Final Report: Root Exudate Biostimulation for Polyaromatic Hydrocarbon Phytoremediation

EPA Grant Number: R829479C020
Subproject: this is subproject number 020 , established and managed by the Center Director under grant R829479
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: The Consortium for Plant Biotechnology Research, Inc., Environmental Research and Technology Transfer Program
Center Director: Schumacher, Dorin
Title: Root Exudate Biostimulation for Polyaromatic Hydrocarbon Phytoremediation
Investigators: Thomas, John C. , Rugh, Clayton
Institution: University of Michigan - Dearborn
EPA Project Officer: Lasat, Mitch
Project Period: October 1, 2003 through September 30, 2004 (Extended to December 31, 2007)
RFA: The Consortium for Plant Biotechnology Research, Inc., Environmental Research and Technology Transfer Program (2001) RFA Text |  Recipients Lists
Research Category: Hazardous Waste/Remediation , Targeted Research


University of Michigan–Dearborn

  1. Isolate microbes from polyaromatic hydrocarbon (PAH) laden soils at the Miller Road site in Dearborn, Michigan .
  2. Isolate and partially characterize root exudates from several plants being used for phytoremediation.
  3. Develop an exudate mediated growth-stimulation screen for microbes using a microtiter plate. Using this assay, determine a dose for PAH destruction screening (see 4.).
  4. Develop a bench-top PAH degradation screening method and apply the method to at least 100 microbes for the 2003, 2004, and 2005 seasons from beneath selected plant species. This screen compared microbes beneath plants to un planted plots or control soils.
  5. Recover samples from beneath the planted and unplanted plots tested for microbes and analyze for total and individual PAHs using gas chromatography/mass spectrometry (GC/MS) methods.

Michigan State University

  1. Identify plant species exhibiting enhanced PAH biodegradation in contaminated soils.
  2. Quantify biodegrading bacterial community response in rhizosphere soils and purified root exudates of superior phytoremediation plant species.

Develop DNA-based tools for rapid identification and quantification of PAH biodegrader bacteria in rhizosphere and root exudate-stimulated microbial consortia.

Summary/Accomplishments (Outputs/Outcomes):

University of Michigan

Exudate collection was made from healthy pre flowering adult plants obtained from seeds or removed from the field and grown for at least 4 weeks in hydroponic culture. Use of hydroponics greatly facilitated the recovery of root secretions. Before or after exudate collection there was no observed change in the color of roots. These same roots continued to grow, produce side roots, and elongate throughout the experiment. The only exception was the last Swamp Goldenrod exudate sample, 1-11-04. In this sample, a majority of the root-mass was suspended above the nutrient solution 2 weeks after the 12-18-03 exudate collection. The lower one-quarter of the root mass did remain in the remaining nutrient solution. Plants were given new nutrient solution and the exudate recovered 14 days later (1-11-04). Roots continued to grow; however, there was some slight browning of roots.

Exudate characterization was accomplished with the bicinchoninic acid (BCA) Protein Assay Kit (Pierce, Rockford, IL). Protein values averaged about 20 μg/ml with the exception of Swamp Goldenrod 1-11-04, which had over 170 μg/ml total protein (Table 1). Anthrone reagent detected sugars, which were very abundant in most of the root exudates examined. Swamp Goldenrod samples (particularly 1-11-04) contained greater levels of carbohydrate than other exudate samples (Table 1). Amino acid levels in root exudates varied with species and time of sampling, again with Swamp Goldenrod 1-11-04 having the highest levels recorded (Table 1). When individual amino acids were examined, proline was enhanced in many exudate samples (Figure 1). Swamp Goldenrod samples 11-18-03 and 12-2-03 contained more lysine and methionine, later switching to a preponderance of asparagine in the most effective exudate obtained, 1-11-04 (Figure 1). Amino acid profiles of the NE Aster and Boneset exudates showed less dramatic changes as the plants aged (data not shown).

