Central role of dynamic tidal biofilms dominated by aerobic hydrocarbonoclastic bacteria and diatoms in the biodegradation of hydrocarbons in coastal mudflats

Abstract

Mudflats and salt marshes are habitats at the interface of aquatic and terrestrial systems that provide valuable services to ecosystems. Therefore, it is important to determine how catastrophic incidents, such as oil spills, influence the microbial communities in sediment that are pivotal to the function of the ecosystem and to identify the oil-degrading microbes that mitigate damage to the ecosystem. In this study, an oil spill was simulated by use of a tidal chamber containing intact diatom-dominated sediment cores from a temperate mudflat. Changes in the composition of bacteria and diatoms from both the sediment and tidal biofilms that had detached from the sediment surface were monitored as a function of hydrocarbon removal. The hydrocarbon concentration in the upper 1.5 cm of sediments decreased by 78% over 21 days, with at least 60% being attributed to biodegradation. Most phylotypes were minimally perturbed by the addition of oil, but at day 21, there was a 10-fold increase in the amount of cyanobacteria in the oiled sediment. Throughout the experiment, phylotypes associated with the aerobic degradation of hydrocarbons, including polycyclic aromatic hydrocarbons (PAHs) (Cycloclasticus) and alkanes (Alcanivorax, Oleibacter, and Oceanospirillales strain ME113), substantively increased in oiled mesocosms, collectively representing 2% of the pyrosequences in the oiled sediments at day 21. Tidal biofilms from oiled cores at day 22, however, consisted mostly of phylotypes related to Alcanivorax borkumensis (49% of clones), Oceanospirillales strain ME113 (11% of clones), and diatoms (14% of clones). Thus, aerobic hydrocarbon biodegradation is most likely to be the main mechanism of attenuation of crude oil in the early weeks of an oil spill, with tidal biofilms representing zones of high hydrocarbon-degrading activity.

Fig 1

Location map of the sampling site on the Colne Estuary (Essex, United Kingdom) (51°50.5′N, 0°58.4′E) used for the oil spill mesocosms. (Coastline map drawn and digitized by Steven McMellor, with the complete map created using several Geographic Information System layers publicly available from Natural England [copyright, Natural England 2012].)

Fig 2

Schematic representation of the experimental tidal mesocosms. (a) Surface view of nine polypropylene containers (32.5 by 42 by 25.5 cm) holding 20 sediment cores (9.8 cm high and 6.5 cm in diameter) that were placed above reservoir containers. Three different treatments were carried out in triplicate, as follows: (i) autoclaved sediment cores with Forties crude oil (Colne oil killed [COK]), (ii) sediment cores with Forties crude oil (Colne oil live [COL]), and (iii) sediment core without oil (Colne live [CL]). Tair, air temperature probe; Tsub1, temperature probe in sediment; Tsub2, temperature probe in water reservoir tank; Tsub3, temperature probe in water tank containing cores; Solm, tube solar energy sensors; QS, spot sensors for solar energy in the photosynthetically active waveband. (b) Side view of the upper container holding the sediment cores and the lower reservoir, illustrating the flow of seawater. At low tide, the pump was off, and seawater drained to the level of the tube connected to the pump. At high tide, the pump was on, and seawater flowed continuously through a tube passing through the hole. The tube was curved downwards into the upper tank to prevent the transfer of oil from the upper tank to the reservoir.

Fig 3

Proportion of the higher taxa in mudflat mesocosms (top 1.5 cm of sediment), based on 454 pyrosequencing analysis of bacterial 16S rRNA genes (Greengenes analysis with RDP classification). (Left) Bacterial communities in nonoiled sediments. (Right) Bacterial community in oiled sediments. CL-2, nonoiled sediment, day 2 (899 sequences); COL-2, oiled sediment, day 2 (258 sequences); COL-12, oiled sediment, day 12 (321 sequences); CL-21, nonoiled sediment, day 21 (761 sequences); COL-21, oiled sediment, day 21 (1,492 sequences); Actino, Actinobacteria; Bactero, Bacteroidetes; Chloro, Chloroflexi; Diatom, Bacillariophyceae; AlphaP, Alphaproteobacteria; GammaP, Gammaproteobacteria; DeltaP, Deltaproteobacteria.

Fig 4

Principal-component analysis of 454 pyrosequencing data and cluster analysis of T-RFLP data from sediment samples. Data are based on directly amplified 16S rRNA genes from Bacteria. (a) PCA plot based on P1 and P2 components. (b) PCA plot based on P1 and P3 components. The first principal component (P1) explains 35.8% of the variation between the conditions, the second (P2) explains 26%, and the third (P3) explains 23%. PCA was conducted by using the UniFrac metric algorithm and was based on a neighbor-joining phylogenetic tree. (c) Cluster analysis based on Bray-Curtis similarity taking into account the square root transformation of T-RF relative abundances (based on the mean height of the T-RFs). Oiled sediments are indicated by closed squares, and nonoiled sediments are indicated by open squares.

Fig 5

T-RFLP profiles of triplicate tidal biofilms harvested in oiled (biofilm-COL) and nonoiled (biofilm-CL) mesocosms on the last day of the experiment (day 22). The profiles were obtained with primer 63F and restriction enzyme AluI; for clarity, data for primer 1389R are not shown. The peak height represents the relative abundance of each T-RF, and the horizontal scale represents the T-RF length in nucleotides. The highlighted T-RF peak (size, 35 nucleotides) from the oiled tidal biofilms dominates each of the three communities. In the nonoiled tidal biofilms, however, a double peak is observed at this position. By correlating in silico T-RF positions from cloned sequences, it is clear that the 35-nucleotide T-RF in the oiled floating biofilm comes from Alcanivorax, while the double peak at a similar position in the nonoiled biofilm comes from various other groups, including Gammaproteobacteria, Bacteroidetes, as well as species distantly related to cultured organisms.