Administration of bleomycin via the oropharyngeal aspiration route leads to sustained lung fibrosis in mice and rats as quantified by UTE-MRI and histology

Abstract

Pulmonary fibrosis can be experimentally induced in small rodents by bleomycin. The antibiotic is usually administered via the intratracheal or intranasal routes. In the present study, we investigated the oropharyngeal aspiration of bleomycin as an alternative route for the induction of lung fibrosis in rats and mice. The development of lung injury was followed in vivo by ultrashort echo time magnetic resonance imaging (UTE-MRI) and by post-mortem analyses (histology of collagen, hydroxyproline determination, and qRT-PCR). In C57BL/6 mice, oropharyngeal aspiration of bleomycin led to more prominent lung fibrosis as compared to intranasal administration. Consequently, the oropharyngeal aspiration route allowed a dose reduction of bleomycin and, therewith, a model refinement. Moreover, the distribution of collagen after oropharyngeal aspiration of bleomycin was more homogenous than after intranasal administration: for the oropharyngeal aspiration route, fibrotic areas appeared all over the lung lobes, while for the intranasal route fibrotic lesions appeared mainly around the largest superior airways. Thus, oropharyngeal aspiration of bleomycin induced morphological changes that were more comparable to the human disease than the intranasal administration route did. Oropharyngeal aspiration of bleomycin led to a homogeneous fibrotic injury also in rat lungs. The present data suggest oropharyngeal aspiration of bleomycin as a less invasive means to induce homogeneous and sustained fibrosis in the lungs of mice and rats.

Figure 1. HPLC/MS spectra of bleomycin (0.075 mg/mL) freshly prepared and after six days at 4°C acquired on an Acquity Ultra performance LC instrument (Waters, Milford, Massachusetts, USA).

An Acquity HSS T3 1.8 µm 2.1×50 mm column was used at 60°C; eluent A (water +0.05% formic acid +3.75 mM ammonium acetate) and eluent B (acetonitrile +0.04% formic acid) were used with a gradient of 5 to 98% B in 1.4 min and 1.0 mL/min flow. Bleomycin A2: C55H84N17O21S3+ with a mass of 1415.56 g/mol; bleomycin B2: C55H84N20O21S2 with a mass of 1425.52 g/mol.

Figure 2. Details of segmentation procedure.

For each slice, a border is drawn manually to limit the segmentation of high intensity signals to the lung. Images correspond to a rat at A) baseline and B) day 7 after OA of bleomycin (2 mg/kg).

Figure 3. Picrosirius-stained histological slices of C57BL/6 mice observed at bright-field (left column) and at polarization (right column).

The high intensity areas in the polarized view, corresponding to collagen, coincide well with the red areas in the bright-field view. In this work, collagen has been quantified by examining histological slices at bright-field. A) Saline-challenged mouse (OA administration). B) IN administration of bleomycin (6×0.25 mg/kg). C) OA administration of bleomycin (6×0.25 mg/kg).

Figure 4. Detection of bleomycin-induced lung injury by UTE-MRI in BALB/c mice.

A) Measurements (4-min-acquisition-time) on comparable slices before and at different timepoints (5, 9, and 23 days) after OA of 6 × 1.0 mg/kg bleomycin. Arrows point to bleomycin-induced lung injury. B) Comparison of UTE acquisitions with different numbers of averages in a bleomycin-treated animal. The volumes of signals in the lungs evaluated from these images were comparable for all acquisition conditions: 1 average (285.4 µL), 2 averages (281.7 µL), 4 averages (282.6 µL), 8 averages (280.4 µL). White arrows point to bleomycin induced lung injury.

Figure 5. Comparison of OA to IN administration of bleomycin (6×0.25 mg/kg) in C57BL/6 mice (n = 8 per group) by in vivo MRI and post-mortem analyses.

