Modeling host-microbiota interactions in disease using organoid co-cultures

Abstract

Adult stem cell (ASC)-derived organoids are miniature replicas of human epithelial organs. These organoids represent the molecular characteristic of their epithelium of origin, and thus, they can be used as proxies to study human biology in a dish. Many human epithelia, like the gastrointestinal tract or the airways, are in close contact with the microbiota. This microorganismal community contributes to the physiological functioning of our organism, but specific bacterial species are pathogenic. Because of this, the co-culture of ASC organoids with diverse bacteria species is emerging as a powerful tool to study bacterial pathogenesis in vitro. Thus, the work presented in this thesis aims to establish and characterize organoid-bacteria co-culture models to gain mechanistic insight on how pathogenic bacteria cause disease.

In the first part, we review the current knowledge on bacterial associations with colorectal cancer (CRC), existing mechanistic insight on how these may contribute to the disease (Chapter 1), and the state-of-the-art approaches to study bacteria-epithelium interactions using co-culture models (Chapter 2). Next, we describe the methods used for human intestinal organoid work (Chapter 3), and the establishment of organoid-bacterial co-cultures that we developed (Chapter 4).

In the second part, we use these co-culture models to study the role of CRC-associated pks+ E. coli in the disease (Chapter 5). The pks operon confers these bacteria the ability to produce the genotoxin colibactin. We discovered that colibactin-producing pks+ E. coli induces mutations in organoids, in the shape of mutational signatures. These signatures can be understood as specific mutational footprints left in the genome by colibactin, and can be used as a proxy to detect colibactin-mutagenesis in other samples. This led to the identification, for the first time, of bacteria-induced mutations in cancer samples, that were particularly enriched in a subset of CRC patients. Additionally, our results suggest that colibactin-induced mutations might affect APC, the main CRC driver. Furthermore, we assessed the mutagenic ability of a pks+ probiotic strain, by an approach based on the DNA motif characteristic of colibactin-induced mutations (Chapter 6). Beyond CRC, the increasing rate of antimicrobial resistant (AMR) strains is becoming a major problem worldwide. Particularly, AMR strains of P. aeruginosa cause chronic infections in individuals with cystic fibrosis, contributing to its high mortality rate (Chapter 7.1). We established co-cultures between the P. aeruginosa and airway organoids in 2D. We characterized the transcriptional response to this interaction, from both the epithelium and the bacterial side, using Dual RNA sequencing (Chapter 7.2). This revealed epithelial-induced changes in the bacteria related to metabolism, expression of virulence factors and antibiotic resistance. Importantly, several of these processes are recapitulated in vivo. Thus, this co-culture can serve as a platform for future studies on P. aeruginosa pathogenesis on cystic fibrosis, and as an improved model to test novel antimicrobials.

The last chapter (Chapter 8) discusses the implications of these findings, putting them in the light of most current knowledge of the field of host-microbiota interactions in disease, and devises future steps to make the field move towards more clinically relevant discoveries.