SOIL MICROBIAL ACTIVITY AND PLANT/MICROBE SYMBIOSES AS
INDICATORS FOR ECOLOGICAL EFFECTS OF BIOREMEDIATION BIOTECHNOLOGY
W. F. Pfender,a S. P. Maggard,b and L. S. Watrudc,(1)
aKansas State University, Department of Plant Pathology, 414 Throckmorton Hall, Kansas State
University, Manhattan, Kansas 66506. (913) 532-6176, FAX (913) 532-5692;
bMantech Environmental Technology Inc., 200 SW 35th Street, Corvallis, OR 97333. (503)
754-4600, FAX (503) 754-4799; and
cUS EPA Environmental Research Laboratory, 200 SW 35th Street, Corvallis, OR 97333. (503)
754-4874, FAX: (503) 754-4799, firstname.lastname@example.org
Pentachlorophenol-contaminated soil was amended with a strain of Pseudomonas capable
of biodegradation of the pollutant. After bioremediation was complete, soil was tested for
indicators of ecological condition. Several direct-toxicity measures (plant seed germination and
root elongation, earthworm survival) showed the bioremediated soil to be indistinguishable from
clean reference soil. However, preliminary results suggest that plant-microbe symbioses may have
been affected in bioremediated soil. That is, Rhizobium-induced nodules on roots of birdsfoot
trefoil were significantly smaller on plants grown in bioremediated soil and formation of VA
mycorrhizae in bioremediated and polluted soils was increased above that in clean soil. In
addition to these effects on symbiosis formation, bioremediated soil differed from clean soil in the
ability of the soil microflora to metabolize lignocellulose (the major form of carbon entering most
soil ecosystems). Plant-microbe symbiosis formation and integrated soil microflora activity may
have greater sensitivity than direct toxicity as measures of soil ecological condition following
bioremediation by an introduced biotechnology product. Follow-up studies using a broader range
of plants and microbial symbionts are planned.
Key Words: Bioremediation, mycorrhizae, Rhizobium, pentachlorophenol, symbiosis
Non-engineered and engineered bacterial and fungal agents and increasingly plants, are
being evaluated and developed for in situ remediation of pollutants in surface soils. Attractions of
in situ remediation include (a) potentially lower cost as compared to the use of bioreactors,
chemicals or incineration methods (b) reduced environmental exposure to pollutants in
contaminated soils, since they would not need to be transported to treatment sites or facilities; (3)
potential for treatment of larger areas containing low to moderate levels of pollutants. (King et
al., 1992; USEPA, 1992; Watrud et al., 1994; Manasse and Watrud, 1994).
One of the objectives of the USEPA Biotechnology Risk Assessment Program is to
develop methods that can be used to assess the ecological effects of releasing biotechnology
products including bioremediating agents, to the environment. Accordingly, in our research, a set
of laboratory assays are being developed to assess the ecological effects of bioremediating agents.
Initially, the methods are being developed with non-genetically engineered organisms; as
genetically-engineered strains become available, the methods can be tested with those as well.
To develop our tests, we have selected pentachlorophenol (PCP) as the pollutant. Most
microorganisms lack the ability to degrade it (Kitunen et al., 1987; Mueller et al., 1991),
therefore it is a reasonable target for bioremediation by nonindigenous organisms. However,
several different fungi (Lamar and Dietrich, 1990) and bacterial strains (Middeldorp et al., 1990;
Seech et al., 1991; Radehaus and Schmidt, 1992) have been reported to degrade PCP, so there
are organisms available for us to test. We are using a bacterial strain, Pseudomonas, and the
white-rot fungus, Phanerochaete, as bioremediating agents.
