Meta-mass shift chemical profiling of metabolomes from coral reefs

Significance Coral reef taxa produce a diverse array of molecules, some of which are important pharmaceuticals. To better understand how molecular diversity is generated on coral reefs, tandem mass spectrometry datasets of coral metabolomes were analyzed using a novel approach called meta-mass shift chemical (MeMSChem) profiling. MeMSChem profiling uses the mass differences between molecules in molecular networks to determine how molecules are related. Interestingly, the same molecules gain and lose chemical groups in different ways depending on the taxa it came from, offering a partial explanation for high molecular diversity on coral reefs. Untargeted metabolomics of environmental samples routinely detects thousands of small molecules, the vast majority of which cannot be identified. Meta-mass shift chemical (MeMSChem) profiling was developed to identify mass differences between related molecules using molecular networks. This approach illuminates metabolome-wide relationships between molecules and the putative chemical groups that differentiate them (e.g., H2, CH2, COCH2). MeMSChem profiling was used to analyze a publicly available metabolomic dataset of coral, algal, and fungal mat holobionts (i.e., the host and its associated microbes and viruses) sampled from some of Earth’s most remote and pristine coral reefs. Each type of holobiont had distinct mass shift profiles, even when the analysis was restricted to molecules found in all samples. This result suggests that holobionts modify the same molecules in different ways and offers insights into the generation of molecular diversity. Three genera of stony corals had distinct patterns of molecular relatedness despite their high degree of taxonomic relatedness. MeMSChem profiles also partially differentiated between individuals, suggesting that every coral reef holobiont is a potential source of novel chemical diversity.

[1]  R. Deberardinis,et al.  Cellular Metabolism and Disease: What Do Metabolic Outliers Teach Us? , 2012, Cell.

[2]  Stephen L. Johnson,et al.  After the feature presentation: technologies bridging untargeted metabolomics and biology. , 2014, Current opinion in biotechnology.

[3]  M. Bonn,et al.  Hydration strongly affects the molecular and electronic structure of membrane phospholipids. , 2012, The Journal of chemical physics.

[4]  R. Knight,et al.  Molecular cartography of the human skin surface in 3D , 2015, Proceedings of the National Academy of Sciences.

[5]  Russ Greiner,et al.  Competitive fragmentation modeling of ESI-MS/MS spectra for putative metabolite identification , 2013, Metabolomics.

[6]  Pieter C Dorrestein,et al.  Illuminating the dark matter in metabolomics , 2015, Proceedings of the National Academy of Sciences.

[7]  Forest Rohwer,et al.  Re-evaluating the health of coral reef communities: baselines and evidence for human impacts across the central Pacific , 2016, Proceedings of the Royal Society B: Biological Sciences.

[8]  Richard A. Berk,et al.  An Introduction to Ensemble Methods for Data Analysis , 2004 .

[9]  P. Falkowski,et al.  Membrane lipids of symbiotic algae are diagnostic of sensitivity to thermal bleaching in corals. , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[10]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[11]  Safwat Ahmed,et al.  Mechanism of action of antiepileptic ceramide from Red Sea soft coral Sarcophyton auritum. , 2015, Bioorganic & medicinal chemistry letters.

[12]  N. Carballeira,et al.  The Fatty Acid Composition of Tropical Marine Algae of the Genus Halimeda (Chlorophyta) , 1999 .

[13]  J. Lindon,et al.  Systems biology: Metabonomics , 2008, Nature.

[14]  Nuno Bandeira,et al.  Mass Spectrometry-Based Visualization of Molecules Associated with Human Habitats. , 2016, Analytical chemistry.

[15]  Florent E. Angly,et al.  Microbial Ecology of Four Coral Atolls in the Northern Line Islands , 2008, PloS one.

[16]  Juho Rousu,et al.  Metabolite identification and molecular fingerprint prediction through machine learning , 2012, Bioinform..

[17]  Kristian Fog Nielsen,et al.  Sharing and community curation of mass spectrometry data with Global Natural Products Social Molecular Networking , 2016, Nature Biotechnology.

[18]  N. Knowlton,et al.  Diversity and distribution of coral-associated bacteria , 2002 .

[19]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[20]  Ernest Fraenkel,et al.  Revealing disease-associated pathways by network integration of untargeted metabolomics , 2016, Nature Methods.

[21]  S. Böcker,et al.  Searching molecular structure databases with tandem mass spectra using CSI:FingerID , 2015, Proceedings of the National Academy of Sciences.

[22]  Richard D. Smith,et al.  Clustering millions of tandem mass spectra. , 2008, Journal of proteome research.

[23]  J. Navarro,et al.  Lipids of some Caribbean and Red Sea corals: total lipid, wax esters, triglycerides and fatty acids , 1993 .

[24]  P. Dorrestein,et al.  Biosynthetic origin of natural products isolated from marine microorganism–invertebrate assemblages , 2008, Proceedings of the National Academy of Sciences.

[25]  R. Knight,et al.  Mass Spectrometry Based Molecular 3D-Cartography of Plant Metabolites , 2017, Front. Plant Sci..

[26]  Joe Wandy,et al.  Topic modeling for untargeted substructure exploration in metabolomics , 2016, Proceedings of the National Academy of Sciences.

[27]  William H. Gerwick,et al.  Retrospective analysis of natural products provides insights for future discovery trends , 2017, Proceedings of the National Academy of Sciences.

[28]  P. Dorrestein,et al.  Correction to ‘Metabolomics of reef benthic interactions reveals a bioactive lipid involved in coral defence’ , 2016, Proceedings of the Royal Society B: Biological Sciences.