The potential of the diverse chemistries present in natural products (NP) for biotechnology and medicine remains untapped because NP databases are not searchable with raw data and the NP community has no way to share data other than in published papers. Although mass spectrometry techniques are well-suited to high-throughput characterization of natural products, there is a pressing need for an infrastructure to enable sharing and curation of data. We present Global Natural Products Social molecular networking (GNPS, http://gnps.ucsd.edu), an open-access knowledge base for community wide organization and sharing of raw, processed or identified tandem mass (MS/MS) spectrometry data. In GNPS crowdsourced curation of freely available community-wide reference MS libraries will underpin improved annotations. Data-driven social-networking should facilitate identification of spectra and foster collaborations. We also introduce the concept of ‘living data’ through continuous reanalysis of deposited data.

During bacterial cannibalism, a differentiated subpopulation harvests nutrients from their genetically identical siblings to allow continued growth in nutrient-limited conditions. Hypothesis-driven imaging mass spectrometry (IMS) was used to identify metabolites active in a Bacillus subtilis cannibalism system in which sporulating cells lyse nonsporulating siblings. Two candidate molecules with sequences matching the products of skfA and sdpC, genes for the proposed cannibalistic factors sporulation killing factor (SKF) and sporulation delaying protein (SDP), respectively, were identified and the structures of the final products elucidated. SKF is a cyclic 26-amino acid (aa) peptide that is posttranslationally modified with one disulfide and one cysteine thioether bridged to the α-position of a methionine, a posttranslational modification not previously described in biology. SDP is a 42-residue peptide with one disulfide bridge. In spot test assays on solid medium, overproduced SKF and SDP enact a cannibalistic killing effect with SDP having higher potency. However, only purified SDP affected B. subtilis cells in liquid media in fluorescence microscopy and growth assays. Specifically, SDP treatment delayed growth in a concentration-dependent manner, caused increases in cell permeability, and ultimately caused cell lysis accompanied by the production of membrane tubules and spheres. Similarly, SDP but not SKF was able to inhibit the growth of the pathogens Staphylococcus aureus and Staphylococcus epidermidis with comparable IC 50 to vancomycin. This investigation, with the identification of SKF and SDP structures, highlights the strength of IMS in investigations of metabolic exchange of microbial colonies and also demonstrates IMS as a promising approach to discover novel biologically active molecules.etabolic exchange describes the process of exchanging signals or nutrients between cells or populations and is a common feature of all living systems. Bacteria produce a wide array of signaling molecules to control metabolic as well as morphological and developmental changes in either an interspecies or intraspecies manner (1). Bacillus subtilis, for example, has a complex life cycle and thrives in diverse living conditions ranging from soil, contaminated wounds, and the intestinal tract (2-4). To accommodate this, B. subtilis dedicates ∼10% of its genome to the production of specific molecules involved in intra-and interspecies metabolic exchange (5). Two of these molecules are sporulation delaying protein (SDP) and sporulation killing factor (SKF), which, based on genetic experiments, are proposed to lyse a subpopulation of B. subtilis cells to provide nutrients for the remaining cells, a process referred to as bacterial cannibalism (6-10). This behavior is dependent on Spo0A, a master transcriptional regulator that also controls biofilm formation and sporulation (6-13).We set out to characterize these cannibalistic compounds to establish their roles in the B. subtilis life cycle and to understand their structure ...

The ability to correlate the production of specialized metabolites to the genetic capacity of the organism that produces such molecules has become an invaluable tool in aiding the discovery of biotechnologically applicable molecules. Here, we accomplish this task by matching molecular families with gene cluster families, making these correlations to 60 microbes at one time instead of connecting one molecule to one organism at a time, such as how it is traditionally done. We can correlate these families through the use of nanospray desorption electrospray ionization MS/MS, an ambient pressure MS technique, in conjunction with MS/MS networking and peptidogenomics. We matched the molecular families of peptide natural products produced by 42 bacilli and 18 pseudomonads through the generation of amino acid sequence tags from MS/MS data of specific clusters found in the MS/MS network. These sequence tags were then linked to biosynthetic gene clusters in publicly accessible genomes, providing us with the ability to link particular molecules with the genes that produced them. As an example of its use, this approach was applied to two unsequenced Pseudoalteromonas species, leading to the discovery of the gene cluster for a molecular family, the bromoalterochromides, in the previously sequenced strain P. piscicida JCM 20779 T . The approach itself is not limited to 60 related strains, because spectral networking can be readily adopted to look at molecular family-gene cluster families of hundreds or more diverse organisms in one single MS/MS network.MS/MS molecular networking | mass spectrometry | microbial ecology T ens of thousands of sequenced microbial genomes or rough drafts of genomes are available at this time, and this number is predicted to grow into the millions over the next decades. This wealth of sequence data has the potential to be used for the discovery of small bioactive molecules through genome mining (1-6). Genome mining is a process in which small molecules are discovered by predicting what compound will be genetically encoded based on the sequences of biosynthetic gene clusters. However, the process of mining genetically encoded small molecules is not keeping pace with the rate by which genome sequences are being obtained. In general, genome mining is still done one gene cluster at a time and requires many person-years of effort to annotate a single molecule. The time and significant expertise that current genome mining requires also make genome mining very expensive. In light of this extensive effort and cost, alternative approaches to genome mining and annotating specialized metabolites must be developed that not only take advantage of the sequenced resources available and make it efficient to perform genome mining on a more global scale but also enable the molecular analysis of unsequenced organisms. Such methods will then significantly reduce the cost of genome mining by increasing the speed with which molecules are connected to candidate genes and using resources already available. Here, we put fo...

Proteins in the YidC/Oxa1/Alb3 family have essential functions in membrane protein insertion and folding. Bacillus subtilis encodes two YidC homologs, one that is constitutively expressed (spoIIIJ/yidC1) and a second (yqjG/yidC2) that is induced in spoIIIJ mutants. Regulated induction of yidC2 allows B. subtilis to maintain capacity of the membrane protein insertion pathway. We here show that a gene located upstream of yidC2 (mifM/yqzJ) serves as a sensor of SpoIIIJ activity that regulates yidC2 translation. Decreased SpoIIIJ levels or deletion of the MifM transmembrane domain arrests mifM translation and unfolds an mRNA hairpin that otherwise blocks initiation of yidC2 translation. This regulated translational arrest and yidC2 induction require a specific interaction between the MifM C-terminus and the ribosomal polypeptide exit tunnel. MifM therefore acts as a ribosome-nascent chain complex rather than as a fully synthesized protein. B. subtilis MifM and the previously described secretion monitor SecM in Escherichia coli thereby provide examples of the parallel evolution of two regulatory nascent chains that monitor different protein export pathways by a shared molecular mechanism.