Bidirectional role of IL-6 signal in pathogenesis of lung fibrosis

Abstract

Background: Various signals are known to participate in the pathogenesis of lung fibrosis. Our aim was to determine which signal is predominantly mobilized in the early inflammatory phase and thereafter modulates the development of lung fibrosis.

Methods: Mice received a single dose of 3 mg/kg body weight of bleomycin (BLM) and were sacrificed at designated days post-instillation (dpi). Lung homogenates and sections from mice in the early inflammatory phase were subjected to phospho-protein array analysis and immunofluorescence studies, respectively. Bronchoalveolar lavage fluid (BALF) from mice was subjected to an enzyme-linked immunosorbent assay (EIA) for interleukin (IL)-6 and evaluation of infiltrated cell populations. The effects of endogenous and exogenous IL-6 on the BLM-induced apoptotic signal in A549 cells and type 2 pneumocytes were elucidated. In addition, the effect of IL-6-neutralizing antibody on BLM-induced lung injury was evaluated.

Results: Phospho-protein array revealed that BLM induced phosphorylation of molecules downstream of the IL-6 receptor such as Stat3 and Akt in the lung at 3 dpi. At 3 dpi, immunofluorescence studies showed that signals of phospho-Stat3 and -Akt were localized in type 2 pneumocytes, and that BLM-induced IL-6-like immunoreactivity was predominantly observed in type 2 pneumocytes. Activation of caspases in BLM-treated A549 cells and type 2 pneumocytes was augmented by application of IL-6-neutralizing antibody, a PI3K inhibitor or a Stat3 inhibitor. EIA revealed that BLM-induced IL-6 in BALF was biphasic, with the first increase from 0.5 to 3 dpi followed by the second increase from 8 to 10 dpi. Blockade of the first increase of IL-6 by IL-6-neutralizing antibody enhanced apoptosis of type 2 pneumocytes and neutrophilic infiltration and markedly accelerated fibrosis in the lung. In contrast, blockade of the second increase of IL-6 by IL-6-neutralizing antibody ameliorated lung fibrosis.

Conclusions: The present study demonstrated that IL-6 could play a bidirectional role in the pathogenesis of lung fibrosis. In particular, upregulation of IL-6 at the early inflammatory stage of BLM-injured lung has antifibrotic activity through regulating the cell fate of type 2 pneumocytes in an autocrine/paracrine manner.

Fig. 1

Effect of BLM on phosphorylation of signaling proteins in lung. Lysates (400 μg) from the lungs of mice instilled with PBS or BLM (3 dpi) were subjected to Pathscan Antibody Array. a Typical profile of BLM-induced protein phosphorylation. Proteins with increased phosphorylation levels in response to BLM are boxed. In particular, phospho-Akt and -Stat3 are boxed by a double-line. b Phosphorylation levels of these proteins in the lungs of mice instilled with PBS (white bars) and BLM (black bars). Using a densitometer, each signal was normalized to the positive internal controls included in the array glass and expressed in arbitrary units. Average signal of the positive internal controls is expressed as 10 arbitrary units. Data are shown as mean ± S.E.M. of three independent experiments. The difference between the two groups (PBS and BLM) was statistically significant (*P < 0.05) by Student’s t-test for unpaired values

Fig. 2

Localization of phospho-Akt and -Stat3 and expression of IL-6 at early inflammatory stage of BLM-injured lung. a Lung sections of mice instilled with PBS or BLM (3 dpi) were stained with anti-phospho-Akt or anti-phospho-Stat3 antibody in combination with anti-proSP-C antibody followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for phospho-Akt and -Stat3; 594 for SP-C). Arrowheads indicate phospho-Akt⁺SP-C⁺ and phospho-Stat3⁺SP-C⁺ cells. b Lung sections from mice instilled with BLM (3 dpi) were stained with anti-IL-6 antibody in combination with anti-proSP-C or anti-Iba1 antibody followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for IL-6; 594 for SP-C and Iba1). Arrowheads indicate IL-6⁺SP-C⁺ and IL-6⁺Iba1⁺ cells. Asterisk indicates a cell with macrophage-like morphological features. c Lung sections from mice instilled with BLM (3 dpi) were stained with anti-IL-6, anti-proSP-C and anti-Iba1 antibodies followed by reaction with Alexa 350-, 488- and Alexa 594-conjugated second antibodies. Data are shown as mean ± S.E.M. (n = 6). Nuclei were stained with DAPI. Similar results to the immunofluorescence profiles in (a) and (b) were observed in four independent experiments

Fig. 3

Endogenous IL-6 suppresses BLM-induced apoptotic signal in alveolar epithelial cells. a Effect of human IL-6-neutralizing antibody, a PI3K inhibitor LY294002, and a Stat3 inhibitor S3I-201 on BLM-induced caspase 8 activation in A549 cells. Lysates (20 μg) from cells treated with or without BLM were subjected to WB analysis for detection of cleaved caspase 8. Application of human IL-6-neutralizing antibody, control IgG, LY294002 and S3I-201 to the cells was performed 30 min prior to BLM treatment. The same lysates were subjected to WB analysis to determine the amount of β-actin as an internal control. b Exogenous IL-6 inhibited BLM-induced caspase 8 activation in A549 cells in a dose-dependent manner. The amount of β-actin of each lane was determined as an internal control. c Effect of mouse IL-6-neutralizing antibody, LY294002, and S3I-201 on BLM-induced caspase 3 activation in primary cultured SP-C⁺ cells. The basic procedure was the same as that for (a)

