Resilience of coral-associated bacterial communities exposed to fish farm effluent

Abstract

Background: The coral holobiont includes the coral animal, algal symbionts, and associated microbial community. These microbes help maintain the holobiont homeostasis; thus, sustaining robust mutualistic microbial communities is a fundamental part of long-term coral reef survival. Coastal pollution is one major threat to reefs, and intensive fish farming is a rapidly growing source of this pollution.

Methodology & principal findings: We investigated the susceptibility and resilience of the bacterial communities associated with a common reef-building coral, Porites cylindrica, to coastal pollution by performing a clonally replicated transplantation experiment in Bolinao, Philippines adjacent to intensive fish farming. Ten fragments from each of four colonies (total of 40 fragments) were followed for 22 days across five sites: a well-flushed reference site (the original fragment source); two sites with low exposure to milkfish (Chanos chanos) aquaculture effluent; and two sites with high exposure. Elevated levels of dissolved organic carbon (DOC), chlorophyll a, total heterotrophic and autotrophic bacteria abundance, virus like particle (VLP) abundances, and culturable Vibrio abundance characterized the high effluent sites. Based on 16S rRNA clone libraries and denaturing gradient gel electrophoresis (DGGE) analysis, we observed rapid, dramatic changes in the coral-associated bacterial communities within five days of high effluent exposure. The community composition on fragments at these high effluent sites shifted towards known human and coral pathogens (i.e. Arcobacter, Fusobacterium, and Desulfovibrio) without the host corals showing signs of disease. The communities shifted back towards their original composition by day 22 without reduction in effluent levels.

Significance: This study reveals fish farms as a likely source of pathogens with the potential to proliferate on corals and an unexpected short-term resilience of coral-associated bacterial communities to eutrophication pressure. These data highlight a need for improved aquaculture practices that can achieve both sustainable industry goals and long-term coral reef survival.

Figure 1. A map of the study sites and surface water in vivo chlorophyll a measurements averaged over 4 consecutive weekly samplings.

Note that transplant site “Fish Pens” did not have corals placed there since no live coral currently exists at that site. It was the location of all water and sediment sampling for the experiment as a representative fish pen.

Figure 2. Denaturing gradient gel electrophoresis (DGGE) image of one coral colony (#1) at all sites (Far-1&2, Near-1&2, and Reference) across all time points (T-0, T-5 days, T-22 days).

Arrows indicate Desulfovibrio bands that were sequence verified. There were two separate fragments sampled from each colony at T0. Both samples are shown on this gel as T0 a and b. Samples from the high effluent sites (Near-1&2) at T-5 days were the only visible Desulfovibrio bands.

Figure 3. DGGE image showing Roseobacter band (>97% sequence similarity) at all time points in the Fish Pen water, but only T-5 days at site Near-1 for coral 1.

Figure 4. The distribution of phylotypes by class from 16S rRNA clone libraries.

Libraries have been pooled by site (i.e. libraries from all fragments at a given time point from Far-1 and Far-2 are represented by “Far” and the same is true for “Near”) and each individual time point (T0, T-5 days, T-22 days) is shown separately. The number of sequences represented is denoted as “n = ”. Spirochaetes, previously seen only in corals infected with diseases, and Clostridia are both present even though fragments showed no visible signs of disease. Spirochaetes sequences are only present at T-5 days and predominantly at high effluent sites.