Table 1. Exudate Characteristics. Protein was determined using the Pierce BCA Protein Assay Kit (Pierce, Rockford IL). Carbohydrate was determined using the anthrone method (Scott TA, Melvin EH, 1953). Amino acid analysis was performed in the University of Michigan Protein Structure Facility (Dr. Henriette A. Remmer). Data are expressed as per volume of exudate-reconstituted exudate (0.5 ml per plant).


Protein mg/ml Exudate

Carbohydrate mg/ml Exudate

Total Amino Acids nM/ml Exudate

Swamp Goldenrod 11-18-03

24.8 +/- 6.6

785.1 +/- 109.2


Swamp Goldenrod 12-12-03

19.5 +/- 1.1

921.9 +/- 23.5


Swamp Goldenrod

174.0 +/- 25

8,613.0 +/- 631.6


NE Aster 8-15-03

38.5 +/- 4.3

124.0 +/- 20.0


NE Aster 9-2-03

20.1 +/- 0.1

44.4 +/- 11.0


NE Aster 9-16-03

15.0 +/- 3.2

151.7 +/- 51.2


Boneset 10-13-03

7.4 +/- 1.1

178.9 +/- 64.6


Boneset 10-25-03

24.8 +/- 6.6

193.9 +/- 40.6


Amino Acid (% nmol) from Swamp Goldenrod Exudate

Figure 1. Amino Acid (% nmol) from Swamp Goldenrod Exudate. Plants were grown in hydroponic solutions (changed every 2 weeks). After exudate recovery, plants were allowed to recover for 4 weeks. Early Swamp Goldenrod indicates exudates from 11-18-03 and 12-18-03. Exudate 1-11-04 was obtained from previously stressed plants.

Microbes from the field-grown NE Aster, Boneset, Swamp Goldenrod, and Bluestem were screened and counted. Soil samples from 6 to 8 inches below these plants were recovered in July 2003, July 2004, and August 2005. Microbes from soil of the subplot of single -species NE Aster (lower elevation) and double -species NE Aster/ Bluestem ( upper elevation ) were recovered, as well as the single -species Boneset ( lower elevation ) and single -species Green Bulrush ( lower elevation ). Control soil was obtained from underneath the tarp liner beneath the gravel road, between the ( lower elevation ) Bulrush, NE Aster, and Boneset plots. In addition, an unplanted within-plot control was also used in 2005 -2006. Microbes were recovered, plated on yeast extract peptone glucose (YEPG) medium, and more than 100 colonies per subplot were isolated to homogeneity. The colony forming units (CFUs) are depicted in Figure 2 .

Microbes beneath Boneset and Green Bulrush increased significantly from 2003 -2004 to the 2004 -2005 season. In the last season, there was a precipitous decline in bacterial population numbers. This particular year was the warmest as well as the six th driest summer in the Detroit metropolitan area on record (National Weather Service Forecast Office; Detroit-Pontiac ).

Microbial populations under NE Aster seemed more constant from year to year (Figure 2). Because all the soil in all plots began as a common mixture, changes in microbial populations over time are likely the direct result of particular plant -microbe interaction(s) as well as localized microbial species shifts.

Microbial Population Density Under Phytoremediating Plants for Three Seasons

Figure 2. Microbial Population Density Under Phytoremediating Plants for Three Seasons. The “nt” is not tested. Data are expressed as the mean and standard deviation of microbial dilutions and triplicate plate counts in at least two experiments.

Exudates and Growth. Initially a set of 40 microbes from the NE Aster/Bluestem plot in 2003-2004 were screened for growth promotion in the presence of NE Aster exudate (9-16-03) 0.7% (v/v) in YEPG medium with growth measured as an increase in optical density at 600 nm over time. Seven isolated microbes were found to be growth responsive to exudate. In growth stimulation versus dose experiments with the NE Aster exudate, an exudate dose of 0.7% - 0.07% (v/v) was found to enhance growth of all seven individual microbes (Figure 3). For this reason, 0.35% (v/v) exudate was chosen for further PAH degradation enumeration because this value was above the growth stimulation threshold.

