FEMS Microbiology Ecology 41 (2002) 191^197

www.fems-microbiology.org

Interactions of earthworms with indigenous and bioaugmented

PCB-degrading bacteria
Ekawan Luepromchai a , Andrew C. Singer b , Ching-Hong Yang b ,
David E. Crowley a;b;
a
Environmental Toxicology Graduate Program, University of California, Riverside, CA 92521, USA
b
Department of Environmental Sciences, University of California, Riverside, CA 92521, USA

Received 19 February 2002; received in revised form 19 April 2002; accepted 2 May 2002

First published online 10 July 2002

Abstract

Partial bioremediation of polychlorinated biphenyl (PCB)-contaminated soils has been achieved using bioaugmentation with PCB-
degrading bacteria and earthworms. To further study the contribution of earthworms to bioremediation, an experiment was conducted in
which the changes in indigenous and bioaugmented PCB-degrading bacteria were analyzed during treatment of contaminated soil using
earthworms (Pheretima hawayana) alone or in combination with the PCB-degrading bacteria, Ralstonia eutrophus and Rhodococcus sp.
ACS. Bacteria used for bioaugmentation were induced with carvone and salicylic acid in culture and were repeatedly applied every 3^4
days to the surface of unmixed, 20-cm long soil columns containing 100 ppm Aroclor 1242. After 9 weeks of treatment, the soil bacterial
communities were analyzed using PCR primers for the bph genes. Results showed that approximately 50% of the PCBs were removed in
the top 9 cm using a combination of earthworms and bioaugmentation, whereas bioaugmentation or earthworms applied alone were
effective only for removing PCBs from the top 3 cm of the soil columns. Enhanced removal of PCBs caused by earthworms was associated
with an increase in the population size of culturable, indigenous biphenyl-degrading bacteria, and an increase in the level of the bphA and
bphC genes. The results suggest that earthworms facilitate PCB bioremediation by enhancing the dispersal of PCB-degrading bacteria in
bioaugmented columns, as well as providing environmental conditions that favor the growth and activity of indigenous PCB-degrading
bacteria. 8 2002 Published by Elsevier Science B.V. on behalf of the Federation of European Microbiological Societies.

Keywords : bph gene probe; Bacterial community analysis ; PCB bioaugmentation; Earthworm

1. Introduction nous soil micro£ora. This can be accomplished by addi-

Bioremediation is an environmentally sound, and poten-
tially low cost method to clean up soils that are contam-
inated with polychlorinated biphenyls (PCBs) [1]. None-
theless, there is still concern over the reliability and
general e⁄cacy of bioremediation that has precluded the
commercial application of these methods [2^4], and there
is a consensus that further improvements will require a
better understanding of the biological processes that facil-
itate bioremediation [2,5,6]. Bioremediation can employ
either bioaugmentation in which additional bacteria are
added to the soil, or may be achieved by biostimulation
of contaminant degrader populations within the indige-
* Corresponding author. Tel. : +1 (909) 7873785;
Fax : +1 (909) 7873993.
E-mail address : crowley@mail.ucr.edu (D.E. Crowley).
0168-6496 / 02 / $22.00 8 2002 Published by Elsevier Science B.V. on behalf of the Federation of European Microbiological Societies.
PII : S 0 1 6 8 - 6 4 9 6 ( 0 2 ) 0 0 2 9 4 - 5
tion of supplemental nutrients, addition of cosubstrates
that induce enzymes for degradation, the use of surfac-
tants to increase the bioavailability of the contaminant,
or by altering the soil physical and chemical conditions
to enhance microbial activity. During bioaugmentation,
indigenous bacteria may play a complementary role in
the degradation process. In this regard, molecular tech-
niques for microbial community analysis are particularly
useful for monitoring changes in microbial populations
during xenobiotic degradation processes and the response
of bacterial communities to di¡erent bioaugmentation and
biostimulation treatments [7^10].
In prior research in our laboratory, methods for biore-
mediation of PCBs have been developed that use repeated
applications of PCB-degrading bacteria to contaminated
soil. Since PCBs are not readily utilized as carbon sources
for growth, their degradation is enhanced by addition of a
cosubstrate such as biphenyl which induces expression of

