Culture-Independent Analysis of Bacterial Community Composition during Bioremediation of Crude Oil-Polluted Soil

Aim: To use cultivation-independent techniques based on DGGE of PCR-amplified 16S rRNA gene and to evaluate bacterial community composition during bioremediation of crude oil-polluted soil.
Study Design: Molecular fingerprints of bacterial populations involved in the active phase of crude oil biodegradation were generated with DGGE after 16S rRNA gene amplification.
Place and Duration of Study: Department of Microbiology and Plant Pathology, University of Pretoria, South Africa, between March and August 2008.
Methodology: Crude oil-degrading bacteria in soil microcosms contaminated with 4% crude oil and then biostimulated with nitrogen-phosphorus-potassium inorganic fertilizer (NPK: designated PN soil), calcium ammonium nitrate (designated PU soil) and poultry droppings (designated PP soil) respectively were characterized with PCR of the gene for the small subunit (SSU) of the bacterial ribosome. Total culturable heterotrophic and hydrocarbon utilizing bacteria were enumerated using plate count and Bushnell Haas media. Total organic carbon content was measured throughout the study period to indirectly determine the effect of microbial activity on carbon content in biostimulated treatments as against controls. Gas chromatography was used to monitor hydrocarbon degradation with time while electron microscopy examined community richness during hydrocarbon degradation. Reamplified dominant DGGE bands (550bp) were cleaned up and sequenced using an ABI 3130XL genetic analyzer. Electropherograms were inspected with Chromas Lite 2.01. Sequence identification was performed using BLAST.
Results: Dendogram of the DGGE bands constructed using Jaccard coefficient algorithm revealed that communities from PU and PP-amended soils each formed distinct clades whereas PN treated soil showed less association when compared with PU and PP respectively. Fifty distinct bands were excised, reamplified by PCR and sequenced. Sequence analysis revealed the presence of phylogenetically distinct known hydrocarbon degrading bacteria like Corynebacterium spp., Dietzia spp., Janibacter sp. low G+C Gram positive bacterial clones Nocardioides spp., Rhodococcus erythropolis and uncultured bacterial clones. Forty successful sequence data obtained from the excised DGGE bands were submitted to GenBank database under accession numbers GU451069 to GU451108. Chromatograms of the residual hydrocarbons in test treatments and controls showed that biodegradation occurred markedly in treated soils in this order PN>PU>PP while no significant loss was observed in the oil-contaminated control on days zero and 42 respectively. Bacterial counts increased significantly in PN, PU and PP treatments and not in controls PC and OC. Total organic carbon increased appreciably in PN, PU and PP respectively from day zero to day 28. Electron micrographs of microbial consortia in the nutrient-amended soils revealed presence of active populations induced by biostimulation as against the sparsely populated controls.
Conclusion: The results suggest that nutrient amendment stimulates and selects indigenous soil bacteria that are able to degrade petroleum hydrocarbons.