Approximately 9.8 percent of over 100 cultured microbes from NE Aster/Bluestem soil responded to the NE Aster exudate, while 5.7 percent of over 100 microbes beneath Boneset responded to NE Aster e xudate. Extensive growth analysis was not continued. Interestingly, none of the microbes showing growth acceleration in the presence of Aster exudate showed PAH degradation (as described below). UPD 1, 6, and 7 (verified PAH -degrading microbes) were growth-stimulated by NE Aster exudate (data not shown).

Figure 3. Growth Response of Selected Microbes From the NE Aster/Bluestem Plot to NE Aster Exudate on Microbes. Microbes were initially identified as “growth enhanced” in preliminary experiments. The legend indicates the μl of exudate per 150 μl medium compared to medium alone. Triplicate samples were shaken at room temperature throughout the experiment. “Media only” refers to no bacteria (medium-alone). Th ese data reflect the mean of triplicate samples of growth measured at 8 hours compared to media alone samples.

PAH Degradation by Individual Microbes Beneath Phytoremediation Plants. To measure PAH degradation, a color-change assay was adapted for the use of root exudates. The color-change assay used a mixture of phenanthrene, fluorene, and dibenzothiophene. Known PAH -degrading organisms (UPD 1, 6, and 7) produced dark colored wells in the presence of the PAH cocktail, medium, and 0.35% (v/v) of Swamp Goldenrod (1-11-04) exudate. Other exudates also stimulated the production of colored compounds, albeit less intensely than Swamp Goldenrod 1-11-04. For this reason, Swamp Goldenrod exudate (1-11-04) was used for large -scale screening of soil microbes beneath phytoremediation plants from the 2003-2004 and 2004-2005 seasons. The next most potent exudate was from Boneset 11-15-03 (data not shown).

After analyzing over 100 individual microbes from beneath each plant species for the 2004-2005 and the 2005-2006 seasons, the change in the percentage of PAH degrader organisms was determined. T he overall percentage of PAH degraders was greater in 2004-2005 than 2005-2006 (Table 2). This is also reflected in the percentage degraders in unplanted controls (20% versus 2.7%). The fraction of degraders in NE Aster was similar each season. Using the controls as a seasonal control, the ratio of percent degraders over control indicated that plants helped maintain a degrader population in 2005-2006, unlike the two unplanted controls.

Table 2. % of PAH Degraders as Determined by Scores of 2 or Higher in Color-Change Assays With Swamp Goldenrod 1-11-04 Exudate. Over 100 individual organisms from each subplot were screened.


% PAH degraders 2004-2005

Relative to unplanted control

% PAH degraders 2005-2006

Relative to unplanted control

Unplanted control (under road tarp)





Within-plot control

Not tested

Not tested



Green B ulrush




















Boneset, NE Aster, and Green Bulrush maintained significant degrader populations in 2005-2006, unlike the control subplots. The one “higher elevation” subplot (NE Aster/Bluestem) was less able to promote PAH degraders in both years surveyed.

Determination of Soil PAHs and Total PAH Levels. Soil analysis of PAH content was done using U.S. Environmental Protection Agency method 8270C (GC/MS) by Kemron Environmental Services, Marietta, Ohio. Figure 4 demonstrates the mean and standard deviation of at least three samples from each plot tested.

Analysis of Total PAHs From Samples 6-8 Inches in Depth All Taken in September 2005. “Pre-planting 2002” represents the values obtained from mixed soil prior to planting in 2002

Figure 4. Analysis of Total PAHs From Samples 6-8 Inches in Depth All Taken in September 2005. “Pre-planting 2002” represents the values obtained from mixed soil prior to planting in 2002. Using an unpaired 2-tail student t -test, ** indicates this sample is significantly different than the adjoining within-plot control sample (P > 0.028).

