PG also anchors other cell envelope components and intimately participates in cell growth and cell division processes [1]. Nevertheless, PG is also an Achilles’ heel for Bacteria, as some environmental organisms produce molecules that inhibit PG synthesis. The mold Penicillium notatum was shown by Alexander

Fleming to produce penicillin, a PG synthesis inhibitor and the first antibiotic used to treat bacterial infections in humans [30]. Vancomycin is another PG synthesis inhibitor produced by the soil bacterium Streptomyces orientalis[31]. However, PG is found in the vast majority of bacteria, including bacterial organisms living in the same niches as antibiotic-producing organisms. Accordingly, we observed that the absence of Cilengitide PG correlates with the intracellular life style and genome reduction [32]. In addition, free-living PG-less Bacteria and Archaea organisms use various osmoadapation strategies, such as the intracellular accumulation of inorganic ions, salt-tolerant enzymes or the accumulation of selected negative or neutral organic

molecules [33, 34] to maintain cell shape despite the absence of PG. Archaea cell walls could also contain other polymers, such as pseudomurein, methanochondroitin, KPT-8602 in vivo heterosaccharide and glutaminylglycan, participating in the mechanical strength of the cell wall [19]. Conclusions The exploration of PG in bacteria shows great heterogeneity in PG content. Genome analysis with ancestral reconstructions and phylogenetic comparative analyses offer a neutral tool to explore this heterogeneity and trace Acetophenone the evolutionary history of PG. These analyses also allowed the identification of genes that could be used to

predict functional features. Methods A1155463 Screening the CAZY database We extracted the GH23, GH73, GH102, GH103, GH10, GT28 and GT51 gene content for each genome available in CAZy in April 2011 [7], i.e., 1 398 Bacteria genomes distributed among 21 phyla, 42 Eukaryota genomes, 101 Archae genomes and 103 Viruses genomes. This database is updating regularly GenBank finished genomes for their content in carbohydrate active enzymes, providing with their EC number, gene name and product description. We then searched for the simultaneous presence of one GT28, one GT51 and at least one GH as evidence for PG metabolism. To assess the predictive value of this minimal 3-gene set, we correlated its bioinformatic detection with biological evidence for the presence of PG. We searched biological evidence for the presence of PG by screening Pubmed [35] using ‘peptidoglycan’, ‘cell wall’, ‘life style’ and the name of the genus as keywords. We further explored the HAMAP website [36], GenBank database [37] and Genome OnLine Database GOLD [38] for additional strain and genomic information.