MATERIALS AND METHODS
Our approach in this work is to contaminate soil with PCP, treat the contaminated soil
with bioremediation agents in the laboratory, then test the bioremediated soil for parameters of
interest. Results of these laboratory experiments will ultimately be compared with results from
PCP bioremediation sites in the field, to validate our set of assays. Thus far, we have worked
with bioremediating agents only in laboratory microcosms. Beginning in the summer of 1994,
laboratory validation studies will be conducted in conjunction with a field demonstration of PCP
There are two major questions we are asking. First, do the organisms produce any
biotransformation products that may cause a problem? The experiments to determine this are
done with 14C radiolabelled PCP, and we do a mass-balance determination on the fate of the
pollutant. The second type of question we are studying is whether there are adverse ecological
effects from the bioremediation, due to toxic metabolites or other activities or effects of the
introduced organisms. This presentation will focus on the question of ecological effects
measurement. But first, we will present briefly the results of our mass-balance experiments, to
provide the context for the ecological assays.
The mass-balance experiment is done to determine whether the bioremediation agent
mineralizes the pollutant to harmless products, whether it produces known toxic materials, or
whether it produces unknown metabolites with unknown toxicity. These experiments are done
with 14C radiolabelled PCP. Briefly, these are the steps in the experiments: we contaminate soil
with uniformly-labelled PCP (specific activity approximately 1%) using a total of 175ppm PCP in
the soil. We apply the bioremediation agent or the appropriate control treatment: controls consist
of sterile liquid medium for the bacterial strains or sterile wood-chips for the fungus which is
grown on sawdust. The treated soil is incubated in closed jars with a large headspace, and the
headspace is flushed twice weekly to aerate the system and to collect headspace gas for analysis
of CO2 and organic volatile compounds. At the beginning, middle and end of the incubation
period, we collect soil samples for analysis. Soil is extracted with a one-week methylene-chloride
treatment, followed by Soxhlett extractions with methylene chloride and then by hexane/methanol.
These extracts are analyzed by LSC and by HPLC. After extracting as much of the material from
soil as possible, the residual, nonextractable level of radiolabel is determined by analytical
combustion of the soil sample.
We used two types of bioremediation agents in these experiments, and we found that the
white-rot fungus mineralized only a small proportion of the PCP to CO2; most of the radiolabelled
PCP was converted to pentachloroanisole and was extractable. The bacterial strain, a
Pseudomonas sp. #SR3 originally isolated by Sol Resnick while working at the EPA Gulf Breeze
Lab, mineralized most of the PCP to carbon dioxide after 6 weeks. A proportion of the original
pollutant remained as extractable PCP, and about 15% of it was converted to a nonextractable
material. The possible ecological hazard with this situation, as an example, is that the remaining
PCP or its soil-bound metabolites are toxic to some aspect of the soil-associated biota, and/or that
the soil biota has been changed by the introduction of the Pseudomonas in such a way that normal
biological functions of the soil are disrupted. To determine whether the bioremediation renders
the soil suitable for plant growth, we conducted the second type of experiment with soil that had
been contaminated with non-radiolabelled PCP and subsequently bioremediated by the
For these experiments, we needed to produce bioremediated soil, then assess the
ecological parameters of interest. As in the mass-balance experiments, the soil was contaminated
with PCP at 175 ppm, but the PCP was not radiolabelled. Bacterial inoculum was added as a
suspension, and the soil was incubated with periodic aeration for 6 weeks. The experiment
included two types of controls: PCP-contaminated soil that was not treated with the bacterial
strain, in order to see what the ecological condition was for soil that was incubated as the test soil
but without bioremediation; and non-contaminated soil, to have a baseline for the ecological
measurements from nonpolluted soil incubated as the test soil was. At the end of 6 weeks'
incubation, samples of the soil were taken and analyzed for PCP level. Having ascertained that
the extractable PCP level was similar to what we had seen in the mass-balance experiment, we
took samples of this bioremediated soil for our bioassays and ecological measurements.
RESULTS AND DISCUSSION
The assays we develop are intended to measure the ecological condition of soil with
respect to primary production, that is, plant growth. There are several types of indicators we are
using to evaluate the ability of post-bioremediation soil to support plant life. Some assays are for
direct toxicity to plants - we are using seed germination and root elongation assays, exposing 5
genera of plants to whole soil (not simply to extracts from the soil). We also expose earthworms
to whole soil and test for survival during 3 weeks. These direct toxicity tests show that after
bioremediation the soil is similar to the non-polluted check soil. That is, there is no significant
direct toxicity to seed germination, plant root elongation, or earthworm survival in soil that has
been bioremediated by the Pseudomonas isolate. In the polluted, nonbioremediated soil there is
significant toxicity as measured by all of these parameters.