FEMSEC 1379 7-8-02

192 E. Luepromchai et al. / FEMS Microbiology Ecology 41 (2002) 191^197
the bph genes encoding for biphenyl degradation. This can
also be achieved using selected monoterpenes or salicylic
acid that serve as cosubstrates for cometabolism of PCBs,
but which are less toxic than biphenyl [11]. Another im-
provement in these methods has been the co-application of
a surfactant that mobilizes PCBs and serves as a growth
substrate for the soil inoculants and the indigenous bac-
teria [12]. Lastly, the most recent innovation has been the
use of earthworms to enhance the dispersal of PCB-de-
grading bacteria that are applied to the surface of the
contaminated soil. In prior research with earthworms
[13], the disappearance of PCBs appeared to be due to a
combination of e¡ects including enhanced dispersal of the
inoculum, improved soil aeration and increased soil car-
bon and nitrogen supply. Interestingly, in control treat-
ments, earthworms alone were found to promote PCB
degradation, which suggested that earthworms also in£u-
ence the indigenous biphenyl-degrading bacteria through
their activities in the soil. To further study this phenome-
non, the research reported here examines the interactions
between earthworms and the PCB-degrading bacteria dur-
ing bioremediation of a PCB-contaminated soil using bio-
augmentation with Ralstonia eutrophus, Rhodococcus sp.
ACS and their spent culture media that contains nutrients,
inducing substrates, and surfactant. Changes in PCB-de-
grading bacterial populations were monitored by polymer-
ase chain reaction (PCR) detection of the bphA and bphC
genes and enumeration of bacteria that can use biphenyl
as a growth substrate. The results suggest complex inter-
actions in which earthworms stimulated both the bioaug-
mented and indigenous bacteria to degrade PCBs, and
enhanced the overall e⁄cacy of removing PCBs below
the soil surface by mixing the soil and bacteria.
2. Materials and methods
2.1. Soil microcosm preparation
Twelve 30-cm long, 5.08-cm diameter brass columns
were prepared as soil microcosms. The columns were
closed on the bottom end with an open weave fabric
and were partially ¢lled with a layer of 140 g of uncon-
taminated soil that could be extracted and analyzed at the
end of the experiment to monitor any movement of PCBs
in water that leached from the upper layer of the contam-
inated soil. The pristine soil barrier also served to mini-
mize gas di¡usion from below into the soil column, there-
by more closely simulating oxygen concentrations found at
depth in the ¢eld. A layer of glass wool and two pieces of
copper mesh (1 mm mesh) were added to completely cover
the surface of the lower layer of the soil and to preclude
the movement of earthworms into the lower soil zone.
Each tube was then ¢lled with 0.6 kg of PCB-contaminat-
ed soil that was prepared from Dello loamy sand (typic
Psammaquent, sandy alluvium, 0.90% carbon, 0.08% ni-
FEMSEC 1379 7-8-02
trogen, pH 7.7) spiked with 100 ppm Aroclor 1242 as
previously described [13]. Once the columns were ¢lled,
they contained 20 cm of soil that was packed to within
5 cm of the top of the columns. The soil columns were
allowed to equilibrate for 4 weeks prior to starting the
treatments. Two days before applying the earthworms
and bioaugmented bacteria, each column was watered
with 70 ml of deionized water and 70 ml of minimal salts
(MS) medium [14] to establish a 12% soil water content.
The weights of the soil columns were noted to allow for
the determination of water loss over time. The soil mois-
ture contents were adjusted gravimetrically to their initial
levels every 3^4 days.
2.2. Bioremediation treatments
The experiment consisted of a 2U2 factorial design :
with and without earthworms and bioaugmentation. The
treatment employing bioaugmentation used two PCB-de-
grading bacteria, R. eutrophus H850 and Rhodococcus sp.
ACS, which were added along with the spent media from
their cultures. The control treatment received MS medium
only. In the soil columns with the earthworms, two adult
Pheretima hawayana earthworms (approximately 0.55 g
earthworm31 ) were placed in the soil, and were fed every
3^4 days with approximately 1 g of rolled oats. Similar
amounts of oats were added to the columns without earth-
worms. Three replicate columns were prepared for each of
the four treatments for a total of 12 columns. The soil
columns were maintained in a plant growth chamber at
20‡C with 70% relative humidity, a light intensity of 500
WE m32 s31 and a 16/8 h day/night cycle.
2.3. Inoculum preparation
Microbial cultures were grown in 250-ml Erlenmeyer
£asks containing 100 ml of culture medium, and were
grown on a rotary shaker at 250 rev min31 . The Rhodo-
coccus sp. ACS culture medium contained 100 mg l31
carvone, and 1000 mg l31 sorbitan trioleate dissolved in
MS medium ; whereas, the R. eutrophus H850 cultures
contained 500 mg l31 salicylic acid and 1000 mg g31 sor-
