Person:

Gerber, Georg

The human gut microbiota comprise a complex and dynamic ecosystem that profoundly affects host development and physiology. Standard approaches for analyzing time-series data of the microbiota involve computation of measures of ecological community diversity at each time-point, or measures of dissimilarity between pairs of time-points. Although these approaches, which treat data as static snapshots of microbial communities, can identify shifts in overall community structure, they fail to capture the dynamic properties of individual members of the microbiota and their contributions to the underlying time-varying behavior of host ecosystems. To address the limitations of current methods, we present a computational framework that uses continuous-time dynamical models coupled with Bayesian dimensionality adaptation methods to identify time-dependent signatures of individual microbial taxa within a host as well as across multiple hosts. We apply our framework to a publicly available dataset of 16S rRNA gene sequences from stool samples collected over ten months from multiple human subjects, each of whom received repeated courses of oral antibiotics. Using new diversity measures enabled by our framework, we discover groups of both phylogenetically close and distant bacterial taxa that exhibit consensus responses to antibiotic exposure across multiple human subjects. These consensus responses reveal a timeline for equilibration of sub-communities of micro-organisms with distinct physiologies, yielding insights into the successive changes that occur in microbial populations in the human gut after antibiotic treatments. Additionally, our framework leverages microbial signatures shared among human subjects to automatically design optimal experiments to interrogate dynamic properties of the microbiota in new studies. Overall, our approach provides a powerful, general-purpose framework for understanding the dynamic behaviors of complex microbial ecosystems, which we believe will prove instrumental for future studies in this field.

The gut microbiome plays an important role in immune function and has been implicated in several autoimmune disorders. Here we use 16S rRNA sequencing to investigate the gut microbiome in subjects with multiple sclerosis (MS, n=60) and healthy controls (n=43). Microbiome alterations in MS include increases in Methanobrevibacter and Akkermansia and decreases in Butyricimonas, and correlate with variations in the expression of genes involved in dendritic cell maturation, interferon signalling and NF-kB signalling pathways in circulating T cells and monocytes. Patients on disease-modifying treatment show increased abundances of Prevotella and Sutterella, and decreased Sarcina, compared with untreated patients. MS patients of a second cohort show elevated breath methane compared with controls, consistent with our observation of increased gut Methanobrevibacter in MS in the first cohort. Further study is required to assess whether the observed alterations in the gut microbiome play a role in, or are a consequence of, MS pathogenesis.

Longitudinal studies of the microbiota are important for discovering changes in microbial communities that affect the host. The complexity of these ecosystems requires rigorous integrated experimental and computational methods to identify temporal signatures that promote physiologic or pathophysiologic responses in vivo. Employing a murine model of infectious colitis with the pathogen Citrobacter rodentium, we generated a 2-month time-series of 16S rDNA gene profiles, and quantitatively cultured commensals, from multiple intestinal sites in infected and uninfected mice. We developed a computational framework to discover time-varying signatures for individual taxa, and to automatically group signatures to identify microbial sub-communities within the larger gut ecosystem that demonstrate common behaviors. Application of this model to the 16S rDNA dataset revealed dynamic alterations in the microbiota at multiple levels of resolution, from effects on systems-level metrics to changes across anatomic sites for individual taxa and species. These analyses revealed unique, time-dependent microbial signatures associated with host responses at different stages of colitis. Signatures included a Mucispirillum OTU associated with early disruption of the colonic surface mucus layer, prior to the onset of symptomatic colitis, and members of the Clostridiales and Lactobacillales that increased with successful resolution of inflammation, after clearance of the pathogen. Quantitative culture data validated findings for predominant species, further refining and strengthening model predictions. These findings provide new insights into the complex behaviors found within host ecosystems, and define several time-dependent microbial signatures that may be leveraged in studies of other infectious or inflammatory conditions.

Elucidating functions of commensal microbial genes in the mammalian gut is challenging because many commensals are recalcitrant to laboratory cultivation and genetic manipulation. We present Temporal FUnctional Metagenomics sequencing (TFUMseq), a platform to functionally mine bacterial genomes for genes that contribute to fitness of commensal bacteria in vivo. Our approach uses metagenomic DNA to construct large-scale heterologous expression libraries that are tracked over time in vivo by deep sequencing and computational methods. To demonstrate our approach, we built a TFUMseq plasmid library using the gut commensal Bacteroides thetaiotaomicron (Bt) and introduced Escherichia coli carrying this library into germfree mice. Population dynamics of library clones revealed Bt genes conferring significant fitness advantages in E. coli over time, including carbohydrate utilization genes, with a Bt galactokinase central to early colonization, and subsequent dominance by a Bt glycoside hydrolase enabling sucrose metabolism coupled with co-evolution of the plasmid library and E. coli genome driving increased galactose utilization. Our findings highlight the utility of functional metagenomics for engineering commensal bacteria with improved properties, including expanded colonization capabilities in vivo.

