Characterization of a thermophilic acidogenic microbial community treating organic waste and producing the high commodity chemical n-caproate using a combination of culturomics and meta-omics

Proposed outline of the dissertation: This research aims to develop a biological model that explains the metabolic interactions in the biological anaerobic waste digestion reactor C1 of the MELiSSA system. This system, based on a lake ecosystem, aims at recycling waste by mainly microbial reactions, during long term space missions. Four main activities are performed to obtain a validated biological metabolic model of the complex microbial community of the C1 reactor. The work will start from background information about the dominant key stone populations present in the system under steady state performance. First, the effects of non-steady conditions (pH or temperature purtubations) on the C1 community composition are examined using metagenomics/metaproteomics. This will indicate whether or not other species have to be considered as key-stone stone besides those already known. Second, the meta-omic data from the first experiment, will be used to detail key-stone species/pathways/metabolic networks by cooccurrence analysis and key-stone individual genomes will be reconstructed. Third, the substrate specific role and metabolic functions of the key-stone species in C1 will be elucidated/verified/detailed by using time lapse 16S rRNA gene/transcript sequencing and DNA/protein Stable Isotope Probing after incubation with individual waste compounds and structures. Finally, the contribution of C1 keystone species to specific substrate conversion will be examined by studying their activities and interactions in artificial co-cultures after their (metagenome based) isolation from the complex community. Provisional description of the objective(s) of the doctoral dissertation (research plan), the methodology, timing and any previously accomplished research related to the present proposal: The Micro-Ecological Life Support System Alternative (MELiSSA) is a concept developed by the European Space Agency (ESA), that evolved out of the need for a regenerative life support system for long term space missions. The concept is inspired on a lake ecosystem and is conceived as a closed loop system consisting of 4 biological compartments (C1 and C4) that through combined activity of different organisms recycles organic waste to new food for the space crew (Hendrickx et al., 2006). The robustness of the MELiSSA loop relies on the building of robust, structured and predictive mathematical models which can only be implemented through a deep knowledge of the composition, behavior, metabolisms, kinetics, limitations, inhibitions, etc. of each subsystem. The C1 compartment is the first compartment in the cycle. A thermophilic anaerobic microbial consortium liquefies the solid waste and produces ammonium, volatile fatty acids (VFA’s), CO2 and minerals. The overall aim of this research is to obtain a validated biological metabolic model that describes the metabolic interactions within the complex microbial community of the MELiSSA waste degradation compartment C1. The research will start from available background information from the host lab that describes a preliminary microbial network reporting on the potential key-stone species and functionality (mainly based on high abundancy) in the C1 reactor at steady state performance deduced from metaomic data acquired from multiple reactor operations under standard conditions. More specifically, the research will validate and detail (1) this preliminary microbial network, (2) the corresponding metabolic pathways and (3) the identified key species/functions/biomarkers of C1. In order to reach these objectives, the following activities (as WPs) will be performed. In WP1, C1 reactors will be operated at a mode similar to this used to obtain the background information except for small deviations in the composition of the reactor feed thereby mimicking changes in the astronauts diet and waste (Poughon et al., 2013). Aside reactor performance parameters, the C1 microbial community composition will be monitored by metagenomics/metaproteomics. In other reactor experiments transient more drastic (pH or temperature) perturbations will be implemented. In addition, the community composition of a mother C1 reactor that aims at producing C1 reactor inoculum will be monitored to detail long-term community stability. These experiments will inform us about the dynamics and resilience of the C1 community upon minor and major perturbations. It will be examined whether the key-stone species and functionalities remain/reappear after perturbations and whether or not other species have to be considered as key-stone. Moreover, additional and needed sufficient sequence information will be gathered for detailed pathway and microbial/metabolic network analysis in WP2. In WP2, the meta-omic information obtained in WP1 and the background meta-omic information will be used to (1) further identify and detail key-stone species, (2) to reconstruct key-stone individual genomes using suitable metagenomic binning approaches (Kang et al., 2015) and (3) to detail keystone pathways/metabolic networks in the C1 community. Comparative and longitudinal meta-omic cooccurrence analysis will be performed to indicate other key-stone species additional to the highabundance species, i.e., non-dominant species that co-exist with the dominant species (Faust et al., 2012). Similarly, co-occurrence of functionality will be analyzed. Key to the success of co-occurrence analysis is the availability of sequence information on sufficient samples/replicates and the application of well-chosen perturbations/conditions (as performed in WP1). The annotation analysis of the individual key-stone species genomes and the mapping of the meta-proteomes on those will detail potential functions/activities/interactions of the key species/pathways going on in the C1 reactor and infer potential microbial networks and even additional key-stone species based on pathway complementation (Faust and Raes, 2012; Roume et al., 2015). WP3 will be used to further elucidate/verify/detail the substrate specific role and metabolic functions of the key-species in C1. In a first task, in small-scale batch experiments, the C1 community is fed with individual substrates present in the standard feed and time lapse 16S rRNA gene/transcript sequencing will pinpoint those key-stone species that proliferate indicating their role in the utilization of particular substrates. Tested substrates will be components of the standard feed like feces, salad, toilet paper as well as basic chemical structures like cellulose, glucose, individual VFAs, proteins, peptides etc. In a second task, the latter compounds are fed as 13C-labeled variants and time lapse DNA/protein Stable Isotope Probing (DNA/protein-SIP) (Jehmlich et al., 2016; Chen and Murrell, 2010) will be used to indicate those species and pathways (and the sequel) that actively convert the respective substrates. In parallel, community samples are subjected to a metabolomic analysis to indicate the presence/activity of metabolic pathways for the different substrates (Zimmermann et al., 2015; Bouvin et al., 2015). The DNA/protein SIP and metabolome information will be linked to deduce substrate-specific metabolic networks in C1. Since redundancy in function is expected, i.e., several species can be involved in conversion of the same structure, the respective contribution of some key-stone species to substrate conversion will be unknown. A first indication is provided by the relative abundance of the respective species in the 13CDNA and 13C-protein fraction, possible with a confirmation of intermediate metabolites from metabolome analysis, and their growth rates as monitored by targeted qPCR. Evidence will be obtained in WP4 by examining substrate conversion and growth rates of the key-stone species in isolation and within reconstructed artificial communities. To this end, key-stone species will be isolated using an innovative metagenomics based isolation approach, i.e., from the reconstructed genomes and potential metabolic activities (see WP2), media that selectively grow the key stone species are defined (Gutleben et al., 2017). This appears feasible since most of the dominant keystone species are Clostridia that were cultured before (Kersters et al., 1994). Alternatively, we will use similar (genomic based) organisms from culture collections. Together with information from WP2 and WP3, this information will lead to an interactive multispecies metabolic model of the C1 community.