However, we have indications that some other measures of plant and soil biological
function may be more sensitive indicators of ecological condition. These are (1) plant-microbe
symbioses, and (2) carbon processing ability of the soil community.
Formation of plant/microbe symbioses. Symbiotic nitrogen fixation is the most efficient
way for fixed nitrogen to get into the biosphere. We tested effects of bioremediated soil on this
symbiosis by planting Rhizobium-coated legume seeds into tubes containing the test soil, and
incubating for 4 weeks while the roots grew. We found that in bioremediated soil, there appeared
to be a reduction in number of Rhizobium-induced nodules, but it was not statistically significant.
There were, however, significantly fewer large nodules formed on the roots grown in
bioremediated soil, as compared with the control treatment. We had previously found, in
preliminary assays with PCP-contaminated soil, that 175 ppm of PCP completely prevented
nodule formation, and 25 ppm PCP caused a significant reduction in total number of nodules.
This same level of 25 ppm PCP had been non-toxic by most direct measures, i.e. earthworm
survival, seed germination, and root elongation for most plant species. The reduction in nodule
size we saw in the bioremediated soil may have been due to toxicity of residual PCP or
metabolites, or to antagonistic interactions between Rhizobium and the altered microflora in the
rhizosphere of plants grown in bioremediated soil.
Another key symbiosis is formed between most plant species and vesicular-arbuscular
mycorrhizal (VAM) fungi. VAM fungi provide plants with essential phosphorus, and may protect
roots against drought stress or infection by root pathogens. We used the fungus Glomus
deserticola and a grass host. We grew the plants in tubes containing test soil amended with a
known inoculum of the VAM fungus (100 spores per 5 g soil), and measured % root infection
after 6 weeks incubation. Here we found that the extent of infection by the symbiont is not
reduced by PCP at our starting concentration (175 ppm), nor is it reduced in bioremediated soil.
In fact, we saw that the presence of PCP produced a significant increase in mycorrhizal infection,
and that the bioremediated soil was similarly stimulatory. The PCP may have killed some
microbes that are antagonistic to the mycorrhizal fungi, and this change in soil microflora may
have persisted even after the PCP was removed by the Pseudomonas. Another possibility is that
the increased levels of infection represent a shift from a beneficial relationship to a pathogenic
one, as sometimes occurs when mycorrhizal host plants are stressed; this could be resolved by
further experiments comparing mycorrhizal and non-mycorrhizal plants, to determine whether the
highly-infected mycorrhizal plants in this soil are producing less biomass than non-mycorrhizal
plants in the same soil. Whatever the case, we note here that a change in mycorrhizal infection due
to PCP contamination was not rectified by the bioremediation of the PCP by Pseudomonas SR3.
As estimated by vital staining with fluorescein diacetate, the active microbial biomass was
changed by PCP and by bioremediation of PCP in soil. In PCP treated controls, there was very
little active fungal biomass. Similarly, population of active bacteria was slightly lower in PCP-treated soil than in noncontaminated soil. Bioremediation of the PCP had an influence on
bacterial and fungal populations: total active fungal biomass was greater in bioremediated soil
than in either control soil, and bioremediation resulted in a total active bacterial biomass that
appeared somewhat larger, though statistically similar to, that in the non-contaminated soil.
