Molecular assessment of oyster microbiomes and viromes reveals their potential as pathogen and ecological sentinels

Abstract

Oyster aquaculture world-wide is a booming industry that can provide many benefits to coastal habitats, including economic, ecosystem-level, and cultural benefits. Oysters present several risks for human consumption, including transmission of parasites, and bacterial and viral pathogens. Oyster microbiomes are well-defined, but their connection to the incidence of pathogens, humans or others, is unclear. Furthermore, viruses associated with oysters are largely unknown, and their connection to humans, animals, and ecosystem health has not been explored. Here, we employed a One Health framework and modern molecular techniques, including 16S rRNA amplicon and metagenomic sequencing, to identify links between changes in the microbial and viral communities associated with oysters and the incidence of pathogens detected in oyster tissues and their surrounding environments. In addition, we adapted the BioFire® FilmArray®, commonly used in hospitals, to determine the presence of human pathogens within the sampled oysters. We detected known human pathogens in 50 % of the oysters tested. Within the genomic datasets, we noted that pathogens of humans, animals, and plants in oysters were shared with the nearby water and sediments, suggesting a sink-source dynamic between the oysters and their surroundings. 16S rRNA gene analysis revealed that while oysters share common microbial constituents with their surrounding environments, they enrich for certain bacteria such as Mycoplasmatales, Fusobacteriales, and Spirochaetales. On the contrary, we found that oyster viromes harbored the same viruses in near equal relative abundances as their surrounding environments. Our results show how oysters could be used not only to determine the risk of human pathogens within coastal estuaries but also how oyster viruses could be used as ecosystem-level sentinels.

Keywords: Aquaculture; Food safety; Microbiome; Oysters; Viruses.

Fig. 1

Location of the sampled oyster reefs in Galveston Bay, an estuary in the upper coast of Texas discharging in the western Gulf of Mexico. Blue are subtidal samples where oysters remain submerged throughout tidal fluctuations, and red are intertidal samples where low tides can expose oysters to air. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

(A) Taxonomic composition of the viromes of oysters and their surrounding water and sediment in four locations in Galveston Bay. The total number of viral contigs identified by VIBRANT is indicated above each bar. The top 10 most abundant viral families within all metagenomes are displayed separately; all other viral families are grouped into “Other Viruses”. (B) NMDS analysis of the family-level taxonomic composition of the viromes of the oysters and surrounding water and sediment.

Fig. 3

(A) Taxonomic composition of the microbial communities based on 16S rRNA gene sequencing shared between the oyster and their surrounding water and sediment. (B) LefSe analysis results for significantly (α < 0.05) different microbial orders with an LDA score of at least 5 between the three types of samples. (C) NMDS analysis of the presence and absence of ASVs in the samples.

Fig. 4

Comparison of the NMDS analysis comparing (A) the KEGG modules identified from the 16S rRNA gene sequence data using Picrust2 and (B) the AMG functional profiles identified by VIBRANT. Hierarchical clustering of the KEGG metabolisms found in AMGs from (C) all samples and (D) excluding the sediment and bacterial water samples. Samples from viromes are indicated by V in parentheses.

Fig. 5

A) Pathogens detected using the BF-GI and LyoKit in red. Only 3 of the 22 pathogens tested for were identified. Column names correspond to the samples in Supp. Table 2 and the color of the text corresponds to post-harvest treatments, individual quick freezing (blue), high hydrostatic pressure (red), and gamma irradiation (green), and no treatment (black). B) Pathogens identified by comparing 16S rRNA gene sequences to the MBPD database [35]. Column names are colored according to pathogen type in MBPD. Colors of boxes represent the sample type, and in the top row, a summarized representation of non-oyster presence is used. Lightly colored columns are pathogens that were not detected in oysters but were detected in the water or sediment. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)