Fig. 4

Endogenous IL-6 modulates BLM-sensitive cytokine production in type 2 pneumocytes. a Typical expression profile of cytokines. Primary cultured type 2 pneumocytes from mice were treated with BLM in the presence or absence of mouse IL-6-neutralizing antibody, and the supernatants were subjected to Mouse Cytokine Antibody Array C1. Application of mouse IL-6-neutralizing antibody and control IgG to the cells was performed 30 min prior to PBS or BLM treatment. The 12 molecules induced by BLM (BLM + control IgG) are numbered and indicated with the correct location in the membrane map, compared with the unstimulated control (PBS + control IgG). Among the 12 proteins, IL-6-neutralizing antibody (BLM + anti-IL-6)-induced and -reduced molecules are indicated by squares with a double-line and a single-line, respectively. IL-6-neutralizing antibody-insensitive molecules are indicated by a broken-line. We confirmed that the array profile of unstimulated control (PBS + control IgG) showed no difference from that of PBS alone. b Changes in expression of IL-6-neutralizing antibody-sensitive molecules observed in WB array analyzed by densitometer. Using a densitometer, each signal was normalized to the positive internal controls included in the array membrane and expressed in arbitrary units. Average signal of the positive internal controls is expressed as 100 arbitrary units. Data are shown as mean ± S.E.M. of three independent experiments. *P < 0.05, significantly different from value of BLM + control IgG group (ANOVA followed by Tukey's test)

Fig. 5

Time course of BLM-induced IL-6 level in BALF. BALF from mice treated with BLM were subjected to ELISA for mouse IL-6 at the indicated dpi. IL-6 level in BALF prepared from mice with PBS treatment is shown as negative control (PBS). Data are shown as mean ± S.E.M. (n = 4). *P < 0.05, significantly different from the PBS value (ANOVA followed by Tukey's test)

Fig. 6

Blockade of IL-6 at early inflammatory stage of BLM-injured lung enhances apotosis of type 2 pneumocytes and alveolitis. Mice were divided into four groups: PBS group, BLM group, BLM + control IgG group, and BLM + anti-IL-6 group. a Apoptotic type 2 pneumocytes were elucidated by counting YO-PRO-1-permeable cells at 3 dpi. Data are shown as mean ± S.E.M. of twenty sections from each individual (n = 5). *P < 0.05, significantly different from value of BLM-treated lung (ANOVA followed by Tukey's test). b Effect of IL-6-neutralizing antibody on change in number of infiltrated cells into the lung. Data are shown as mean ± S.E.M. (n = 5). *P < 0.05, significantly different from value of BLM-treated lung (ANOVA followed by Tukey's test)

Fig. 7

Blockade of IL-6 at early inflammatory stage of BLM-injured lung accelerates lung fibrosis. Mice were divided into four groups: PBS group, BLM group, BLM + control IgG group, and BLM + anti-IL-6 group. a Effect of IL-6-neutralizing antibody on BLM-induced histopathological changes in the lung. Lung sections from each group at 7 and 14 dpi were stained with Masson’s trichrome to visualize fibrotic lesions. b Semi-quantitative measurement of lung fibrotic change. Eight mice in each group were used. Data are shown as scores of sixteen sections (two sections/mouse). Bars represent median values. *P < 0.05, significantly different from value of BLM-treated lung (ANOVA followed by Tukey's test). The difference between the two groups (7 and 14 dpi of BLM) was statistically significant (#) by Student’s t-test for unpaired values. n.s., no significant difference

Fig. 8

Blockade of IL-6 at early fibrotic stage of BLM-injured lung ameliorates lung fibrosis. Mice were divided into two groups: BLM + control IgG group, and BLM + anti-IL-6 group. a Body weight (BW) of BLM-instilled mice with or without IL-6-neutralizing antibody treatment. Time-dependent change in BW is expressed as percentage of BW of each mouse just before BLM instillation. *P < 0.05, significantly different from BLM + control IgG group at each indicated time point by Student’s t-test. Data represent mean ± S.E.M. (n = 6). b Survival rate of BLM-instilled mice with or without IL-6-neutralizing antibody treatment. Data represent mean ± S.E.M. (n = 12). c Effect of IL-6-neutralizing antibody on BLM-induced histopathological change in the lung. Lung sections from each group at 14 dpi were stained with Masson’s trichrome to visualize fibrotic lesions. d Semi-quantitative measurement of lung fibrotic change. Six mice in each group were used. Data are shown as mean ± S.E.M. of twelve sections (two sections/mouse). *P < 0.05, significantly different from BLM + control IgG group by Student’s t-test

Fig. 9

Localization of phospho-Stat3 and expression of IL-6 at early fibrotic stage of BLM-injured lung. Lung sections from mice instilled with BLM (8 dpi) were used for the following experiments. a Masson’s trichrome staining. b Lung sections were stained with anti-IL-6 and anti-proSP-C antibodies followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for IL-6; 594 for SP-C). Arrowhead indicates IL-6⁺SP-C⁺ cell. c Lung sections were stained with anti-IL-6 and anti-Iba1 antibodies followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for IL-6; 594 for Iba1). Arrowheads indicate IL-6⁺Iba1⁺ cells. d Lung sections were stained with anti-IL-6 and anti-S100A4 antibodies followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for IL-6; 594 for S100A4). e Lung sections were stained with anti-proSP-C and anti-phospho-Stat3 antibodies followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for SP-C; 594 for phospho-Stat3). f Lung sections were stained with anti-SMA and anti-phospho-Stat3 antibodies followed by reaction with Alexa 488- or Alexa 594-conjugated second antibodies (488 for SMA; 594 for phospho-Stat3). Nuclei were stained with DAPI. Similar results to the histopathological changes and immunofluorescence profiles were observed in four independent experiments