide-induced apoptosis, and neutralization of amphiregulin in M1-CM by amphiregulin Abs significantly enhanced the proapoptotic effect of lipopolysaccharide on MLE12 cells. The absence of protective effect of M1-CM on apoptosis of MLE12 cells might be explained by the fact that high concentrations of proinflammatory cytokines and molecules in M1-CM might promote the lipopolysaccharide-induced proapoptotic effect of MLE12 cells, thereby abolishing the protective effect of amphiregulin in M1-CM on MLE12 cells. These findings suggested that, without the protection of amphiregulin secreted by M1 macrophages in lipopolysaccharide-induced ALI, lung epithelial cells might otherwise have suffered from more severe damage.

Amphiregulin, by exclusively binding to its receptor EGFR, induces phosphorylation of EGFR. In this study, lipopolysaccharide-induced phosphorylation of EGFR and AKT in lung tissues was increased by exogenous amphiregulin but decreased by amphiregulin neutralization. The enhancing effect of exogenous amphiregulin on the phosphorylation was consistent with our in vitro findings that both the recombinant amphiregulin and the amphiregulin protein secreted by M1 AMs directly activated EGFR and AKT in pulmonary epithelial cells. Moreover, the simultaneous phosphorylation of both EGFR and AKT by amphiregulin is in agreement with a previous study demonstrating that phosphorylation of EGFR could further activate the AKT signaling pathway.³⁰  We also found that, amphiregulin protected the pulmonary epithelial cells against lipopolysaccharide-induced apoptosis, thereby strengthening the epithelial barrier, which was coincident with a previous finding that decreased AKT phosphorylation led to more severe damage in lipopolysaccharide-challenged lung tissues.³¹  Westphalen et al.³¹  proposed that the activated AMs could prevent lung injury by initiating Ca²⁺-dependent AKT activation in epithelia at the onset of lipopolysaccharide-induced inflammation. Whether amphiregulin activates AKT signal pathway in a Ca²⁺-dependent manner warrants further investigation. Overall, this study demonstrated that amphiregulin was capable of promoting the phosphorylation of EGFR and AKT in mouse lung epithelial cells under lipopolysaccharide challenge and, as a result, preventing the lung against ALI.

In summary, different from the common concept that classically activated AMs just exert a damaging effect on the lung tissues, this study showed that that classically activated AMs have a protective effect on the lung tissues by expressing a high-level of amphiregulin, which reduced influx of protein-rich fluid into alveolar space, inhibited infiltration of inflammatory cells, down-regulated the expression of inflammatory cytokines through activation of EGFR-AKT pathway, and resultantly protected lung tissues against lipopolysaccharide-induced ALI by strengthening the epithelial barrier.

The authors thank Jinghui Zhang, M.D., the Laboratory of General Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, for technical assistance.

This study was supported by grants 81201454, 30930089, and 30972837 from the National Natural Science Foundation of China (Beijing, China).

The authors declare no competing interests.