Carbon processing by soil microbes. The ability of soil microorganisms to process
organic carbon is a critical aspect of soil ecosystem functioning. The rate of processing for
lignocellulose (the major form of organic carbon entering the soil as detritus) affects the level of
soil organic matter, with consequent effects on soil structure, and moisture- and nutrient-holding
capacities. Respiration from lignocellulose was significantly depressed in the bioremediated soil,
in comparison with that in the nonpolluted control soil. Because degradation of lignin and
cellulose is primarily a fungus-mediated activity, this inhibition of lignocellulose respiration
indicates an adverse effect on fungi or their metabolism. PCP is toxic to most fungi. To see
whether the problem is lack of fungal inoculum to recolonize the remediated soil, we re-inoculated the bioremediated soil with a small amount (5%) of fresh, noncontaminated soil and
incubated it for an additional 6 weeks. But at the end of that time, the bioremediated soil still had
a significantly reduced respiration from lignocellulose. One could speculate that in the
bioremediated soil, there is an altered population of fungi that normally process lignocellulose,
and/or an altered population of bacteria that are inhibitory to these fungi.
These observed effects, in formation of plant-microbe symbioses and in lignocellulose
metabolism, are not mediated by reduced pH. Reduced pH might be expected due to the
degradation of PCP; but measurements of pH of postbioremediation soil were found to be the
same as that of the nonpolluted control soil.
In summary, in PCP contaminated soil that was bioremediated by a bacterial strain that
mineralizes approximately 75% of the PCP to CO2, there was relatively little residual toxicity in
the soil as measured by seed germination and root elongation of 5 plant species, and by toxicity to
earthworms. There was, however, a change in lignocellulose-based respiration. The change in
ability of the soil to process lignocellulose may possibly reflect a change in community
composition of the fungi, which are the predominant organisms involved in lignocellulose
degradation. Finally, and perhaps most importantly, it was observed that plant-microbe symbioses
may be sensitive indicators of soil ecological disruption. Rhizobium nodule formation and VAM
fungal infection were affected in the remediated soil even though several other plant related
toxicity measures (germination and root elongation) were unaffected. Rhizobial and VAM fungi
plant/microbe symbioses are ecologically important partnerships; particularly in natural or low
agronomic input ecosystems where they represent respectively, the major means of nitrogen and
phosphorous inputs to vegetation in terrestrial ecosystems. Our results reported above suggest
that plant symbioses may also be useful as indicators for the ecological effects and environmental
safety of biotechnology products such as bioremediating agents.
King RB, Longard GM, Sheldon JK (1992) Practical Environmental Remediation. Lewis
Publishers, Boca Raton, LA.
Kitunen V, Valo R, Salkinoja-Salonen M (1987) Contamination of soil around wood-preserving
facilities by polychlorinated aromatic compounds. Environ. Sci. Tech. 21:96-101.
Lamar RT, Dietrich DM (1990) In situ depletion of pentachlorophenol from contaminated soil by
Phanerochaete spp. Appl. Environ. Microbiol. 56:3093-3100.
Manasse RS, Watrud LS (1994) Current Methods in Terrestrial Revegetation. Office of Research
and Development, Washington, DC 600/R-94/000.
Middeldorp P, Briglia M, Salkinoja-Salonen M (1990) Biodegradation of pentachlorophenol in
natural soil by inoculated Rhodococcus chlorophenolicus. Microb. Ecol. 20:123-139.
Mueller JG, Lantz SE, Blattman BO, Chapman PJ (1991) Bench-scale evaluation of alternative
biological treatment processes for the remediation of pentachlorophenol- and creosote-contaminated materials: solid-phase bioremediation. Environ. Sci. Technol. 25:1045-1055.
Radehaus P, Schmidt S (1992) Characterization of a novel Pseudomonas sp. that mineralizes high
concentrations of pentachlorophenol. Appl. Environ. Microbiol. 58:2879-2885.
Seech AG, Trevors JT, Bulman TL (1991) Biodegradation of pentachlorophenol in soil: the
response to physical, chemical, and biological treatments. Can. J. Microbiol. 37:440-444.
USEPA (1992) Bioremediation of Hazardous Wastes. Office of Research and Development,
Washington, DC. 600/R-92/126.
Watrud LS, Metz SG, Fischhoff DA (1994) Engineered Plants in the Environment. Office of
Research and Development, Washington, DC. 600-R-94/XXX.
1. For Reprints: Lidia S. Watrud, US EPA Environmental Research Laboratory, 200 SW
35th Street, Corvallis, OR 97333,