A New Model of Acute Exacerbation of Experimental Pulmonary Fibrosis in Mice

Abstract

Rationale: idiopathic pulmonary fibrosis (IPF) is the most severe form of fibrosing interstitial lung disease, characterized by progressive respiratory failure leading to death. IPF's natural history is heterogeneous, and its progression unpredictable. Most patients develop a progressive decline of respiratory function over years; some remain stable, but others present a fast-respiratory deterioration without identifiable cause, classified as acute exacerbation (AE).

Objectives: to develop and characterize an experimental mice model of lung fibrosis AE, mimicking IPF-AE at the functional, histopathological, cellular and molecular levels.

Methods: we established in C57BL/6 male mice a chronic pulmonary fibrosis using a repetitive low-dose bleomycin (BLM) intratracheal (IT) instillation regimen (four instillations of BLM every 2 weeks), followed by two IT instillations of a simple or double-dose BLM challenge to induce AE. Clinical follow-up and histological and molecular analyses were done for fibrotic and inflammatory lung remodeling analysis.

Measurements and main results: as compared with a low-dose BLM regimen, this AE model induced a late burst of animal mortality, worsened lung fibrosis and remodeling, and superadded histopathological features as observed in humans IPF-AE. This was associated with stronger inflammation, increased macrophage infiltration of lung tissue and increased levels of pro-inflammatory cytokines in lung homogenates. Finally, it induced in the remodeled lung a diffuse expression of hypoxia-inducible factor 1α, a hallmark of tissular hypoxia response and a major player in the progression of IPF.

Conclusion: this new model is a promising model of AE in chronic pulmonary fibrosis that could be relevant to mimic IPF-AE in preclinical trials.

Keywords: acute exacerbation; animal model; inflammation; lung fibrosis.

Figure 1

Experimental protocol, survival and weight curves. (A) Eight-week-old C57BL6/J male mice received six IT instillations (arrows) of bleomycin (BLM) at 0.8 IU/g or PBS every two weeks to form the BLM (pink arrow) and PBS groups (white arrow), respectively. In parallel, a group of mice received four IT instillations of 0.8 UI/g BLM, and then two IT instillations of BLM at 1.6 UI/g, thus forming the acute exacerbation group (BLM-AE, pink-hatched black arrow). Mice were sacrificed at day 90 (D90) for analyses. (B) Body-weight monitoring was established throughout the experiment, the difference in weight compared to the weight measured on day 0 (D0) before the IT instillations of PBS (n = 6, black line), BLM, (n = 6, dotted pink line) and BLM-AE (n = 7, pink line) groups are reported on a graph. A Friedman test followed by a Dunn’s multiple comparisons test was performed to estimate the difference in mice weight as compared to initial weight, *** p < 0.001. (C) Mouse survival was recorded every day until the end of the experimental design for the different groups (PBS, black line; BLM, dotted pink line; and BLM-AE, pink line) and plotted on a Kaplan–Meier curve. (D) Mouse peripheral oxygen saturation (SpO2) was measured by infrared pulse oximetry. (E) Lung compliance was measured by plethysmography. (D,E) Data were presented as a box plot representing the median ± interquartile range. Raw data were submitted to one-way ANOVA test followed by Newman–Keuls to compare each group (control PBS group (white bar), BLM group (pink bar) and BLM-AE group (pink hatched black bar)). ns: not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, respectively.

Figure 2

Histological study and analysis of collagen deposits. (A,C,E) Representative mapping of hematoxylin and eosin-stained from the PBS (A), BLM (C) and BLM-AE (E) groups to evaluate global lung architecture. (B,D,F) Representative Sirius red and fast green lung sections from the PBS (B), BLM (D) and BLM-AE (F) groups (scale bar corresponding to 100 µm) to evaluate collagen deposits and lung remodeling. (G) Lung injured area was estimated by quantification of sections density after H&E staining using HistoLab® image analysis software and expressed as the percentage of the total lung section. Injury quantification was normalized to the mean value of control group and presented as a percentage. (H) Light microscopy quantification of Sirius red staining reported to fast green staining quantification. (I) Quantification of soluble collagen in the right lung middle lobe by the Sircol method. (G–I) Data were presented as a box plot representing the median ± interquartile range. Raw data were submitted to one-way ANOVA test followed by Newman–Keuls to compare each group (control PBS group (white bar), BLM group (pink bar) and BLM-AE group (pink hatched black bar)), ns: not significant, * p < 0.05, ** p < 0.01, *** p < 0.001, respectively.

