Rapid Formation of Microbe-Oil Aggregates and Changes in Community Composition in Coastal Surface Water Following Exposure to Oil and the Dispersant Corexit

Abstract

During the Deepwater Horizon (DWH) oil spill, massive quantities of oil were deposited on the seafloor via a large-scale marine oil-snow sedimentation and flocculent accumulation (MOSSFA) event. The role of chemical dispersants (e.g., Corexit) applied during the DWH oil spill clean-up in helping or hindering the formation of this MOSSFA event are not well-understood. Here, we present the first experiment related to the DWH oil spill to specifically investigate the relationship between microbial community structure, oil and Corexit®, and marine oil-snow in coastal surface waters. We observed the formation of micron-scale aggregates of microbial cells around droplets of oil and dispersant and found that their rate of formation was directly related to the concentration of oil within the water column. These micro-aggregates are potentially important precursors to the formation of larger marine oil-snow particles. Therefore, our observation that Corexit® significantly enhanced their formation suggests dispersant application may play a role in the development of MOSSFA events. We also observed that microbial communities in marine surface waters respond to oil and oil plus Corexit® differently and much more rapidly than previously measured, with major shifts in community composition occurring within only a few hours of experiment initiation. In the oil-amended treatments without Corexit®, this manifested as an increase in community diversity due to the outgrowth of several putative aliphatic- and aromatic-hydrocarbon degrading genera, including phytoplankton-associated taxa. In contrast, microbial community diversity was reduced in mesocosms containing chemically dispersed oil. Importantly, different consortia of hydrocarbon degrading bacteria responded to oil and chemically dispersed oil, indicating that functional redundancy in the pre-spill community likely results in hydrocarbon consumption in both undispersed and dispersed oils, but by different bacterial taxa. Taken together, these data improve our understanding of how dispersants influence the degradation and transport of oil in marine surface waters following an oil spill and provide valuable insight into the early response of complex microbial communities to oil exposure.

Keywords: MOSSFA; deepwater horizon; marine oil-snow; micro-aggregate; oil and corexit®.

Figure 1

Cell abundance (top), micro-aggregate abundance (middle), and cytochrome P450 gene copy abundance (bottom) observed in the four mesocosm treatments. Columns represent the pooled mean of replicated measurements with error bars representing the pooled standard error.

Figure 2

Estimated oil equivalents (EOE) within the WAF, CEWAF, and DCEWAF mesocosms. Points represent the average of triplicate measurements. Error bars represent standard deviation. EOE measurements in the Control mesocosms were below the detection limit and are not shown.

Figure 3

NMDS plot of the shifts in the microbial community structure observed in the four mesocosm treatments. Darker colors represent later time-points for each treatment, which is further highlighted with the overlaid arrows, which indicate the direction of community succession over time.

Figure 4

Relative abundances of the 30 most abundant OTUs. Each bar is the average of triplicate treatments. Tick marks on the x-axes are the same as those in Figure 1, demarking each experimental time point, taken every 12 h. Color key is the same as Figure 1: gray is Control, orange is WAF, blue is CEWAF, and green is DCEWAF. When no data is presented, this reflects the absence of the OTU in that treatment.

Figure 5

Observed changes in phylogenetic diversity, calculated as the inverse-Simpson index, over time for each of the mesocosm treatments.

Figure 6

Correlation network analysis of OTUs within the mesocosm experiment. OTU nodes are colored according to the treatment in which they were most abundant. Rectangular nodes indicate a known or putative hydrocarbon degrading taxon. Asterisks denote taxa which contain both hydrocarbon degrading and non-hydrocarbon degrading members. The width of the lines connecting the nodes are proportional to correlation strength (from 0.70 to 0.99) with red and green lines indicating a negative or positive correlation, respectively. For simplicity of viewing, we only display OTUs that display one-step correlations to OTU7 and OTU20 (A) and OTU10 (B) to highlight connections between putative hydrocarbon degraders prevalent in WAF and Corexit amended treatments.

Figure 7

Time until a significant increase in aggregate abundance was observed within the mesocosms, plotted vs. the concentration of oil (EOE mg L⁻¹). This was defined as the moment when the average aggregate concentration became greater than the limit of quantification (calculated here as M + 3·SD). Error bars represent standard deviation.