Gene Content of Seawater Microbes is a Strong Predictor of Water Chemistry Across the Great Barrier Reef
PhD candidate Marko Terzin from James Cook University and the Australian Institute of Marine Science (AIMS) has spent a few months in 2023 in our lab to complement his learning about mixOmics techniques to analyse his data. He is back this year to finish off his PhD.
In this study, Marko analysed seawater metagenomics samples to better understand the Great Barrier Reef health. The study is a great example of application of MINT to integrate and accommodate for unwanted variation of four sampling trips across different seasons.
- We integrated physico-chemical variables with microbial functions and environmental parameters using sPLS and MINT
- We identified robust seawater microbial biomarkers indicative of reef health
- We identified distinct seasonal variations in surface water chemistry with particulate organic matter produced by season-specific microorganisms
Gene Content of Seawater Microbes is a Strong Predictor of Water Chemistry Across the Great Barrier Reef (2025). Marko Terzin, Steven J. Robbins, Sara C. Bell, Kim-Anh Lê Cao, Renee K. Gruber, Pedro R. Frade, Nicole S. Webster, Yun Kit Yeoh, David G. Bourne, Patrick W. Laffy. Accepted in Microbiome.
Abstract
Background. Seawater microbes (bacteria and archaea) play essential roles in coral reefs by facilitating nutrient cycling, energy transfer, and overall reef ecosystem functioning. However, environmental disturbances such as degraded water quality and marine heatwaves, can impact these vital functions as seawater microbial communities experience notable shifts in composition and function when exposed to stressors. This sensitivity highlights the potential of seawater microbes to be used as indicators of reef health. Microbial indicator analysis has centred around measuring the taxonomic composition of seawater microbial communities, but this can obscure heterogeneity of gene content between taxonomically similar microbes, and thus microbial functional genes have been hypothesised to have more scope for predictive potential, though empirical validation for this hypothesis is still pending. Here, we establish a functional baseline of seawater microbiomes across outer Great Barrier Reef (GBR) sites to compare the diagnostic value between taxonomic and functional information in inferring continuous physico-chemical metrics in the surrounding reef.
Results. Integrating gene-centric analyses with 17 physico-chemical variables (temperature, salinity, and particulate and dissolved nutrients) across 48 reefs revealed that associations between microbial functions and environmental parameters were twice as stable compared to taxonomy-environment associations. Distinct seasonal variations in surface water chemistry were observed, with nutrient concentrations up to 3-fold higher during austral summer explained by enhanced production of particulate organic matter (POM) primarily by Synechococcus , whereas in winter, nutrient levels were lower and POM production was also attributed to Pr ochlor ococcus . Additionally, heterotrophic microbes (e.g., Rhodospirillaceae , Burkholderiaceae , Fla v obacteriaceae , and Rhodobacter aceae ) were enriched in reefs with elevated dissolved organic carbon (DOC) and phytoplankton-derived POM, encoding functional genes related to membrane transport, sugar utilisation, and energy metabolism. These microbes likely contribute to the coral reef microbial loop by capturing and recycling nutrients derived from Synechococcus and Pr ochlor ococcus , ultimately transferring nutrients from picocyanobacterial primary producers to higher trophic levels.
Conclusion. This study reveals that functional information in reef-associated seawater microbes robustly associates with physico-chemical variables than taxonomic data, highlighting the importance of incorporating microbial function in reef monitoring initiatives. Our integrative approach to mine for stable seawater microbial biomarkers can be expanded to include additional continuous metrics of reef health (e.g., benthic cover of corals and macroalgae, scale reef metagenomics datasets beyond the GBR.