Multi-omics eco-surveillance of bacterial community function in legacy contaminated estuary sediments

Abstract

Complex legacy contamination is a major issue for many estuaries, with toxicity affecting change in bacterial communities, and their provision of associated goods and services. Sequencing surveys of bacterial community composition provide inferred function; however, additional insights may be generated by measurement of realised metabolic phenotypes. We apply multi-omics (genomics, lipidomics, and metabolomics), with traditional sediment quality analyses, to characterise sediment-associated bacterial communities in an estuary subject to legacy metal contamination (Zn, Hg, As, Cd, Cu and Pb). Analyses of bacterial composition and inferred function (genomics) are coupled with measurements of realised bacterial phenotype (metabolomics and lipidomics) at multiple industrialised and reference sites. At sites with the highest sediment metal concentrations (NTB), we also observed increased abundances of hydrocarbon and sulphuric acid metabolites, indicating additional sediment contamination. Bacterial phyla across sampled sites were dominated by Proteobacteria and Desulfobacteria. NTB sites were enriched with metabolically versatile, cooperative and biofilm forming phyla including, Zixibacteria, Spirochaetota, SAR324 clade, Proteobacteria, Latescibacterota, Desulfobacterota, Deferrisomtota and Acidobateriota; with inferred functions characterised by sulphur metabolism, pathways associated with the degradation of complex organic molecules, and fermentation. Reference sites were characterised by enhanced vitamin biosynthesis, cell wall, cofactor and carbohydrate biosynthesis, and CO2 fixation. Measured metabolic phenotypes at NTB sites supported predicted functions, with most consistent change observed to naphthalene and aminobenzoate degradation pathways and carbohydrate metabolism (galactose, amino and nucleotide sugar). Change in NTB metabolite profiles was most highly correlated with sediment Hg concentrations, indicative of toxic exposure and potential for Hg methylation. Lipid profiles generated further insight into potential functional (hydroxy fatty acids) and community level change (ceramide phosphoethanolamines, unsaturated glycerides). Multi-omics outputs provided insights into bacterial community functions, modes of contaminant toxicity and expressed mechanisms of adaptation, necessary to better inform management decisions and predictive models in increasingly human-influenced environments.