Establishment of Coral-Bacteria Symbioses Reveal Changes in the Core Bacterial Community With Host Ontogeny

Abstract

Bacterial communities are fundamental symbionts of corals. However, the process by which bacterial communities are acquired across the life history of corals, particularly in larval and early juvenile stages, is still poorly characterized. Here, transfer of bacteria of the Scleractinian coral Acropora digitifera from adults to spawned egg-sperm bundles was analyzed, as well as acquisition across early developmental stages (larvae and newly settled spat), and 6-month-old juveniles. Larvae were reared under manipulated environmental conditions to determine the source (maternal, seawater, or sediment) of bacteria likely to establish symbiotic relationships with the host using amplicon sequencing of the 16S rRNA gene. Maternal colonies directly transferred bacteria from the families Rhodobacteraceae, Cryomorphaceae, and Endozoicimonaceae to egg-sperm bundles. Furthermore, significant differences in the microbial community structure were identified across generations, yet the structure of the coral bacterial community across early life history stages was not impacted by different environmental rearing conditions. These data indicate that the uptake and structure of bacterial communities is developmentally, rather than environmentally, regulated. Both maternal coral colonies and ubiquitous bacteria found across environmental substrates represent a potential source of symbionts important in establishing the coral microbiome. Uniquely, we report the presence of variation with ontogeny of both the core and resident bacterial communities, supporting the hypothesis that microbial communities are likely to play specific roles within the distinct life history stages of the coral host.

Keywords: 16S rRNA; bacterial communities; core bacterial communities; metabarcoding; seawater; sediment; symbiosis; transmission.

FIGURE 1

Sampling and experimental design. (A) Acropora digitifera adult coral colonies were relocated from the reef to the Coral Bay jetty. (B) A few hours prior to spawning colonies were placed into individual tubs containing seawater. (C) Egg–sperm bundles were collected for fertilization, and (D) 8 h old larvae were transferred to four experimental treatments for larval rearing: RMD_SW, reduced microbial cell density seawater; RMD_SW/Sed, reduced microbial cell density seawater and sediment; SW, seawater, SW/Sed, seawater and sediment. Seawater (SW) and sediment (Sed) samples were collected from the collection site and used in tanks during the manipulative experiment. (E) Once settled onto terracotta tiles, newly settled spat were deployed back onto the reef for 6 months. (i–iii,v,vi): DNA symbols indicates where various samples (n = 5 in all cases) were taken for 16S rRNA gene sequencing. (iv) Samples of SW and RMD_SW (n = 3) were preserved in 2% formalin at 4°C and bacterial density was determined via microscopy. Symbols are from the Integration and Application Network (ian.umces.edu/symbols/).

FIGURE 2

The number of OTUs unique to or shared between (A) all five maternal coral colonies (adult) and the released egg-sperm bundles (bundle), and between each individual coral colony and resulting bundles (B–F). The total number of OTUs associated with each group is represented in parentheses. Relative abundance of the 90 bacterial OTUs shared among adult colonies and egg-sperm bundles (G) (represented in A). Sequences were classified at family level where possible, with unclassified sequences indicated at higher taxonomic resolution. Numbers in parenthesis demark the number of different OTUs within the respective families/taxa.

FIGURE 3

Relative abundance of the bacterial taxa within the core (present in 80% of the samples) and the resident (present in 50% of the samples) microbiome in: (A) Adult, adult colonies, (B) Larvae, 2- and 4-day-old larvae; (C) Spat, newly settled spat; and (D) Juvenile, 6-month-old juveniles. Sequences were classified at family level where possible, with unclassified species indicated at higher taxonomic resolution. Families within each pie chart are organized from the lowest (*) to the highest abundance (clockwise direction).

FIGURE 4

Unique and shared OTUs between the various life history stages and the surrounding environments of seawater and sediment. (A) Adult, adult coral colonies; (B) L2, 2-day-old larvae; (C) L4, 4-day-old larvae; (D) Spat, newly settled spat, and (E) Juvenile, 6-month-old juveniles. SW, seawater from the reef; SED, sediment from the reef, RMD_SW, reduced microbial density seawater (filtered 0.22 μm). The total number of OTUs associated with each sample type is represented in parentheses. Abbreviations are SW and Sed represent samples taken: “Pre” prior to spawning, “Tank” from the tank experiment, “Post” near juveniles on the reef.

FIGURE 5

Non-metric multidimensional scaling (nMDS) of entire microbial communities associated with Acropora digitifera early life stages and reared among different environmental conditions. Development stages are indicated as: L2, 2-day-old larvae (squares); L4, 4-day-old larvae (diamonds); Spat, newly settled spat (circles); Juvenile, 6-month-old juveniles (cross). While the colors indicated the rearing environment: RMD_SW, reduced microbial cell density seawater (blue); RMD_SW/Sed, reduced microbial cell density seawater and sediment (light blue); SW, seawater (orange); SW/Sed, seawater and sediment (red).

FIGURE 6

Shared and unique microbial communities associated with A. digitifera early life stages: L2, 2-day-old larvae; L4, 4-day-old larvae; Spat, newly settled spat; Juvenile, 6-month-old juveniles. (A) The number of OTUs unique to or shared between the development phases. The total number of OTUs associated with each stage is in parentheses. (B) Relative abundance of the 109 bacterial OTUs shared among early life stages. Sequences were classified at family level where possible, with unclassified sequences indicated at higher taxonomic resolution. Numbers in parenthesis demark the number of different OTUs within the respective families/taxa.

FIGURE 7

Microbial communities shared across all life history stages of Acropora digitifera. (A) The number of OTUs unique to or shared between stages, with the total number of OTUs associated with each group represented in parentheses. (B) Relative abundance of the 28 bacterial OTUs shared among all generations. Sequences were classified at a family level where possible, with unclassified species indicated at higher taxonomic resolution. Numbers in parenthesis demark the number of different OTUs within the respective families. Adult, adult colonies; Bundle, egg-sperm bundles; L2, 2-day-old larvae; L4, 4-day-old larvae; Spat, newly settled spat; Juvenile, 6-month-old juveniles.