The Effect of Nitrogen, Sulfur, and Phosphorus Compounds on Bioremediation of Oil Spills by Pseudomonas fluorescens and Bacillus subtilis

Bioremediation involves microbial metabolism of harmful substances. This study was undertaken to enhance oilspill bioremediation through the addition of sodium sulfate, sodium nitrate, and sodium phosphate to samples containing Pseudomonas fluorescens and Bacillus subtilis. Since these bacteria often utilize nitrates during cellular respiration and nitrate ammonification, respectively, it was hypothesized that sodium nitrate would increase the amount of hydrocarbon metabolites present in samples more than sodium sulfate or sodium phosphate. Samples containing different combinations of bacteria, .1 and .01 molar solutions of the above nutrients, salt water, and crude-oil-substitute hexadecane (utilized to replicate the long hydrocarbon chains found in petroleum) were shaken at 27°C for seven or 14 days. Bacterial concentration was estimated by turbidity measurements. Each organic layer was analyzed with gas chromatography to identify any smaller hydrocarbons. The data from two trials partially disprove the hypothesis. While the treatment of .1 M sodium nitrate caused one of the greatest increases in bacterial concentration with P. fluorescens and B. subtilis, the chromatographic data from these treatments differed only marginally from the positive controls and other treatments. Additionally, the data suggested that substantial metabolite formation occurred only in P. fluorescens samples with .1 M and .01 M treatments of sodium phosphate. However, the .1 M nutrient combination and .01 M sodium sulfate treatments caused the greatest hexadecane disappearance. These results are relevant to the enhancement of bioremediation techniques; a need exemplified by the Deepwater Horizon catastrophe. Introduction Marine oil spills represent one of the most devastating accidents occurring worldwide, as demonstrated by the Deepwater Horizon explosion on April 20, 2010. Preventing these spills entirely would be practically impossible; however, preparing for them can limit the damage they cause. Increasing the rate at which oil spills in the ocean can be cleaned would save millions of organisms and acres from lasting harm. This project was completed to increase the effectiveness of bioremediation techniques by the addition of inorganic nutrients to Pseudomonas fluorescens and Bacillus subtilis for use during oil spills. According to the EPA, almost 14,000 oil spills are reported each year1. Many different techniques are used to attempt to clean oil spills. Booms–barriers that extend about three feet below the surface of the water can be used to contain the oil, while The Effect of Nitrogen, Sulfur, and Phosphorus Compounds on Bioremediation of Oil Spills by Pseudomonas fluorescens and Bacillus subtilis Meghan Shea1*, Sandra Litvin2, and Anastasia Chirnside3 Student1, Teacher2: Unionville High School, Kennett Square, PA 19348 Mentor3: University of Delaware, Newark, DE 19716 *Corresponding author: [email protected] INTERNSHIP ARTICLE skimmers act as vacuums to remove this contained oil1. Sorbents, or materials that absorb or adsorb oil, are also commonly used to remove the final traces of oil from an oil spill1. However, the technique that requires the least human effort and is often the most successful in the long run is bioremediation1. According to the Civil Engineering Department of Virginia Tech, bioremediation is “the application of biological treatment to the cleanup of hazardous chemicals in the soil and surface or subsurface waters”2. Certain bacteria are able to feed on the contamination, which in this case is long hydrocarbon chains, and use the energy for growth and reproduction2. Through this process, the oil is metabolized into water and carbon dioxide. While this process usually occurs naturally, it can be stimulated through the addition of nutrients and microbes, which is known as bioaugmentation. Common organisms used in the bioaugmentation of petroleum are Pseudomonas, Proteus, Bacillus, Penicillum, and Cunninghamella. For this experiment, two rod-shaped, aerobic, heterotrophic microbes, Pseudomonas fluorescens and Bacillus subtilis, were used to degrade oil. P. fluorescens colonizes soil, water, and plant surface environments and is motile due to multiple polar flagella3. This Gram-negative