Vibrio Ecology in the Neuse River Estuary, North Carolina, Characterized by Next-Generation Amplicon Sequencing of the Gene Encoding Heat Shock Protein 60 (hsp60)

Abstract

Of marine eubacteria, the genus Vibrio is intriguing because member species are relevant to both marine ecology and human health. Many studies have touted the relationships of Vibrio to environmental factors, especially temperature and salinity, to predict total Vibrio abundance but lacked the taxonomic resolution to identify the relationships among species and the key drivers of Vibrio dynamics. To improve next-generation sequencing (NGS) surveys of Vibrio, we have conducted both 16S small subunit rRNA and heat shock protein 60 (hsp60) amplicon sequencing of water samples collected at two well-studied locations in the Neuse River Estuary, NC. Samples were collected between May and December 2016 with enhanced sampling efforts in response to two named storms. Using hsp60 sequences, 21 Vibrio species were identified, including the potential human pathogens V. cholerae, V. parahaemolyticus, and V. vulnificus Changes in the Vibrio community mirrored seasonal and storm-related changes in the water column, especially in response to an influx of nutrient-rich freshwater to the estuary after Hurricane Matthew, which initiated dramatic changes in the overall Vibrio community. Individual species dynamics were wide ranging, indicating that individual Vibrio taxa have unique ecologies and that total Vibrio abundance predictors are insufficient for risk assessments of potentially pathogenic species. Positive relationships between Vibrio, dinoflagellates, and Cyanobacteria were identified, as were intraspecies associations, which further illuminated the interactions of cooccurring Vibrio taxa along environmental gradients.IMPORTANCE The objectives of this research were to utilize a novel approach to improve sequence-based surveys of Vibrio communities and to demonstrate the usefulness of this approach by presenting an analysis of Vibrio dynamics in the context of environmental conditions, with a particular focus on species that cause disease in humans and on storm effects. The methods presented here enabled the analysis of Vibrio dynamics with excellent taxonomic resolution and could be incorporated into future ecological studies and risk prediction strategies for potentially pathogenic species. Next-generation sequencing of hsp60 and other innovative sequence-based approaches are valuable tools and show great promise for studying Vibrio ecology and associated public health risks.

Keywords: Vibrio; amplicon sequencing; hsp60; microbial ecology; public health.

FIG 1

Relative abundances of Vibrio species in the NRE from May through December 2016 using the hsp60 gene (A) and 16S rRNA gene (B). Panels designate sampling site (stations 70 or 120) and depth (surface or bottom water). Numbers in parentheses indicate the number of sampling events per month. There were no 16S reads assigned to the Vibrio genus in several samples, as indicated by blank data slots.

FIG 2

Temperature, salinity, average daily river discharge, total Vibrio abundance, organic nutrients, and total chlorophyll a for all sampling dates, stations, and depths in the NRE during the study period. Organic nutrient data are averaged across stations and depths for each sampling date. Named storm events Tropical Storm Colin (June 2016) and Hurricane Matthew (October 2016) are marked by vertical lines.

FIG 3

(A) Distance-based redundancy analysis (dbRDA) plot of the fitted distLM model for Vibrio communities in the NRE. Each point in the ordination represents the Vibrio community in a given water sample. The distance between points is the Bray-Curtis distance (dissimilarity) between Vibrio communities. Vectors denote significant environmental gradients (P < 0.05) in the distLM. Points associated with increased river discharge after Hurricane Matthew are labeled according to the number of days post-Matthew the sample was taken. The ordination is also presented as bubble plots in which point size is proportional to the relative abundances of V. parahaemolyticus (B), V. vulnificus (C), and V. cholerae (D). The bubble key indicates bubble size for 100, 70, 40, and 10% of the range for each species.

FIG 4

Spearman rank correlation plot showing correlations between Vibrio taxa and between Vibrio taxa and phytoplankton photopigments. The black dotted line distinguishes Vibrio-Vibrio relationships from Vibrio-pigment relationships. Only significant correlations (P < 0.05) are plotted.

FIG 5

Network plots showing positive and negative Spearman rank correlations (Spearman's r > 0.4) between Vibrio species identified using hsp60 and the top 15 most abundant bacterial phyla identified using 16S rRNA for all samples (A) and for samples associated with Hurricane Matthew (B). Vector width is proportional to the strength of the correlation.

FIG 6

Sampling locations in the Neuse River Estuary (NRE) in eastern North Carolina.