Figure 3

Lung histopathology in the BLM group. Illustration of lung lesions observed in different mice after repetitive low-dose BLM IT instillations. (A) Subpleural distribution with inward extension of the dense lung remodeling (red arrow) with cystic airways remnants (H&E bar 1500 µm). (B) Higher magnification of the A area framed in yellow showing a neat demarcation (red arrow) with the normal lung (scale bar 500 µm) magnified 4x in the insert. (C) Subpleural remodeling showing alveolar collapse especially in the framed area (H&E bar 500 µm). (D,E) Dense fibrotic areas with cystic airways reaching the pleural limit (stars). (E) is a magnification of the (D) framed area (Sirius red/green staining; (D): scale bar 1500 µm; (E): scale bar 300 µm).

Figure 4

Lung histopathology in the BLM-AE groups. Illustration of lung lesions observed in different mice after double-dose BLM intratracheal instillations. (A) Subpleural dense cellular and fibrotic remodeling observed by Masson’s staining. Scale bar 2000 µm. (D) Magnification of the red framed area in A, scale bar 200 µm. (B,C,E,F) Magnification of the yellow framed area in A showing mononuclear cell infiltration. (B) Perivascular mononuclear cell infiltration (arrowheads). Scale bar 100 µm. (C,E) Acute alveolar cell lesions (arrows) and blue matrix deposits. (C) Scale bar 100 µm, (E) scale bar 150 µm, (F) “Reactive” airway cells (arrowhead) compare to normal cells (arrows). Scale bar 150 µm. (G–I) Subacute lesions in a dense remodeled area. (G) Masson’s staining. Scale bar 200 µm. (H) Magnification of the yellow framed area in G showing an interalveolar bud (star) layered by atypical epithelial cells (arrow). Scale bar 50 µm. (I) Red-stained intercellular collagen deposits (arrowhead) within the cellular infiltration (Sirius red/fast green staining, scale bar 100 µm).

Figure 5

Identification of macrophage and lymphocyte cell populations. (A,D,G) Immunostained representative lung sections for F4-80 macrophages, (B,E,H) CD3 T lymphocytes and (C,F,I) CD19 B lymphocytes identification in the PBS (A–C), BLM (D–F), BLM-AE (G–I) groups. Scale bar corresponding to 100 µm. A representative image of the PBS, BLM and BLM-AE groups is shown. After BLM instillation, infiltration of F4-80 macrophages is observed in the BLM group (D) and an upsurge is shown after exacerbation (BLM-AE, (G)). T lymphocyte infiltration was observed in the BLM group ((E,F), respectively) and after exacerbation ((H,I), respectively). (J) Quantification of macrophages, (K) CD3-LT and (L) CD19-LB in the control PBS group (white bar), BLM group (pink bar) and BLM-AE group (pink hatched black bar). Six mice from the PBS group, six mice from the BLM group and seven mice from the BLM-AE were photographed. For each field in each mouse and in each condition, the average number of positive cells for F4-80, CD3 and CD19 immunolabelling was reported to the total number of cells present. All values are represented as median ± interquartile range; one-way ANOVA analysis followed by Newman–Keuls was performed. ns: not significant *** p ≤ 0.001.

Figure 6

Quantification of pro-inflammatory cytokines in whole lung extracts: Quantification of KC (A), IL-1β (B), TNF-α (C) and IL-6 (D) pro-inflammatory proteins in whole-lung protein lysates from the control PBS group (white bar), BLM group (pink bar) and BLM-AE group (pink hatched black bar). All values are represented as median ± interquartile range; one-way ANOVA analysis followed by Newman–Keuls was performed. ns: not significant, ** p ≤ 0.005, *** p ≤ 0.001.

Figure 7

Hypoxia-induced factor HIF-1α immunostaining and expression level of target genes. (B–E) Photographs of 5 µm serial lung sections immunostained for HIF-1α protein and a rabbit isotype. (Scale bar corresponding to 50 µm). A representative image of the PBS (B), BLM (C), and BLM-AE (D) groups was shown. When reported to the staining of isotype (A), no significant HIF-1α staining was observed in the BLM group as compared to the PBS control group. Significant increase staining of HIF-1α is observed in the BLM-AE group mainly localized in modified alveolar epithelial cells (black arrow) (D). (E) mRNA expression of BNIP3 and (F) mRNA of SERPINE1 (PAI-1) quantified by RT-qPCR from the control PBS group (white bar), BLM group (pink bar) and BLM-AE group (pink hatched black bar). Effect of BLM or BLM-AE was presented as fold induction normalized to the mean value of PBS group and reported to 1. All values are represented as median ± interquartile range; one-way ANOVA analysis followed by Newman–Keuls was performed. ns: not significant, * p ≤ 0.01.