bacterium can utilize nitrate as an electron accepter in place of oxygen during cellular respiration3. B. subtilis, a Gram-positive bacterium with peritrichous flagella, also inhabits soil and water4. In place of crude or motor oil, both of which contain additives that react with different aspects of this experiment making data collection difficult, HD (see figure 1 for abbreviations and acronyms) was utilized to replicate the long hydrocarbon chains found in petroleum. HD is an alkane hydrocarbon with a 16-carbon chain. Often, nitrogen, phosphorus, and sulfur are the limiting factors of bioremediation in aquatic environments. Therefore, sodium nitrate (NaNO3), sodium phosphate (Na3PO4), and sodium sulfate (Na2SO4) were used as sources of the above nutrients in order to determine which has the greatest effect on the bioremediation of hydrocarbons by P. fluorescens and B. subtilis. All three inorganic nutrients are soluble in water, making them effective for aquatic oil spills. In P. fluorescens, 30 Figure 1. Abbreviations and acronyms. Meghan Shea, Sandra Litvin, and Anastasia Chirnside Page 2 of 8 31 Materials and Methods Inoculating liquid cultures of P. fluorescens and B. subtilis: One liter of nutrient broth was prepared using the standard procedure found on the container and sterilized using an autoclave. 150 ml of nutrient broth were added to each of three clean Erlenmeyer flasks labeled B. subtilis, P. fluorescens, and Spectronic 20 control, respectively. The three flasks were autoclaved. The flask labeled P. fluorescens was inoculated with bacteria from a pure colony from Carolina Biological, using a metal inoculating loop and standard flaming techniques to avoid contamination. The inoculation was repeated twice more with P. fluorescens, re-inoculating the same flask to ensure adequate bacterial growth. The above procedures were repeated with the B. subtilis pure colony and the flask labeled B. subtilis. All three flasks – two inoculated and one control – were incubated at 27oC. Creating a standard spec-20 absorbance curve for each bacterium: A Spec-20 was set to a wavelength of 686 nm and left to warm up for a minimum of 15 minutes. Eleven clean culture tubes were obtained and labeled 1-11. Using different pipettes, turbid P. fluorescens liquid culture and nutrient broth were added to each tube to a total volume of 10.0 ml. The volume of liquid culture was reduced by 1.0 ml in each tube from 10.0 ml to 0.0 ml. Using a vortex mixer, each tube was mixed thoroughly for 1 – 2 seconds. After mixing, 3-4 ml of broth from tube 11 were transferred to a clean Spec-20 tube and the outside of the tube was wiped with a Kimwipe. The tube was placed in the Spec-20 and used as a blank, following the instructions for the specific machine. Starting with tube 10 and going backwards to tube 1, 3-4 ml of each dilution were transferred to the Spec-20 tube and the absorbance was measured. Using the absorbance data, a X-Y scatter plot of dilution factor vs. absorbance was created and a second order polynomial trend line was added to the plot. From the plot, the range over which absorbance was proportional to bacterial concentration was identified, and a second plot was created using only this range. A linear trend line was added to this nitrate is often incorporated in metabolism by serving as an electron acceptor during cellular respiration3. B. subtilis uses nitrate ammonification in anaerobic conditions4. However, the specific use of these nutrients during the hydrocarbon metabolism of P. fluorescens and B. subtilis has not yet been studied. Since the two bacteria utilize nitrate during cellular respiration in anaerobic conditions and nitrate ammonification, respectively, it was hypothesized that sodium nitrate, an inorganic nutrient that provides nitrogen to the microbes, would cause the greatest HD degradation and production of smaller hydrocarbons. To test this hypothesis, samples were prepared containing different combinations of bacteria, .1 and .01 molar solutions of the above nutrients, salt water, and crude-oil-substitute hexadecane. The turbidity of each sample was measured daily using a Spec-20. An absorbance standard curve for each bacterium coupled with serial-dilution plating facilitated the conversion of absorbance to CFU/ml. After shaking at 27°C for 7 or fourteen days, the organic layer was removed from each sample using an