Identification and implications of a core bacterial microbiome in 19 clonal cultures laboratory-reared for months to years of the cosmopolitan dinoflagellate .

Identification of a core microbiome (a group of taxa commonly present and consistently abundant in most samples of host populations) is important to capture the key microbes closely associated with a host population, as this process may potentially contribute to further revealing their spatial distribution, temporal stability, ecological influence, and even impacts on their host's functions and fitness. The naked dinoflagellate is a cosmopolitan and toxic species, which is also notorious in forming harmful algal blooms (HABs) and causing massive fish-kills. Here we reported the core microbiome tightly associated with 19 strains of that were originally isolated from 6 geographic locations along the coast of China and from an estuary of Chesapeake Bay, United States, and have been maintained in the laboratory for several months to over 14 years. Using high-throughput metabarcoding of the partial 16S rRNA gene amplicons, a total of 1,417 prokaryotic features were detected in the entire bacterial microbiome, which were assigned to 17 phyla, 35 classes, 90 orders, 273 families, and 716 genera. Although the bacterial communities associated with cultures displayed heterogeneity in feature (sequences clustered at 100% sequence similarity) composition among strains, a core set of 6 genera were found persistent in their phycospheres, which could contribute up to 74.54% of the whole bacterial microbiome. Three γ-proteobacteria members of the "core," namely, , , and , were the predominant core genera and made up 83.25% of the core bacterial microbiome. The other 3 core genera, , , and , are reported to preferably utilize hydrocarbons as sole or major source of carbon and energy, and two of which ( and ) are recognized as obligate hydrocarbonoclastic bacteria (OHCB). Since OHCB generally present in extremely low abundance in marine water and elevate their abundance mostly in petroleum-impacted water, our detection in cultures suggests that the occurrence of obligate and generalist hydrocarbon-degrading bacteria living with dinoflagellates may be more frequent in nature. Our work identified a core microbiome with stable association with the harmful alga and opened a window for further characterization of the physiological mechanisms and ecological implications for the dinoflagellate-bacteria association.