I worked with the Department of Respiratory Medicine at the Lausanne University Hospital (CHUV). Here, they collected samples and performed 16S rRNA amplicon sequencing on 234 longitudinal bronchoalveolar lavage (BAL) samples from lung transplant recipients over a period of 48 months. I combine this amplicon sequencing data with my culture-based data, host gene expression, patient metadata and machine learning approaches to explore the microbiota of the deep lung. Thereby showing that specific microbial communities or "pneumotypes" are associated with infection, rejection and proper functioning of the transplanted organ. In addition, I cultivated samples on different media under various oxygen concentrations, genotyped individual bacteria. I have created a large biobank called the Lung Microbiota Culture Collection (LUMICOL) from human BAL samples and made in publicly available and have started getting requests to acquire the collection by multiple research groups globally. I combined amplicon sequencing data with bacterial culture results, bacterial and commensal virus numbers, host gene expression data, and patient metadata and used machine learning approaches to explore the microbiota of the deep lung. Next, we wanted to understand how these individual species from the human lower respiratory tract influence human innate immunity. For this, I established a screening approach to identify key lung bacteria and associated immune response in macrophages, the dominant immune cells in lung tissues. For this, I utilised a variant of commercially available cell line called THP-1 human monocytes expressing a secreted embryonic alkaline phosphatase (SEAP) reporter gene under the transcriptional control of Nuclear factor kappa (NFk)-B, the major transcription factor in inflammatory responses. Our aim here was to push the system to be phenotypically similar to the human alveolar macrophages. Hence, I optimised the differentiation procedure and used flow cytometry and mass cytometry in collaboration with Flow Cytometry Core Facility, EPFL, Lausanne, to investigate surface protein markers on macrophages to assess its immunotype. Using these macrophages, we establish a fast, reliable, scalable screening strategy using to investigate the inflammatory potential of diverse bacteria from human lung. We have also made small artifical communities (SACs) by mixing individual bacteria from LUMICOL, which are essentially guided by the observations we made about the different pneumotypes and their taxonomic composition. These SACs were also exposed to macrophages for assessing its inflammatory activity. For the second part of this objective, we initially planned for using animal models. Ongoing work includes host gene expression studies by RNA-seq. Finally, we are now investigating mechanisms by which specific lung bacteria or SACs induce immune changes. For this we aim to know what factors are responsible in both bacteria and host that play important role in these interactions. We have sequenced genomes of major lung bacteria in LUMICOL to know the genes and pathways that exist in the lung microbial ecosystem. We determine the recognition mechanism of lung bacteria by human Pathogen Recognition Receptors (PRRs) and when differentially recognised by macrophages. We are now close to showing that these interactions may be governed by variations in bacterial surface antigens like Lipopolysaccharide (LPS) and Peptidoglycan.
