ABSTRACT
Bacterial communities are highly complex, with interaction networks dictating ecosystem function. Bacterial interactions are constrained by the spatial organization of these microbial communities, yet studying the spatial organization of microbial communities at the single-cell level has been technically challenging. Here, we use the recently developed high-phylogenetic-resolution microbiota mapping by fluorescence in situ hybridization technology to image the gut microbiota at the species and single-cell level. We simultaneously image 63 different bacterial species to spatially characterize the perturbation and recovery of the gut microbiota to ampicillin and vancomycin in the cecum and distal colon of mice. To decipher the biology in this complex imaging data, we developed an analytical framework to characterize the spatial changes of the gut microbiota to a perturbation. The three-tiered analytical approach includes image-level diversity, pairwise colocalization analysis, and hypothesis-driven neighborhood analysis. Through this workflow, we identify biogeographic and antibiotic-based differences in the spatial organization of the gut microbiota. We demonstrate that the cecal microbiota has increased micrometer-scale diversity than the colon at baseline and recovers better from perturbation. Also, we identify potential foundation and keystone species that have high baseline neighborhood richness and that are associated with recovery from antibiotics. Through this workflow, we add a spatial layer to the characterization of bacterial communities and progress toward a better understanding of bacterial interactions leading to improved microbiome modulation strategies. IMPORTANCE: Antibiotics have broad off-target effects on the gut microbiome. When the microbial community is unable to recover from antibiotics, it can lead to increased susceptibility to gastrointestinal infections and increased risk of immunological and metabolic diseases. In this study, we work to better understand how the gut microbiota recovers from antibiotics by employing a recent technology to image the entire bacterial community at once. Through this approach, we characterize the spatial changes in the gut microbiota after treatment with model antibiotics in both the cecum and colon of mice. We find antibiotic- and biogeographic-dependent spatial changes between bacterial species and that many of these spatial colocalizations do not recover to baseline levels even 35 days after antibiotic administration.
Subject(s)
Anti-Bacterial Agents , Bacteria , Cecum , Colon , Gastrointestinal Microbiome , In Situ Hybridization, Fluorescence , Vancomycin , Animals , Gastrointestinal Microbiome/drug effects , Anti-Bacterial Agents/pharmacology , Mice , Bacteria/classification , Bacteria/genetics , Bacteria/drug effects , Bacteria/isolation & purification , Cecum/microbiology , Vancomycin/pharmacology , Colon/microbiology , Ampicillin/pharmacology , Mice, Inbred C57BL , PhylogenyABSTRACT
Longitudinal studies on the gut microbiome that follow the effect of a perturbation are critical in understanding the microbiome's response and succession to disease. Here, we use a dextran sodium sulfate (DSS) mouse model of colitis as a tractable perturbation to study how gut bacteria change their physiology over the course of a perturbation. Using single-cell methods such as flow cytometry, bioorthogonal noncanonical amino acid tagging (BONCAT), and population-based cell sorting combined with 16S rRNA sequencing, we determine the diversity of physiologically distinct fractions of the gut microbiota and how they respond to a controlled perturbation. The physiological markers of bacterial activity studied here include relative nucleic acid content, membrane damage, and protein production. There is a distinct and reproducible succession in bacterial physiology, with an increase in bacteria with membrane damage and diversity changes in the translationally active fraction, both, critically, occurring before symptom onset. Large increases in the relative abundance of Akkermansia were seen in all physiological fractions, most notably in the translationally active bacteria. Performing these analyses within a detailed, longitudinal framework determines which bacteria change their physiology early on, focusing therapeutic efforts in the future to predict or even mitigate relapse in diseases like inflammatory bowel diseases. IMPORTANCE Most studies on the gut microbiome focus on the composition of this community and how it changes in disease. However, how the community transitions from a healthy state to one associated with disease is currently unknown. Additionally, common diversity metrics do not provide functional information on bacterial activity. We begin to address these two unknowns by following bacterial activity over the course of disease progression, using a tractable mouse model of colitis. We find reproducible changes in gut bacterial physiology that occur before symptom onset, with increases in the proportion of bacteria with membrane damage, and changes in community composition of the translationally active bacteria. Our data provide a framework to identify possible windows of intervention and which bacteria to target in microbiome-based therapeutics.
ABSTRACT
From digestion to pathogen resistance and immune system development, the gut microbiota and its collection of microbial genes are redefining what it means to be human. Despite tremendous advances in this field, there is still a limited understanding of how microbial metabolism in the gut impacts human health, which precludes the development of microbiota-targeted therapies. In this article, we discuss the increasing evidence emphasizing the importance of bacterial metabolism in the gut and discuss its intricate links with diet and pharmaceutical compounds leading to altered therapeutic outcomes. We also detail how applying and testing microbial ecology hypotheses will be crucial to fully understand the therapeutic potential of this host-associated community. Going forward, functional and mechanistic studies combining biomedical research, ecology, bioinformatics, statistical modeling, and engineering will be key in our pursuit of personalized medicine.