Fibroblast activation protein is a cellular marker of fibrotic activity in canine idiopathic pulmonary fibrosis

Abstract

Canine idiopathic pulmonary fibrosis (CIPF) is a progressive fibrotic interstitial lung disease of unknown etiology, afflicting aging West Highland white terriers (WHWTs) and leading to progressive respiratory failure. Fibroblast activation protein (FAP), a protease overexpressed in many cancers, is upregulated in idiopathic pulmonary fibrosis in humans. The aim of this study was to investigate FAP as a marker of active fibrosis in lung biopsies from WHWTs affected with CIPF, as well as the potential of plasmatic FAP as a biomarker. After establishing a scoring system to evaluate the severity and activity of fibrosis on histopathological lung sections, anti-FAP immunohistochemistry was performed on healthy and CIPF samples. FAP expression was characterized using both visual and digital quantitative pathology software analyses and then correlated to fibrosis severity and activity. Levels of plasmatic FAP in WHWTs affected with CIPF were measured by enzyme-linked immunosorbent assay and compared with healthy dogs. Lung samples from 22 WHWTs affected with CIPF were collected. According to the fibrosis scoring system, they were classified as cases of mild (5), moderate (9) and severe (8) fibrosis and were attributed scores of fibrosis activity. Fifteen healthy lung samples were classified as non-fibrotic. Healthy lung samples were FAP-negative, whereas fibroblasts were FAP-positive in 20 CIPF samples. FAP immunohistochemical expression correlated mildly with fibrosis severity (p < 0.05; R ² = 0.22) but highly with fibrosis activity scores (p < 0.001; R ² = 0.68). Digital image analysis detected a higher percentage of FAP-positive cells in areas of active fibrosis (p < 0.001) and FAP-positive cells were distributed outside mature fibrosis lesions, clustered in active fibrosis areas or scattered within alveolar septa. On the other hand, plasmatic FAP was significantly lower in dogs affected with CIPF compared with healthy dogs (p < 0.01). In conclusion, this study provides a valuable histological scoring system to assess the severity and activity of fibrosis in CIPF. It demonstrates that FAP is a good cellular marker of fibrotic activity in CIPF, and thus constitutes a promising target to be exploited for diagnostic and therapeutic applications. Additionally, it suggests that plasmatic FAP, although non-specific, could be altered in CIPF.

Keywords: West Highland white terrier; dog; fibroblast activation protein; fibrosis; idiopathic pulmonary fibrosis; immunohistochemistry; lung.

Figure 1

Comparison of FAP immunostaining in canine lung biopsies. No FAP expression in healthy lung (A) and high FAP expression in CIPF (B) and primary pulmonary adenocarcinoma, either in cancer-associated fibroblasts (C) or in cancer cells (D). Immunoperoxidase-diaminobenzidine, hematoxylin counterstain (bar: 100 μm). FAP, fibroblast activation protein; CIPF, canine idiopathic pulmonary fibrosis.

Figure 2

Scatterplot displaying the relationship between the FAP expression index and the grade of active fibrosis. The index of FAP expression (from 0 to 3) is positively correlated (p < 0.001; R² = 0.68) to the score of active fibrosis (from 0 to 3) in lung sections of CIPF. FAP, fibroblast activation protein; CIPF, canine idiopathic pulmonary fibrosis; n, number of cases.

Figure 3

Box-and-whisker plots of collagen content (A) and of FAP-positive cells (B) in areas representing either active (n = 10) or mature fibrosis (n = 10), calculated with quantitative digital analysis in CIPF lung sections. The box represents the median and interquartile range. The whiskers represent the values within 1.5 times the interquartile range. Significance level: *** indicates a p-value below 0.001. FAP, fibroblast activation protein.

Figure 4

Panel showing sequential sections in HE staining (1st row), Picro Sirus red staining (2nd row), anti-FAP immunohistochemistry staining (3rd row) and the superimposition of cell detections (blue: FAP-positive, green: FAP-negative) based on anti-FAP immunohistochemistry onto Picro Sirus red-stained collagen (4th row). Images (A–D) show an area of strongly collagenic mature fibrosis with rare FAP-positive cells (bar: 50 μm). Images (E–H) illustrate an area of active fibrosis with low collagen content and numerous FAP-positive cells (bar: 50 μm). Images (I–L) show an entire section with mixed fibrosis pattern: few FAP-positive cells within highly collagenic mature fibrosis areas, from which less collagenic, FAP-rich areas extend (bar: 500 μm). FAP, fibroblast activation protein.

Figure 5

Box-and-whisker plot of plasma levels of soluble FAP in dogs with CIPF (n = 6) and in healthy dogs (n = 9). The box represents the median and interquartile range, and the whiskers represent the values within 1.5 times the interquartile range. Significance level: ** indicates a p-value below 0.01. FAP, fibroblast activation protein; CIPF, canine idiopathic pulmonary fibrosis.