Metagenomic systems biology of the human gut microbiome reveals topological shifts associated with obesity and inflammatory bowel disease

S. Greenblum, P. J. Turnbaugh, E. Borenstein
2011 Proceedings of the National Academy of Sciences of the United States of America  
The human microbiome plays a key role in a wide range of hostrelated processes and has a profound effect on human health. Comparative analyses of the human microbiome have revealed substantial variation in species and gene composition associated with a variety of disease states but may fall short of providing a comprehensive understanding of the impact of this variation on the community and on the host. Here, we introduce a metagenomic systems biology computational framework, integrating
more » ... omic data with an in silico systems-level analysis of metabolic networks. Focusing on the gut microbiome, we analyze fecal metagenomic data from 124 unrelated individuals, as well as six monozygotic twin pairs and their mothers, and generate community-level metabolic networks of the microbiome. Placing variations in gene abundance in the context of these networks, we identify both gene-level and network-level topological differences associated with obesity and inflammatory bowel disease (IBD). We show that genes associated with either of these host states tend to be located at the periphery of the metabolic network and are enriched for topologically derived metabolic "inputs." These findings may indicate that lean and obese microbiomes differ primarily in their interface with the host and in the way they interact with host metabolism. We further demonstrate that obese microbiomes are less modular, a hallmark of adaptation to low-diversity environments. We additionally link these topological variations to community species composition. The system-level approach presented here lays the foundation for a unique framework for studying the human microbiome, its organization, and its impact on human health. W e humans are mostly microbes. Microbial communities populate numerous sites in the human anatomy and harbor over 100 trillion microbial cells (1). This complex ensemble of microorganisms, collectively known as the human microbiome, plays an essential role in our development, immunity, and nutrition, and has a tremendous impact on our health (2). Among the various body habitats, the most densely colonized is the distal gut. The normal gut flora alone consists of hundreds of bacterial species, collectively encoding an enormous gene set that is 150-fold larger than the set of human genes (3). The gut microbiome plays a key role in many essential processes, including vitamin and amino acid biosynthesis, dietary energy harvest, and immune development (4). Transferring a donor microbiota into a recipient can induce various donor phenotypes [including increased adiposity (5) and metabolic syndrome (6)] or prompt the recovery of a sick recipient (7), suggesting a promising avenue for clinical application via directed manipulation of the microbiome. Characterizing the capacity of the human microbiome, its interaction with the host, and its contribution to various disease states therefore has the potential to provide deep insight into both normal human physiology and human disease, and calls for a predictive systemslevel understanding of community function and structure. Addressing this challenge, worldwide research initiatives (3, 4) have recently started to map the human microbiome, providing insight into previously uncharted species and genes. Specifically, sequencing 16S ribosomal RNA allows researchers to determine the relative abundance of different taxonomic groups in a microbiome (8, 9) . Such surveys have revealed, for example, marked associations between the species composition of the gut microbiome and a variety of host phenotypes (10-12). Species profiles, however, cannot be easily translated into function, because it is not clear how variation in the composition of species in the microbiome affects the metabolic activity of the community and, consequently, the host. In contrast, metagenomic shotgun sequencing of community DNA and a gene-centric comparative approach (8, 13, 14) may capture functional differences in the metabolic potential of the community. Yet, comparative metagenomic analysis of the gut microbiome frequently reveals high functional uniformity across samples and often identifies only a small set of genes or pathways that appear to be associated with certain host states (10, 15). Furthermore, such enriched sets offer preliminary insights into relevant functional differences but may not provide a comprehensive systems-level understanding of the variation and its potential effect on the host-microbiome supraorganism (16, 17). Here, we introduce a unique framework for studying the human microbiome, integrating metagenomic data with a systems-level network analysis. This metagenomic systems biology approach goes beyond traditional comparative analysis, placing shotgun metagenomic data in the context of community-level metabolic networks. Comparing the topological properties of the enzymes in these networks with their abundances in different metagenomic samples and examining systems-level topological features of microbiomes associated with different host states allow us to obtain insight into variation in metabolic capacity. This approach extends the metagenomic gene-centric view by taking into account not only the set of genes present in a microbiome but also the complex web of interactions among these genes and by treating the microbiome as a single "independent" biological system (18). Computational systems biology methods and complex network analyses have been applied widely to study microorganisms, and a variety of approaches have been developed to create genomescale metabolic networks of various microbial species (19) (20) (21) . In this study, we focus on simple connectivity-centered networks that are computationally derived from homology-based large-scale metabolic databases (22) coupled with a topological analysis. These networks form a simplification of the actual underlying metabolic pathways and may be relatively inaccurate and noisy. However, topology-based analysis of such networks has proved powerful for studying the characteristics of single-species metabolic networks and their impact on various functional and evolutionary properties, including scaling (23), metabolic functionality and
doi:10.1073/pnas.1116053109 pmid:22184244 pmcid:PMC3258644 fatcat:nvhch3yd3fet3bfhbdgtjvvrpm