Acetonitrile-based Solid Phase Extraction. Each organic layer was analyzed with gas chromatography (FID) to identify any smaller hydrocarbons, the presence of which would indicate successful bioremediation. new plot. Using the equations of the trend lines from the first and second plots, a correction formula was derived to correct for linearity at high absorbance readings. The entire procedure above was repeated with a turbid B. subtilis culture to obtain a second corrected absorbance curve. Calibrating absorbance measurements with viable cell counts: One liter of nutrient agar was prepared using the standard procedure found on the container, sterilized using an autoclave, and poured into sterile Petri dishes. 16 sterile culture tubes were obtained and labeled 10-1 – 10 -16. Using a 10-ml pipette, 9 ml of sterile nutrient broth were added to each culture tube. 32 nutrient agar Petri dishes were obtained and two were labeled 10-1, two 10-2, etc., through 10-16. Using a Spec-20 and the procedure detailed above, the absorbance of a turbid P. fluorescens culture was recorded. Using a 1-ml micropipette, 1 ml of turbid P. fluorescens culture was transferred to the 10-1 tube. The dilution was mixed thoroughly by vortexing for 1 – 2 seconds. Using a new pipette tip, 1 ml of the 10-1 dilution was withdrawn and transferred to the 10-2 tube. The dilution was mixed using a vortex, and the transferring technique above was repeated until all serial dilutions were made. Using a new pipette tip and starting with the 10-16 dilution, 0.1 ml of the dilution was pipetted onto the appropriately labeled nutrient agar plate and immediately spread with a sterile plate spreader. This procedure was repeated with the duplicate 10-16 plate and all subsequent dilutions. All plates were incubated at 27oC for 48 hours, at which time the colonies of plates with 25 – 250 colonies were counted. Using the colony counts, the colony forming units per ml of the original culture was determined. This value was then divided by the corrected absorbance of the culture to obtain a calibration factor used to convert from absorbance to concentration. The above procedures were then repeated with turbid B. subtilis culture to obtain a second calibration factor. Creating .8 molar solutions of sodium nitrate, sodium phosphate, sodium sulfate, and a combination of the three: Using an analytical balance, 11.364 g of Na2SO4 was measured and added to a 100-ml volumetric flask along with approximately 90 ml of distilled water. The mixture was heated and stirred until the Na2SO4 dissolved completely, and which point distilled water was added to reach a final volume of 100 ml. The above procedures were repeated using 30.402 g of Na3PO4 • 12 H2O and 6.798 g NaNO3 to finish making .8 M solutions. The above procedures were also repeated using 3.835 g Na2SO4, 2.290 g NaNO3, and 10.261 g Na3PO4 • 12 H2O in the same flask to make the combination nutrient solution. All nutrient solutions were autoclaved. Plating Pseudomonas fluorescens and Bacillus subtilis: A 100 μ1 sample of turbid P. fluorescens and turbid B. subtilis cultures, respectively, were transferred to separate sterile agar Petri dishes using standard flaming techniques to avoid contamination. The Petri dishes was spun to spread the culture evenly and placed in an incubator at 27°C. Gram staining P. fluorescens and B. subtilis (figure 2): Using a sterile inoculating loop, 1 loop of P. fluorescens was smeared on a clean microscope slide. The slide and smear were flamed briefly using a Bunsen burner. The smear was covered with crystal violet, the primary stain, for 20 seconds, at which time the smear was rinsed using a wash bottle of distilled water. The smear was then covered with Gram’s iodine solution for one minute, at which time the smear was flooded with isopropyl alcohol until the solvent flowed colorlessly from the slide. Rinsing the slide with a Meghan Shea, Sandra Litvin, and Anastasia Chirnside Page 3 of 8 32 wash bottle for a few seconds stopped the action of the alcohol. Then, the smear was covered with safranin, the secondary stain, for 20 seconds. The smear was washed gently for a few seconds, blotted with bibulous paper, and let dry at room temperature. The above procedures were repeated with 1 loop of B. subtilis. Both slides were viewed under a microscope using oil immersion. Preparing samples: 500 ml of Instant Ocean were prepared using the accompanying standard procedure and sterilized using an autoclave. 