This study shows, for the first time, the effects of crude oil on E. huxleyi and its associated bacterial community under future OA (elevated CO2) conditions. Initially, we show that OA is a positive stressor on E. huxleyi, as elevated CO2 resulted in higher Chl a concentrations (a proxy for growth of the alga) and this is consistent with some previous studies34,35. However, crude oil induced a negative response, and together with elevated CO2 conditions, synergistic effects upon the growth of E. huxleyi that were not observed under ambient CO2 conditions. CO2 is a limiting factor in photosynthesis, therefore higher concentrations can be utilised to compensate for the metabolic stress of maintaining intracellular pH and especially as cells divide and replicate to higher abundance levels. However, whilst the cells may grow in more abundance under elevated CO2, their coccoliths are less calcified due to the relatively lower pH, and as a consequence explain their higher susceptibility to crude oil toxicity. E. huxleyi was reported to adopt the ‘Cheshire Cat’ escape strategy where the cells exist predominantly in a haploid stage that lack in mineralised scales, allowing evasion of viral predation in addition to physical and chemical stress36–38. Our results, however, indicate that calcification provides a physical defence against crude oil toxicity, and the converse (i.e. a lack of mineralised scales) puts the organism at risk of being "killed" when challenged with crude oil. Hydrocarbon adsorption on calcite surfaces has been reported39, so it may be conjectured that thicker, healthier calcite scales would prevent, or if at least reduce the absorption of toxic hydrocarbon chemicals through and into the cell membrane and intracellular region of the cells where it would inevitably cause impairment of cell functioning and ultimately death. As discussed below, the associated bacterial community, and more specifically the oil-degrading community will play a protective role from crude oil toxicity.
Community analysis of the various treatments showed that bacterial species diversity associated with E. huxleyi was not significantly different under elevated compared to ambient CO2 conditions and nor with/without exposure to crude oil. However, certain shifts in the communities, as effected by certain taxa, were observed that delineated these treatments, as observed with the enrichment of Marinobacter, Methylobacterium and unclassified Bacteria in the ambient CO2 treatments after exposure to crude oil. Marinobacter is a genus known to utilize aliphatic and polycyclic aromatic hydrocarbons as a sole source of carbon and energy40–42. Whilst Marinobacter are commonly associated with the degradation of aliphatic and aromatic hydrocarbons in the ocean, they are in fact also capable of utilizing various non-hydrocarbon substrates, such as organic nitrogen-containing compounds (e.g. amino acids, carboxylic acids etc.)43. A decrease in their relative abundance (0–19%) was likely due to increased competition for bioavailable macro- and/or micronutrients. Although primary production increases production of organic carbon, photosynthesis utilises micronutrients such as iron that are co-limiting growth factors for heterotrophic bacteria3,44,45. Not unexpectedly, Marinobacter increased in relative abundance after oil enrichment, but it did so also under elevated CO2 conditions which has not been previously reported. We posit that under future OA conditions, these ubiquitous oil-degrading organisms may not be expected to be affected in their natural response to oil spillage.
Unlike Marinobacter, under elevated CO2 conditions Methylophaga and Methylobacterium became negatively, albeit marginally, impacted following exposure to crude oil, whereas these organisms thrived under the ambient CO2 treatments after exposure to the oil. Whilst some methylotrophs have been shown to become enriched by crude oil or its petrochemical refined products40 and references therein, hitherto this has not been reported for members of the genus Methylobacterium. This genus belongs to a group of organisms known for methylotrophy, in that they rely exclusively on single carbon compounds, such as methanol, methylamine and dimethylsulfide, as a sole source of carbon and energy46. However, genes have been identified within Methylobacterium genomes that are associated with the degradation of hydrocarbons47, and the genome annotation for Methylobacterium extorquens PA1 in the KEGG (http://www.genome.jp/kegg-bin/show_pathway?mex01220) has been shown to possess anaerobic benzene degradation genes. Some methylotrophs, particularly of the genus Methylophaga, are also capable of utilising fructose48 or even hydrocarbons49,50 as sole growth substrates. Furthermore, the presence of hydrocarbon-degrading methylotrophs has been shown amongst communities of bacteria associated with eukaryotic phytoplankton27,51, possibly because eukaryotic phytoplankton can adsorb hydrocarbons from seawater environment or because they synthesise alkenones and other hydrocarbon-like compounds52 and references therein which hydrocarbon-degrading methylotrophs could potentially feed on, although this hydrocarbon-mediated algal-bacterial symbiosis remains unproven. Whilst it remains to be substantiated, we posit that the enrichment of Methylobacterium in the presence of crude oil under ambient CO2 levels may be due to their potential to utilise hydrocarbons as a source of carbon and energy. An alternative explanation for their enrichment may be because these organisms were able to acquire their carbon and energy requirements from extracellular particulate and/or dissolved organic carbon (DOC) released by E. huxleyi; such exudates can offer a rich source of methylated sugars that can be utilised by methylotrophs53,54. It is also noteworthy to mention the release of organic matter exudates by E. huxleyi has been reported to be affected by changes in dissolved CO2 levels55. Whilst we did not analyse for exudate production in our experiments, others showed that elevated CO2 conditions significantly increased the release of transparent exopolymer particles (TEP) and particulate combined carbohydrates (pCCHO) by E. huxleyi55, although the lability of this carbon store to be utilised by methylotrophs, like Methylobacterium, is unknown. At the very least, Methylobacterium was able to survive when challenged with the crude oil, but only under ambient CO2 conditions. As explained earlier, the coccolith scales of E. huxleyi would be in a weakened state, or less robust, to deal with oil exposure under elevated CO2 conditions, due to the detrimental effects caused by a reduced pH to the structure of their calcium carbonate armour, potentially permitting the intracellular entry of toxic hydrocarbons.
