Mouse model validity for studying the impact of tobacco smoke on the human gut microbiota assessed via in silico and experimental approaches.

Abstract

Introduction: The contribution of cigarettes to disease initiation, manifestation and progression is well-established for complex disorders, such as inflammatory bowel disease. However, studying its impact on disease pathophysiology in a controlled setting is challenging in humans, resulting in the application of various model systems, amongst them tobacco smoke-exposed mice. While frequently employed, it is unclear to what extent this model reflects human responses to tobacco smoke.

Methods: Employing a mouse study of experimental nature, we assessed established parameters for monitoring responses to tobacco smoke, paralleled by 16S rRNA gene-based profiling of the murine gut microbiome in n=32 suitable animals. This was supplemented by a case-control study design, based on n=3 publicly available transcriptome datasets, from human oral mucosa, human large airway epithelium and murine lung tissues, where we assessed which components of the response to tobacco smoke observed in mice are functionally comparable to responses seen in humans.

Results: We observed several physiological responses in mice that paralleled human scenarios (weight loss, serum cotinine and Cyp1a1 mRNA expression), serving as a proof of principle. We identified shared microbiome-associated processes: stress related functions were enriched in mice and humans, while other processes, such as inflammatory functions, were discordant. The mouse microbiota showed significant changes in response to tobacco smoke, which mimicked patterns seen in human datasets, such as changes for Lachnospiraceae and Prevotellaceae. In contrast, some families that show significant responses to tobacco smoke in humans, such as Bacteroidaceae, could not be observed in mice.

Conclusions: Considering the high inter-individual variation in humans and the well-controlled conditions in mice, our results suggest that mice, despite the identified limitations, most likely represent a suitable model for studying specific processes, such as stress responses, in the context of tobacco smoke exposure and its impact on the microbiota.