All data are shown as means ± STDEV. A) Relative body weight and time point of death during the study. Animals have been sacrificed directly after their last measurements. Five out of eight mice receiving OA of bleomycin had to be sacrificed prematurely due to excessive bodyweight loss (A). B) Total volume of MRI signals in the lung (µL). The signal at baseline contains contributions from vessels, whereas signals following bleomycin reflect additionally the injury inflicted by the antibiotic. C) Post-mortem determination of hydroxyproline in right lung lobes (day 37). D) Amount of collagen detected by histology of picrosirius in left lung lobes (day 37). Results are expressed as collagen content relative to mean collagen content in the lungs of saline-treated mice (picrosirius factor).

Figure 6. Histology of picrosirius stained lung slices from C57BL/6 mice.

Collagen (green) visualized by picrosirius staining in three slices of the left lung lobe (upper row). Magnified views (×200) of a portion of the corresponding histological slices are shown in the bottom row. A) Stained collagen content in saline treated mouse lung 37 days after administration. B) Stained collagen content in a mouse lung 37 days after last IN bleomycin administration (0.25 mg/kg on six consecutive days). C) Collagen content in a mouse lung 37 days after last OA of bleomycin (0.25 mg/kg on six consecutive days). Compared to the IN administration, the amount of picrosirius was six times higher when bleomycin was oropharyngeally aspirated and the distribution of picrosirus staining was more homogeneous.

Figure 7. Ventral NIRF images of the lungs from two mice 60 min after instillation of the dye Cy5.5 (0.1 mg/mL) either via the IN route (A) or via OA (B).Yellow arrows point to the trachea.

Figure 8. Detection of bleomycin-induced lung injury in C57BL/6 mice by UTE-MRI, picrosirius staining, and hydroxyproline determination.

Animals were treated with 0.1 mg/kg bleomycin (n = 9) or vehicle (n = 5) on six consecutive days via OA. Data are shown as means ± STDEV. A) Relative body weight during the course of the experiment. The statistical significance values ***p<0.001 represent comparisons between saline and bleomycin-treated mice, at each time point. B) Total volume of MRI signals in the lung (µL). The signal at baseline contains contributions from vessels, whereas signals following bleomycin administration reflect additionally the injury induced by the antibiotic. The statistical significance values ***p<0.001 indicate comparison to baseline. C) Post-mortem determination of hydroxyproline in right lung lobes D) Picrosirius factor expressing the amount of collagen relative to the mean collagen content assessed in the lungs of saline-treated mice. The statistical significance value ***p<0.001 means comparison to saline treatment.

Figure 9. Dose adaptation of bleomycin to OA route and comparison of different bleomycin doses in BALB/c mice by UTE-MRI and post-mortem analyses (n = 8 mice for each group except n = 4 mice for histological analysis).

Data are shown as means±STDEV. One-way ANOVA (Two-Way repeated measures ANOVA for MRI-results) with Bonferroni tests was used for statistical analysis. A) Total volume of MRI signals detected in the lung by UTE-MRI The signal at baseline contains contributions from vessels, whereas signals following bleomycin administration reflect additionally the injury induced by the antibiotic. B) Collagen content in left lung lobes detected by picrosirius staining. The picrosirius factor expresses the amount of collagen relative to the mean collagen content in the lungs of saline-treated mice. B) Post-mortem determination of hydroxyproline in right lung lobes (day 23); Data are shown as Mean ± STDEV of µg hydroxyproline per right lung. D-F) mRNA expression level of Col1α1, F4/80, and MMP12 compared to expression of HPRT. Data shown as means±STDEV of fold induction compared to control group.

Figure 10. Oropharyngeal aspiration of 2 mg/kg bleomycin or of saline in Sprague Dawley rats (n = 5 per group); A) Volumes of signals (means±STDEV) detected by UTE-MRI in the lungs of rats before and after bleomycin; Two-way repeated measures ANOVA with Bonferroni tests was performed.

The statistical significance values ***p<0.001, **p<0.01 indicate comparison to baseline as well as comparison to Saline group. B) Signals at baseline contain contributions from vessel. (white arrow in upper panel), whereas signals following bleomycin administration reflect additionally the injury induced by the antibiotic (yellow arrows in lower panel). C) Histological analysis of the left lung lobes. Yellow areas are representing collagen, stained by picrosirius; neither emphysema nor edema was observed. D) Magnified view (×200) of one of the histological slices demonstrating widespread fibrosis as evidenced by picrosirius staining.