Quantitative Metagenomics: use of bacterial strains as internal standards for the analysis of faecal microbials.

Recent technological advances (next-generation sequencing) have enabled new approaches in microbial ecosystem research, including metagenomics: random shotgun sequencing of total DNA from environmental or clinical samples. These developments provided the opportunity to study and monitor various microbial communities associated to the human body at a previously unseen scale, allowing the phylogenetic assessment of the uncultivable fraction of microbial communities and enabling functional insights, both at the level of the individual microorganisms and of the ecosystem as a whole. Metagenomic analyses have linked alterations in the gut microbiota to pathological states such as (low-grade) inflammation, infections, autoimmune diseases, multifactorial disorders, and cancer. Moreover, evidence has been found that gut microbiota composition can serve as an indicator of chronic suboptimal health and wellbeing, either directly linked to suboptimal bowel functioning or extended to general physical and psychological health. The establishment of the association between colon microbiota dysbiosis and disease has caused tremendous excitement in the fields of microbiology and clinical diagnostics. Several research groups rushed forward in order to be the first to discover a specific microbiomebased biomarker or signature linked to the particular disease of their preference. Colon microbiome research has thus witnessed a plethora of disease-focused, medium-scale studies in the last few years, all attempting to establish a firm link between particular species, taxa, bacterial genes, and metabolic pathways and the targeted health condition. Obviously, such studies are indispensible for the further exploration of the colon ecosystem and its impact on host health status and vice versa. However, the excitement concerning the wealth of information unlocked by the introduction of metagenomics in gut microbiota research and the rush toward high-impact publications has left some fundamental issues in microbial ecosystem research unresolved. One of these issues concerns the lack of truly quantitative approaches in metagenomic community analysis. Current metagenomic analyses of the human colon microbiota collect data on relative species abundances and their joint functionalities, in which the abundance of microbial clusters or metabolic pathways is calculated as a fraction of the total sequence library obtained. A critical limitation of this relative approach is that it cannot provide information on the extent or directionality of changes in bacterial abundance or metabolic potential in a comparative analysis. For example, dysbiosis associated to obesity-induced, low-grade inflammation is characterized by a significant reduction of the relative abundance of two dominant bacterial phyla in the gut, namely Firmicutes and Bacteroidetes . However, it is currently impossible to link this observation to either reduced numbers of Firmicutes or blooming of Bacteroidetes. It is clear that a quantitative assessment of the nature of the microbial shift observed is crucial for the further clarification of the link between microbiota and host health. A fundamental challenge in present-day metagenomic colon ecosystem research is to quantitatively assess variation in microbiota composition and metabolic potential across individuals, over time, associated to a diseased or healthy state, or resulting from disease treatment. The predictive/diagnostic potential of metagenomic analyses of body-associated microbial ecosystems and the translation of the technique into medical practice are currently hampered by the explorative/qualitative nature of the methods applied. Only the development of quantitative metagenomic techniques will allow to overcome these hurdles and to exploit the full potential of microbiome research in reproducible and truly quantitative diagnostics Research objectives As sequencing technology nowadays allows the sequencing more than 100 gigabases/day , large-scale characterization of the body-associated microbiota of thousands of individuals has become feasible. However, despite rapid result-driven progress in metagenomics and sequencing approaches, some rather fundamental issues related to technique optimization and adaptation to the specific needs of an ecological context have remained unresolved. Notwithstanding the undeniable potential of the research conducted up to now and the high impact publications resulting from it, it is remarkable that a truly quantitative metagenomic assessment of microbial ecosystems is until today not feasible. Current metagenomic sequencing workflows are fundamentally biased on various levels. It is, for example, far too often assumed that different bacterial taxa present in an environmental sample are all equally susceptible to cell lysis during the community DNA extraction process. Also, to improve sequencing efficiency, preparative steps (i.e. library preparation) in shotgun metagenomics require dilution of sample total DNA to optimal sequencing concentrations and/or equimolar pooling of DNA extracted from multiple samples. Consequently, metagenomes obtained from high microbial density versus low density samples result in roughly comparable output in terms of numbers of reads sequenced, although their respective sequencing depth should vary considerably. As a result of losing track of the communities’ initial densities, the outcome of such analyses is not a census: it only reflects relative species composition within samples, but never their true, quantitative abundance. The human colon ecosystem is considered one of the most densely populated bacterial communities currently described. Bacterial density in the large intestine fluctuates around 1011 bacteria/g - ranging from 1010 to more than 5.1011 – depending on the individual and time of sampling. This implies that one individual might only harbor around 1% of the amount of bacteria present in another. The implications of this variation in total bacterial abundance are currently underestimated and will undoubtedly affect any hypothesis formulated regarding the impact of gut microbiota variability on host health. The current proposal aims at redirecting microbiome research to a truly quantitative approach by implementing internal standard (IS) based normalization in metagenomic analyses. In short, fully sequenced bacterial strains that have not been reported to occur in fecal material will be added in known concentrations to samples before DNA extraction. Through the assessment of IS sequencing depth, this technique allows to extract absolute instead of relative counts from metagenomic datasets - hence enabling true quantitative metagenomics