Smaller and more volatile PAHs may be most easily remediated, whereas the larger PAH molecules seem to be more persistent. A comparison of individual PAH species was made between Green Bulrush and the adjoining unplanted within-plot control. As can be seen in Figure 5, significant amounts of smaller ringed PAHs (fluoranthrene, pyrene, and phenanthrene) and some larger molecular weight PAHs (benzo(a)pyrene and benzo(b) fluoranthene) were removed by this phytoremediation process.

Comparison of the Mean and Standard Deviation of Unplanted Within-Plot Control (N = 4) and Green Bulrush (N = 3) From 2005 -2006. Y-axis = ppb of the PAH species as indicated on the X-axis.

Figure 5. Comparison of the Mean and Standard Deviation of Unplanted Within-Plot Control (N = 4) and Green Bulrush (N = 3) From 2005 -2006. Y-axis = ppb of the PAH species as indicated on the X-axis.

Main Conclusions. Because PAH breakdown (as measured by colored compound formation) was extensive in some microbial isolates in the presence of exudate and glucose, co-metabolism, catabolite activation, or facilitated population expansion seems the most likely role of root exudates. Making this statement, we are aware that cultivable bacterial populations make up only a fraction of the rhizosphere. Furthermore, using an aqueous extraction procedure (0.1% w/v sodium pyrophosphate) may not release all soil organisms capable of in vivo PAH breakdown. Acknowledging these points, it is clear that using the methodology described herein, planted subplots did lead to higher cultivable microbial numbers, a higher fraction of exudate-stimulated PAH degraders, and up to a 30 percent reduction in total PAHs (tPAHs) in the contaminated soil compared to unplanted plots. Using exudates to distinguish the microbial PAH degraders from non degraders in vitro, we conclude that both plants and the microbes beneath them are linked in a cooperative process of exudate production, mono- and dioxygenase synthesis, and microbial community construction that likely contributes to PAH destruction during phytoremediaton.

Michigan State University

Our initial field trials indicated that individual plant species affect PAH biodegradation rates to different extents (Figure 6). Most treatments achieved 20 percent reduction of soil PAH after the first season, though a limited number of planted treatments displayed continued reduction to approximately 40 percent over the 3-year field study, e.g., cardinal flower (Lobelia cardinalis), New England aster (Aster novae-anglicae), green bulrush (Scirus atrovirens), and big bluestem (Andropogon gerardii).

Percentage Reduction of Total PAH (tPAH) Content in the Amended Coke Oven Soils by the Different Planted Phytoremediation Treatments After 1, 2, and 3 Growing Seasons

Figure 6. Percentage Reduction of Total PAH (tPAH) Content in the Amended Coke Oven Soils by the Different Planted Phytoremediation Treatments After 1, 2, and 3 Growing Seasons. The unplanted treatment is shown as “control.” Planted treatments with statistically greater soil tPAH reduction than the unplanted, control treatments after 3 growing seasons (37 months) are indicated with * (α = 0.05).

Numerous laboratory studies have examined the mechanistic basis of PAH phytoremediation. Due to the demonstrated plant-microbe interaction, this process is termed “rhizosphere-assisted bioremediation” and is based upon the fundamental hypothesis:

Effective phytoremediation plant species produce substances in root exudates that induce bacterial biodegrader activity, promoting PAH destruction.

In order to examine this hypothesis with the ultimate goal of optimizing management of root- microbe processes for enhanced PAH bioremediation, we are studying microbial community responses to specific rhizosphere environments and products. To develop a mechanistic understanding of the PAH phytoremediation process, we tested various analytical methods to identify the PAH Spray Metabolism Assay as a reproducible, PAH biodegrader indicator, allowing isolation of confirmed degrader bacterial strains for further study. Soil bacterial extracts were cultured on Petri plates oversprayed with a cloudy residue of phenanthrene, which is the most common “tester” PAH compound, to quantify PAH biodegrader abundance among the different planted soil treatments by formation of cleared zones around the bacterial colonies. We obtained ~2,100 phenanthrene biodegrader 1° isolates representing rhizosphere bacterial community subsamples from each of the 20 different planted field treatments for further study. For broad-spectrum analysis, Spray Assays were repeated using phenanthrene and the additional PAH compounds anthracene, fluoranthene, and pyrene on separate replica plates. The multiple PAH compound study demonstrated that soil bacterial communities from phytoremediation treatments typically had much broader spectrum metabolic capabilities for a wider range of PAH compounds than unplanted or untreated soils (Figure 7).

