Abstract :
[en] Canine idiopathic pulmonary fibrosis (CIPF) mainly affects middle-aged to old dogs from the
West Highland white terrier (WHWT) breed. CIPF is a chronic disease characterized by collagen deposition in the pulmonary interstitium inducing progressive airway failure. Despite numerous investigations, the aetiology of the disease remains unknown, the diagnosis difficult, the pathophysiology misunderstood and the prognosis poor. Currently, there is no curative therapeutic option for that condition and only symptomatic treatments can be used to maintain a quality of life as good as possible for the dogs. CIPF is frequently equated with human idiopathic pulmonary fibrosis (IPF), although the diseases are not strictly identical.
This project was therefore conducted in order to increase pathophysiological knowledges on CIPF and more precisely the ones related to the lung microbiota (LM) and bronchoalveolar lavage fluid (BALF) macrophage clusters. Indeed, macrophages are critical mediators of innate immune responses against bacteria in the lung. They are able to polarize and modulate their phenotypes (inflammatory or pro-fibrotic) to adjust to the microbial environmental conditions. Moreover, recently, studies about IPF have shown that the LM and its impact on the immune system, especially lung macrophages, could have a link with the development, the maintain and the exacerbation of the disease, providing perspectives for designing novel therapeutic strategies.
Before assessing the LM in CIPF, we described the LM in healthy dogs and determined the principal factors potentially able to alter it in healthy conditions, including antimicrobial treatment, age, breed and living conditions. As WHWTs affected with CIPF are frequently referred under antimicrobial drug, we first investigated the short and long-time impact of such treatment on the LM in healthy experimental beagles. We showed that before drug administration, the LM is quite similar among dogs with the predominance of 4 phyla: Proteobacteria, Firmicutes, Bacteroidetes and Actinobacteria. A 10-days oral amoxicillin-clavulanic acid administration induces alteration of the global diversity and the composition of the LM. However, changes nearly disappear 16 days after discontinuation of the drug. The impact of the living conditions and the breed on the LM have then
been assessed by comparing different dogs’ breeds (terriers, WHWTs, beagles, brachycephalic dogs and shepherds) living in different conditions (domestic or experimental). We showed that LM is significantly different depending of the living condition of dogs, while breed has a milder impact. In adult dogs, age doesn’t seem to have an impact on the LM. A core microbiota has been proposed by regrouping data of all published studies related to the LM in healthy dogs and is composed by at least Cutibacterium, Streptococcus, Acinetobacter and Pseudomonas genera. In acute respiratory diseases and particularly in dogs affected with Bordetella bronchiseptica, we showed that 16S rDNA sequencing (the technique used to assess LM) results correlate with results of classical techniques used
to assess bacterial infection in the lung (i.e. polymerase chain reaction and culture). This suggests that the 16S rDNA sequencing is reliable for identifying bacteria involved in canine lung infectious diseases, mainly when rare or slow growing bacteria are concerned. A dysbiosis of the LM is also described in the dogs affected with bordetellosis characterized by a domination of one or two bacteria, a reduction of the diversity and the richness, and a higher bacterial load compared with healthy aged-matched dogs. Lastly, we showed that the LM between healthy WHWTs and WHWTs affected with CIPF is quite similar. Brochothrix, Curvibacter, Pseudarcicella, a genus belonging to Flavobacteriaceae family, Rhodoluna and Limnohabitans genera are increased in WHWTs either healthy or diseased compared with healthy dogs from other breeds. Brochothrix, Pseudarcicella, Curvibacter and a genus belonging to Flavobacteriaceae family are also more abundant in CIPF than
in healthy WHWTs, however, not significantly. We therefore conclude that the presence of a specific LM in WHWTs compared to other breeds may be suspected to be one of the factors that can predispose that breed to CIPF.
In the second part of the thesis, the use of the single-cell mRNA sequencing technique
(scRNA-seq) was first validated in the BALF from healthy dogs. ScRNA-seq is an unbiased and high throughput tool that enables the transcriptomic identification of thousands of single cells at a time and that has never been used in dogs before. By using this technique, we found 14 conserved clusters in the BALF of healthy dogs corresponding to 8 different cell populations: macrophages, lymphocytes, neutrophils, dendritic cells, B cells, mast cells, epithelial cells and cells in division. The scRNA-seq was then used to identify macrophages clusters in WHWTs affected with CIPF compared with healthy WHWTs. In all WHWTs, the same cell populations as described in the previous study were identified. Five clusters of macrophages were found. Among them, two were enriched in pulmonary fibrosis processes compared with other clusters, a cluster of monocytes and a cluster of monocytes-derived macrophages. Pro-fibrotic genes overexpressed in monocytes and monocytes-derived macrophages included CCL2, SPP1, FN1, CCL3, TIMP1, IL1RN, CXCL8 and CCL4, and SFTPC, CCL5, FN1, CXCL8, ATP11A and SPP1, respectively. The differential gene expression in monocytes was not different between CIPF and healthy WHWTs. However, significantly more cells from this cluster were identified in CIPF dogs. Monocytes-derived macrophages were enriched in pulmonary fibrosis but also in angiogenesis and epithelial-mesenchymal transition processes in CIPF compared with healthy
WHWTs. Four pro-fibrotic genes were overexpressed in CIPF compared with healthy WHWTs in that cluster including FN1, SPP1, CXCL8 and PLAU. The presence of those pro-fibrotic macrophage clusters in diseased WHWTs probably participates to the onset and/or the perpetuation of CIPF.
In conclusion, this project allowed to better describe the LM in healthy dogs and its
modifications in WHWTs affected with CIPF in comparison with other lung pathological conditions. The use of the scRNA-seq in dogs’ BALF, after validation revealed the presence of pro-fibrotic macrophage clusters in CIPF compared with healthy WHWTs. Overexpressed pro-fibrotic genes identified in CIPF WHWTs might be used as biomarkers or be targeted for therapeutic treatment, which offers good perspectives for future research about CIPF.
