A novel quantification method of lung fibrosis based on Micro-CT images developed with the optimized pulmonary fibrosis mice model induced by bleomycin

Abstract

Background and aims: Idiopathic pulmonary fibrosis (IPF) is a fibrosing lung disease with unknown etiology, leading to cough and dyspnoea, which is also one of the most common sequelae affecting the quality of life of COVID-19 survivors. There is no cure for IPF patients. We aim to develop a reliable IPF animal model with quantification of fibrosis based on Micro-Computer Tomography (micro-CT) images for the new drug discovery, because different bleomycin administration routes, doses, and intervals are reported in the literature, and there is no quantitative assessment of pulmonary fibrosis based on micro-CT images in animal studies.

Methods: We compared three dosages (1.25 mg/kg, 2.5 mg/kg, and 5 mg/kg) of intratracheal bleomycin administration and experiment intervals (14 and 21 days) in C57BL/6 mice by investigating survival rates, pulmonary histopathology, micro-CT, peripheral CD4⁺ & CD8⁺ cells, and cytokines. Moreover, a simple and reliable new method was developed for scoring fibrosis in live mice based on Micro-CT images by using Image J software, which transfers the dark sections in pulmonary Micro-CT images to light colors on a black background.

Results: The levels of hydroxyproline, inflammation cytokine, fibrotic pathological changes, and collagen deposition in the lungs of mice were bleomycin dose-dependent and time-dependent as well as the body weight loss. Based on the above results, the mice model at 21 days after being given bleomycin at 1.25 mg/kg has optimal pulmonary fibrosis with a high survival rate and low toxicity. There is a significant decrease in the light area (gray value at 9.86 ± 0.72) in the BLM mice, indicating that a significant decrease in the alveolar air area was observed in BLM injured mice compared to normal groups (###p < 0.001), while the Pirfenidone administration increased the light area (gray value) to 21.71 ± 2.95 which is close to the value observed in the normal mice (gray value at 23.23 ± 1.66), which is consistent with the protein levels of Col1A1, and α-SMA. Notably, the standard deviations for the consecutive six images of each group indicate the precision of this developed quantitation method for the micro-CT image taken at the fifth rib of each mouse.

Conclusion: Provided a quantifying method for Micro-CT images in an optimal and repeatable pulmonary fibrosis mice model for exploring novel therapeutic interventions.

Keywords: Bleomycin-induced pulmonary fibrosis mice model; Dosages; Experiment intervals; Quantitative micro-CT assessment.

Fig. 1

Weight and survival rate changes after the instillation of bleomycin in mice at three different dosages on Day 14 or Day 21. A) The study design and working flow. B) There was a significant decrease in the body weights after being given 5 mg/kg, 2.5 mg/kg, or 1.25 mg/kg BLM compared with the normal group. C) The mice in high dosage of BLM (5 mg/kg) have only a 50% survival rate, while dosages of 2.5 mg/kg and 1.25 mg/kg BLM have about 80% survival rate for both 14 and 21 days. Data are shown in means ± SEM（n = 8）. *p < 0.05 and ***p < 0.001 via comparing the normal group.

Fig. 2

Pulmonary fibrosis was developed in all BLM-treated groups after 14 days or 21 days of treatment of BLM. A) Hydroxyproline concentrations were increased in all BLM-treated groups compared with the normal group. The increase of HYP after 21 days was more significant and stable than that of 14 days after BLM treatment. B) The expression levels of α-SMA and Col 1A1 were evaluated by Western blot and shown in bar charts with C) for α-SMA, and D) for Col 1A1. Data were expressed with mean ± SEM (n = 8). *p < 0.05, **p < 0.01and ***p < 0.001 via comparing the normal group.

Fig. 3

Comparison of pathological features of fibrosis after BLM treated 21 days or 14 days. A) The H&E staining (x100) and semi-quantitative evaluation by using the Ashcroft scoring method. B) The Masson staining (x100) and semi-quantitative evaluation by using the Lee H.J scoring method. C) The typical Micro-CT and 3D show representative images of Micro-CT images, and the second line displays images of Micro-CT images in 3D mode. Data are shown by means ± SEM (n = 8). *p < 0.05 and ***p < 0.001 comparing the normal group.

Fig. 4

Different dosages of bleomycin had different effects on the increase of inflammatory cytokines in the pulmonary fibrosis model. Three inflammatory factors included tumor necrosis factor-α (TNF-α) (A), interleukin-1β (IL-1β) (B), and interleukin-6 (IL-6) (C) were elevated in the model groups on Day 14 and remained stable on Day 21 after BLM treatment compared with the normal group. Data are means ± SEM (n = 8). *p < 0.05 and **p < 0.01 via comparing the normal group.

Fig. 5

The altered percentages of CD4⁺ and CD8⁺ T cells were observed in the peripheral blood in the bleomycin-induced pulmonary fibrosis model. A) Compare with the normal group, the percentage of CD4⁺ cells decreased in bleomycin treatment groups. B) And CD8⁺ increased in bleomycin treatment groups. Data were shown as means ± SEM (n = 8). *p < 0.05 and **p < 0.01 comparing the normal group.

Fig. 6

A quantitative method for Micro-CT image by using Image J software. A) The workflow for the quantification of Micro-CT images. The Micro-CT images were taken at the fifth rib of every mouse and converted by image J. The light area (gray value) was calculated for the converted images by Image J. Moreover, the consecutive six images taken at the fifth rib for each mouse from the normal group, BLM group, and PFD treatment BLM group demonstrated that there are very small deviations. Moreover, there is a significant decrease in the light area (gray value) in the image converted by Image J observed in BLM injured mice compared to normal control groups, while the PFD administration increased the light area (gray value). B) The protein levels of Col1A1 and α-SMA in pulmonary tissue correspond to the results of Micro-CT images in Fig. 6A. Data is shown at means ± SEM (n = 6). **p < 0.01, ***p < 0.001 comparing the normal group, ^###p < 0.001 comparing the BLM group.