Sphingomonas are metabolically diverse and capable of utilising hydrocarbons as growth substrates56, but these organisms were not enriched after oil exposure under ambient CO2 conditions, and their relative abundance had noticeably decreased after oil exposure in the elevated CO2 treatment. This genus belongs to the class Alphaproteobacteria, and their decrease in relative abundance under elevated CO2 supports other studies that found OA conditions (from elevated CO2 levels) can negatively impact members belonging to this class57,58. Other genera comprising members with reported hydrocarbon-degrading abilities, such as Halomonas and Alteromonas, and which have been reported associated with E. huxleyi, such as Marivita, Hoeflea, Balneola, Arenibacter, Marinoscillum and Thalassospira22, were also detected but not affected by either CO2 treatment and nor exposure to the oil. We suspect that these organisms found associated with E. huxleyi are incapable of utilising hydrocarbons, which might explain why they were not enriched by the oil in either of the CO2 treatments. On the other hand, members of the genera Prevotella, Streptococcus and Veillonella which have been reported with hydrocarbon-degrading ability59–62, had increased in relative abundance following oil enrichment, but only under the elevated CO2 treatment, but then declined to < 1% abundance by the termination of these experiments. It is unknown to us at present what may have caused this decline and why these changes only occurred in the elevated CO2 treatment.
With atmospheric CO2 concentrations continuing to unabatedly rise and drive OA, our results show that the structuring of bacterial communities living associated with coccolithophores will be quite different in a future ocean, including potential extinctions of some oil-degrading taxa from these communities. Consequently, it is expected that this could have quite profound effects on the fate of spilled crude oil in the ocean, whether from natural seepage or anthropogenic inputs, considering that oil-degrading bacteria are crucial in the biodegradation and ultimate purging of petrochemical pollutants in the ocean. Assessing this by comparing nC17/pristane and nC18/phytane ratios of acid-killed incubations with the same from live incubations, we found that the biodegradation of the aliphatic fraction occurred under both ambient and elevated CO2 conditions. Despite shifts in microbial population dynamics due to elevated CO2 conditions, aliphatic hydrocarbon degradation was indifferent in both ambient and elevated CO2 conditions tested. We note that nC8, nC9, and nC10 were not detected in all treatments, likely due to their loss by volatilisation as expected for these low-molecular-weight hydrocarbons. On the other hand, the aromatic fraction in the crude oil, as measured by our analysis of naphthalene, methylnaphthalene, dimethylnaphthalene, ethylnaphthalene, C3-alkylnaphthalene, phenanthrene, methylphenanthrene, dimethylphenanthrene, ethylphenanthrene and C3-alkylphenanthrene, was not significantly degraded in neither the ambient or elevated CO2 conditions. It is possible that the oil-degrading microbial community associated with E. huxleyi was, for some reason, incapable of inducing the degradation of aromatic hydrocarbons. The metabolic capability to degrade aromatic hydrocarbons, however, would likely have been served by Marinobacter63–66 and references therein, as members of this genus have been reported to utilise polycyclic aromatic hydrocarbons as sole growth substrates. Our experiments were run for 22 days, which was likely too short to begin to detect the initial stages in the biodegradation of the aromatic fraction because, as has been reported in many marine oil spill studies, the biodegradation of the aromatic fraction in crude oil does not often commence until the more labile aliphatics have become almost depleted (e.g. 67). Whilst we cannot conclude on what effects OA conditions might have on the biodegradation of aromatic hydrocarbons, based on our analysis of the aliphatic fraction, our results suggest that the response and biodegradation activities of oil-degrading communities associated with coccoliphores, such as E. huxleyi, would be largely unaltered when challenged with crude oil in future OA conditions.