34 Spec-20 tubes and corresponding stoppers were labeled 1-34 and sterilized through submersion in alcohol. Using aseptic technique, the non-bacterial components in (figure 3) were added to the respective samples using the appropriate pipettes and micropipettes. After the non-bacterial components were added, the absorbance of each tube was determined using a Spec-20 at 686 nm blanked with distilled water. Using aseptic technique, the bacterial components from (figure 3) were added to the respective samples using a 500-μ1 micropipette sterilized with alcohol with sterile tips. The absorbance of each tube was determined once more using a Spec-20 at 686 nm blanked with distilled water. Setting up experimental conditions: The 34 samples were secured in an orbital shaker set at 27°C and 80 RPM. The flasks shook for 14 days, destructively sampling on day 7 and day 14. Every four days, the caps of each tube were quickly removed and replaced to aerate and samples. Every day, the absorbance of each sample was measured using a Spec-20 at 686 nm blanked with distilled water. On day 7 and day 14, the control samples were plated and Gram-stained using procedure described above. Separating waste and hydrocarbon layers using Solid Phase Extraction (SPE): 24 test tubes were placed in an Absorbex (a SPE apparatus) set to waste, and 24 Strata X columns (33-μ polymeric reversed phase, 300 mg) were attached to the top of the Absorbex. To condition, 2 ml of Acetonitrile was run through each column. To equilibrate, 2 ml of water were run through each column. At this point, the Absorbex was changed to collect. 24 samples were run through each labeled column at a rate of 1-2 ml per minute. A few samples did not filter through the columns adequately under vacuum so they were left overnight under vacuum to finish loading. Once the samples were loaded onto the columns, the samples were washed with 2 ml of 40:60 ethanol:water (v/v). Then, the samples were dried for 10 minutes under full vacuum. At this point, the Absorbex was changed to collect and samples were eluted with 4 ml of Acetonitrile. The sample test tubes were removed from the Absorbex, placed in a 40°C water bath and evaporated to dryness under a gentle stream of nitrogen gas. Then, the samples were brought back up into 5 ml of hexane, diluted (0.2 ml sample: 1 ml hexane) into GC vials and capped. The above procedures were repeated until all samples were separated. Analyzing samples using GC (figure 4): 1 microliter of each sample was injected into an Agilent 6890 N Network GC system (University of Delaware) with a Flame Ionization Detector and Agilent 25-m long x 0.2 mm diameter ultra-2 column with a .33-μ1 coating of (5% phenyl)-methylpolysiloxane. The samples were injected with a split ratio of 1/50. The oven was ramped at 8°C/min from 70°C to 200°C and then ramped at 30°C/min from 200°C to 280°C. The gas chromatograms produced were then analyzed using the Sherlock program. Fatty Acid Methyl Ester testing to verify identity of bacteria (adapted from [5]): For FAME analysis, one loop of bacterial cells from the isolated colonies on each plate was added to a 20-ml test tube and closed with a Teflon-lined screw cap. Each tube received 1 ml of saponification reagent and was mixed (vortex mixer) and heated at 100°C for 30 minutes to liberate fatty acids from the cellular lipids. Once cooled, each tube received 2 ml of methylation reagent, was mixed again and heated at 80°C for 10 minutes to form methyl esters of the fatty acids. The FAMEs were extracted from this solution by adding 1.25 ml of one part methyl-tert-butyl ether in one part hexane (v:v) (ultrapure grade, VWR Scientific Products, Bridgeport, NJ) and placing the closed tubes on an end-over-end mixer for 10 minutes. Finally, each tube was centrifuged for 10 min at 10,000g, the bottom aqueous layer removed with a pasteur pipette and the supernatant washed with 3 ml of dilute NaOH (0.27M NaOH). The extracted fatty acids were analyzed by a Hewlett Packard (HP) gas chromatograph (Wilmington, Figure 2. A) Gram-stained B. subtilis under oil immersion before trial. B) Gram-stained B. subtilis under oil immersion after trial. C) Gramstained P. fluorescens under oil immersion before trial. D) Gramstained P. fluorescens under oil immersion after trial. Figure 3.