Percentage of PhenD Bacterial Isolates From Each of the Planted and Unplanted Soil Treatments Capable of Metabolizing Multiple PAH Compounds (% mPAH ZFU)

Figure 7. Percentage of PhenD Bacterial Isolates From Each of the Planted and Unplanted Soil Treatments Capable of Metabolizing Multiple PAH Compounds (% mPAH ZFU). Data bars represent the percentages of specific treatment pools of PhenD isolates capable of degrading anthracene (black bar = AnthD), fluoranthene (gray bar = FlraD), or pyrene (white bar = PyreD) in Spray Test Assays. The unplanted (UNPLNT) and untreated Rouge (UNTROU) soil bacterial isolates are the sets of bars at the far right of the x-axis. Planted soil community isolates are all others. Planted treatments with total multiple PAH degrader percentages higher than unplanted are indicated with * (α = 0.05).

Note that several of the more metabolism-enriching treatments are also those shown to enhance PAH degradation in field trials: AND GER = Andropogon gerardii, AST NOV = Aster novaeanglicae, LOB CAR = Lobelia cardinalis, and SCI ATR = Scirus atrovirens. By contrast, the unplanted (UNP LNT) and untreated Rouge ( UNT ROU) soils displayed extremely limited ranges for multiple PAH metabolism. These observations indicate that, in addition to increased total bacterial or degrader cell densities and more importantly for mixed chemical contaminants such as PAHs, rhizostimulation by some plant species may enhance bacterial metabolic competence against a broader range of target compounds (Rugh, et al., 2005).

DNA-based methods of bacterial quantification (e.g., real-time PCR) could be used to further characterize the differential rhizosphere effect caused by individual or mixed plant species treatments. We hypothesize that plant species-specific rhizosecretions play an additional role for biostimulation of broad-spectrum degrader bacterial strains by promoting enhanced multiple compound biodegrading competency, perhaps by exchange of mobile, dioxygenase-encoding DNA elements among bacterial populations in the contaminated root zone. One approach to characterize the bacterial community response is targeted DNA primer sets for amplification of bacterial strain-specific and generic dioxygenase gene sequences from phytoremediation -treated soil bacterial community and isolate preparations. Effective dioxygenase amplifying primer sets could facilitate subsequent analysis of biodegrader cell density, PAH substrate specificity among isolates and soil consortia, and monitoring of horizontal exchange of PAH metabolism genes among peripheral bacterial populations. Elucidation of these complementary processes will contribute to our understanding and more effective management of plant-based cleanup of PAH -contaminated soil and sediment media.


Scott TA, Melvin EH. Determination of dextran with anthrone. Analytical Chemistry 1953;25(11):1656-1661.

Journal Articles on this Report : 2 Displayed | Download in RIS Format

Other subproject views: All 34 publications 4 publications in selected types All 2 journal articles
Other center views: All 211 publications 48 publications in selected types All 44 journal articles
Type Citation Sub Project Document Sources
Journal Article Rugh CL. Genetically engineered phytoremediation: one man's trash is another man's transgene. Trends in Biotechnology 2004;22(10):496-498. R829479C020 (2005)
R829479C020 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct Full Text
  • Journal Article Rugh CL, Susilawati E, Kravchenko AN, Thomas JC. Biodegrader metabolic expansion during polyaromatic hydrocarbons rhizoremediation. Zeitschrift fur Naturforschung C 2005;60(3-4):331-339. R829479C020 (2005)
    R829479C020 (Final)
  • Abstract from PubMed
  • Abstract: Z. Naturforsch
  • Supplemental Keywords:

    bioremediation, carcinogen, contamination, phytodegradation, pollutant, toxicity, PAH breakdown, in situ bioremediation, phytoremediation, plant exudates,, Scientific Discipline, Waste, TREATMENT/CONTROL, POLLUTANTS/TOXICS, Sustainable Industry/Business, Treatment Technologies, Geochemistry, Technology, Chemicals, New/Innovative technologies, Bioremediation, Agricultural Engineering, bioengineering, biodegradation, root exudate biostimulation, transgenic plants, biotechnology, plant biotechnology, environmental engineering, remediation, PAHs, hydrocarbons, bioacummulation, phytoremediation

    Progress and Final Reports:

    Original Abstract
  • 2004 Progress Report
  • 2005 Progress Report
  • 2006
  • 2007

  • Main Center Abstract and Reports:

    R829479    The Consortium for Plant Biotechnology Research, Inc., Environmental Research and Technology Transfer Program

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R829479C001 Plant Genes and Agrobacterium T-DNA Integration
    R829479C002 Designing Promoters for Precision Targeting of Gene Expression
    R829479C003 aka R829479C011 Biological Effects of Epoxy Fatty Acids
    R829479C004 Negative Sense Viral Vectors for Improved Expression of Foreign Genes in Insects and Plants
    R829479C005 Development of Novel Plastics From Agricultural Oils
    R829479C006 Conversion of Paper Sludge to Ethanol
    R829479C007 Enhanced Production of Biodegradable Plastics in Plants
    R829479C008 Engineering Design of Stable Immobilized Enzymes for the Hydrolysis and Transesterification of Triglycerides
    R829479C009 Discovery and Evaluation of SNP Variation in Resistance-Gene Analogs and Other Candidate Genes in Cotton
    R829479C010 Woody Biomass Crops for Bioremediating Hydrocarbons and Metals
    R829479C011 Biological Effects of Epoxy Fatty Acids
    R829479C012 High Strength Degradable Plastics From Starch and Poly(lactic acid)
    R829479C013 Development of Herbicide-Tolerant Energy and Biomass Crops
    R829479C014 Identification of Receptors of Bacillus Thuringiensis Toxins in Midguts of the European Corn Borer
    R829479C015 Coordinated Expression of Multiple Anti-Pest Proteins
    R829479C016 A Novel Fermentation Process for Butyric Acid and Butanol Production from Plant Biomass
    R829479C017 Molecular Improvement of an Environmentally Friendly Turfgrass
    R829479C018 Woody Biomass Crops for Bioremediating Hydrocarbons and Metals. II.
    R829479C019 Transgenic Plants for Bioremediation of Atrazine and Related Herbicides
    R829479C020 Root Exudate Biostimulation for Polyaromatic Hydrocarbon Phytoremediation
    R829479C021 Phytoremediation of Heavy Metal Contamination by Metallohistins, a New Class of Plant Metal-Binding Proteins
    R829479C022 Development of Herbicide-Tolerant Energy and Biomass Crops
    R829479C023 A Novel Fermentation Process for Butyric Acid and Butanol Production from Plant Biomass
    R829479C024 Development of Vectors for the Stoichiometric Accumulation of Multiple Proteins in Transgenic Crops
    R829479C025 Chemical Induction of Disease Resistance in Trees
    R829479C026 Development of Herbicide-Tolerant Hardwoods
    R829479C027 Environmentally Superior Soybean Genome Development
    R829479C028 Development of Efficient Methods for the Genetic Transformation of Willow and Cottonwood for Increased Remediation of Pollutants
    R829479C029 Development of Tightly Regulated Ecdysone Receptor-Based Gene Switches for Use in Agriculture
    R829479C030 Engineered Plant Virus Proteins for Biotechnology