An important research problem in computational biology is the identification of expression programs, sets of co-expressed genes orchestrating normal or pathological processes, and the characterization of the functional breadth of these programs. The use of human expression data compendia for discovery of such programs presents several challenges including cellular inhomogeneity within samples, genetic and environmental variation across samples, uncertainty in the numbers of programs and sample populations, and temporal behavior. We developed GeneProgram, a new unsupervised computational framework based on Hierarchical Dirichlet Processes that addresses each of the above challenges. GeneProgram uses expression data to simultaneously organize tissues into groups and genes into overlapping programs with consistent temporal behavior, to produce maps of expression programs, which are sorted by generality scores that exploit the automatically learned groupings. Using synthetic and real gene expression data, we showed that GeneProgram outperformed several popular expression analysis methods. We applied GeneProgram to a compendium of 62 short time-series gene expression datasets exploring the responses of human cells to infectious agents and immune-modulating molecules. GeneProgram produced a map of 104 expression programs, a substantial number of which were significantly enriched for genes involved in key signaling pathways and/or bound by NF-κB transcription factors in genome-wide experiments. Further, GeneProgram discovered expression programs that appear to implicate surprising signaling pathways or receptor types in the response to infection, including Wnt signaling and neurotransmitter receptors. We believe the discovered map of expression programs involved in the response to infection will be useful for guiding future biological experiments; genes from programs with low generality scores might serve as new drug targets that exhibit minimal “cross-talk,” and genes from high generality programs may maintain common physiological responses that go awry in disease states. Further, our method is multipurpose, and can be applied readily to novel compendia of biological data.

Predicting dynamics of host-microbial ecosystems is crucial for the rational design of bacteriotherapies. We present MDSINE, a suite of algorithms for inferring dynamical systems models from microbiome time-series data and predicting temporal behaviors. Using simulated data, we demonstrate that MDSINE significantly outperforms the existing inference method. We then show MDSINE’s utility on two new gnotobiotic mice datasets, investigating infection with Clostridium difficile and an immune-modulatory probiotic. Using these datasets, we demonstrate new capabilities, including accurate forecasting of microbial dynamics, prediction of stable sub-communities that inhibit pathogen growth, and identification of bacteria most crucial to community integrity in response to perturbations. Electronic supplementary material The online version of this article (doi:10.1186/s13059-016-0980-6) contains supplementary material, which is available to authorized users.

Manipulation of the gut microbiota holds great promise for the treatment of diseases. However, a major challenge is the identification of therapeutically potent microbial consortia that colonize the host effectively while maximizing immunologic outcome. Here, we propose a novel workflow to select optimal immune-inducing consortia from microbiome compositicon and immune effectors measurements. Using published and newly generated microbial and regulatory T-cell (Treg) data from germ-free mice, we estimate the contributions of twelve Clostridia strains with known immune-modulating effect to Treg induction. Combining this with a longitudinal data-constrained ecological model, we predict the ability of every attainable and ecologically stable subconsortium in promoting Treg activation and rank them by the Treg Induction Score (TrIS). Experimental validation of selected consortia indicates a strong and statistically significant correlation between predicted TrIS and measured Treg. We argue that computational indexes, such as the TrIS, are valuable tools for the systematic selection of immune-modulating bacteriotherapeutics.

Insulin receptors (IRs) and IGF-I receptors (IGF-IR) are major regulators of metabolism and cell growth throughout the body; however, their roles in the intestine remain controversial. Here we show that genetic ablation of the IR or IGF-IR in intestinal epithelial cells of mice does not impair intestinal growth or development or the composition of the gut microbiome. However, the loss of IRs alters intestinal epithelial gene expression, especially in pathways related to glucose uptake and metabolism. More importantly, the loss of IRs reduces intestinal glucose uptake. As a result, mice lacking the IR in intestinal epithelium retain normal glucose tolerance during aging compared with controls, which show an age-dependent decline in glucose tolerance. Loss of the IR also results in a reduction of glucose-dependent insulinotropic polypeptide (GIP) expression from enteroendocrine K-cells and decreased GIP release in vivo after glucose ingestion but has no effect on glucagon-like peptide 1 expression or secretion. Thus, the IR in the intestinal epithelium plays important roles in intestinal gene expression, glucose uptake, and GIP production, which may contribute to pathophysiological changes in individuals with diabetes, metabolic syndrome, and other insulin-resistant states.