Metafoldscan aims at developing a user-friendly interface -experimentally validated- to scan metagenomes and identify fold hits associated to a target protein structurally characterized and with an identified activity. The first ecological system to be scanned will be the human gut microbiota and the first target proteins will be a protein from the commensal bacterium Faecalibacterium prausnitzii called MAM- Microbial Anti-inflammatory Molecule- and the ubiquitous bacterial protein Mfd - Mutation Frequency Decline-, they both associate biological with therapeutic relevance.
Human gut microbiota clusters at least 1,000 different species of known bacteria that attain more than 3 million genes. A balanced composition of microbiota associates with healthy patients whilst reduction in bacterial biodiversity results in dysbiosis and correlates with defective host-microbial interactions and diseases. Among adequate host-bacteria interactions, Faecalibacterium prausnitzii accounts for up to 5% of the microbiota. Notably, this firmicute was recently shown by our partner 1 to exert anti-inflammatory effects in vitro and in vivo by secreted peptides that block nuclear factor NF-kB activation. These bioactive peptides derive from a secreted 15kDa protein called MAM. MAM blocks the NF-kB pathway in vitro and protect mice from chemically-induced colitis. The identification of active molecules involved in the protective effect is of particular interest since F. prausnitzii remains hardly cultivable due to its extreme oxygen sensitivity. Thus, finding other secreted biomolecules with similar anti-inflammatory activity is highly valuable. Conversely, among abnormal host-bacteria interactions, secretion of reactive chemicals such as nitric oxide (NO) induces bacteria DNA damage and gene mutation. To combat host interference into their gene transcription, bacteria possess mechanisms to limit the negative effects where Mfd (Mutation Frequency Decline) plays a pivotal role as it is an evolutionarily-conserved protein that was shown to mediate DNA. Freshly, a role for Mfd in bacterial virulence was reported by our partner 2 who showed that Mfd is required for survival of pathogen bacteria in their host, by protecting them against DNA damage following NO exposure. As Mfd is widely conserved in the bacterial kingdom, these data highlight a mechanism that may be used by a large spectrum of bacteria to overcome the host immune response. Such a finding may allow the development of new and hopefully universal antimicrobial strategies.
MetaFoldScan objectives are:
to build and make available a versatile software to detect structural homologs and accordingly biologically relevant enzymes in ecological system such as human gut microbiota. To our knowledge, no comprehensive solution with flexible and robust browsing of genomes is available. MetaFoldScan is dedicated to allow intensive and robust screening of meta-data to identify structural homologs of target proteins. To that end, the key features are its capacity to browse and select multiple genomes, to adapt a flexible strategy to incorporate structural template(s), to compute high throughput screening and to assemble this pipeline with accurate analysis of hits that associates with rigorous scoring parameters.
to validate the hits from meta-screening of MAM- and Mfd-like folds ie. to validate their biological function, including their probiotic power and possibly to solve their experimental structure.
to achieve innovation and originality with the integration of in silico suite of tools set up in 1) into the user-friendly Galaxy portal. This clearly includes to browse and filter upstream meta-omics data and to step further towards downstream analysis of structural hits with connection to web servers for homology modelling, docking of ligands, or gene co-regulation informations.
In summary, the key issues and challenges are:
Set-up of a suite of functions to scan intensively metagenomes to get structural hits of relevant targets and compute functional annotations.
Validation of the hits and discovery of new enzymes with possibly therapeutic functions.
Integration of the suite of functions into a user-friendly interface implemented in Galaxy.
MaIAGE: Véronique Martin, Sandra Dérozier, Valentin Loux & Gwenaëlle André-Leroux
Micalis: Nalini Ramarao, Jean-Marc Chatel.