bitan trioleate. After approximately 20 h of growth, an
additional 50 Wl (500 mg g31 ) of sorbitan trioleate was
added to the culture £asks to maintain a su⁄cient concen-
tration of the surfactant for desorbing soil-bound PCBs
and to provide the inoculum with additional nutrients fol-
lowing soil inoculation [12]. After the surfactant was
added, the £asks were shaken for 15 min to allow the
sorbitan trioleate to dissolve into the medium. The soil
columns were inoculated with 15 ml of the bacteria sus-
pensions (108 colony-forming units (CFU) ml31 ), which
were applied to the soil surface as a soil drench. The in-
oculation of the soil was repeated biweekly to maintain a
cell density of 106 CFU g31 soil, alternating R. eutrophus
H850 and Rhodococcus sp. ACS cultures at each inocula-
 E. Luepromchai et al. / FEMS Microbiology Ecology 41 (2002) 191^197
tion. Non-inoculated columns that served as the control
treatment were amended with 15 ml of the MS medium.
When the amount of the amendment was not su⁄cient to
replenish the water loss between amendment applications,
additional deionized water was applied. Soil sampling was
conducted 3 days after the 18th amendment (total 9 weeks
treatment). The soils were sampled at three depths by re-
moving the soil from the columns and dividing them into
three fractions taken from 0^3 cm, 3^9 cm, and 9^20 cm
below the soil surface. Soil collected from each depth was
thoroughly mixed, and subsampled for use in bacterial
community and chemical analyses.
2.4. Bacterial community analyses
2.4.1. Biphenyl utilizer enumeration
The population densities of the bacteria that could uti-
lize biphenyl as a carbon source were enumerated using
agar plates. Enumeration was performed by plating 10-
fold serial dilutions of soil samples onto the surface of
agar plates containing MS Bacto agar with biphenyl crys-
tal vapor as the sole carbon source. The plates were incu-
bated at room temperature and numbers of the colonies
were counted after 2 weeks.
2.4.2. Soil DNA extraction
At the end of the experiment, the total DNA of the
actively growing bacteria was immunolabeled with bromo-
deoxyuridine (BrdU), which is a thymidine nucleotide ana-
log that can be used for DNA synthesis by bacteria. To
label the DNA, 4 g soil was amended with 40 Wl of 100
mM BrdU and allowed to incubate at room temperature
for 3 days. The total DNA was then extracted from 1 g
soil samples using a FastDNA SPIN Kit for soil (Bio 101,
Vista, CA, USA). The amount of DNA was estimated
visually after electrophoresis in a 1% agarose gel. The
BrdU-labeled DNA was isolated from the soil DNA using
the immunocapture method previously described by Bor-
neman [15].
2.4.3. PCR detection of bphA and bphC genes
PCR primers for the bphA and bphC genes were selected
for detection of the genes encoding for enzymes in the
PCB-degradation pathway. Biphenyl dioxygenase is the
¢rst enzyme in this pathway and is the product of the
bphA gene, whereas 2,3-dihydroxybiphenyl dioxygenase
is encoded by the bphC gene. Detection of the bphC
gene followed the method by Park et al. [16]. The primer
sequences for the bphC gene were 5P-CTGCACTGCAAC-
GAACGCCAC-3P (primer 1) and 5P-GACAC-
CATGTGGTGGTTGGT-3P (primer 2). The PCR primers
for the bphA gene were designed from a conserved region
that has been identi¢ed in several well-studied PCB de-
graders, including Pseudomonas pseudoalcaligenes KF707,
Comamonas testosteroni, Pseudomonas sp. KKS102, Pseu-
domonas sp. B4, Burkhoderia sp. LB400, Rhodococcus sp.
FEMSEC 1379 7-8-02
193
M5, Rhodococcus globerulus P6, and Rhodococcus sp.
RHA1. The sequences were retrieved from GenBank and
aligned with the software program, PileUp (Genetic Com-
puter Group, Madison, WI, USA). The primer sequences
were A13-gc (5P-CGCCCGCCGCGCGCGGCGGGCG-
GGGCGGGGGCACGGGGGGATGTTCGGCCAGCA-
CATGACG-3P (GC clamp underlined)) and A1r (5P-
GTCAAGAGCGGCAGCAGGAC-3P), which targeted a
230-bp fragment on the N-terminal sequence of the
bphA1 gene coding for the iron sulfur protein in biphenyl
dioxygenase. The GC clamp was attached to primer A13
for the denaturing gradient gel electrophoresis analysis.
Each PCR reaction contained 40 ng DNA, 5 pmol of
each primer, a Ready-To-Go PCR bead (Pharmacia Bio-
tech, Piscataway, NJ, USA), and sterilized distilled water
to a ¢nal volume of 25 Wl. The PCR ampli¢cation was
performed on a PTC-200 thermocycler (MJ Research,
Inc., Watertown, MA, USA) with a touchdown program
to increase the speci¢city of the target. The temperature
pro¢le was 5 min at 95‡C followed by ampli¢cation for 30
cycles with a 0.5‡C decrease in the annealing temperature
after each cycle. The starting cycle consisted of 1 min at
94‡C, 30 s at 65‡C, and 1 min at 72‡C. Twenty cycles of
the same program were conducted with a ¢xed annealing
temperature of 50‡C, and a ¢nal extension for 6 min at
72‡C. The identi¢cations of the correct PCR products
from bphA primers were con¢rmed by DGGE analysis
with denaturing gradients ranging from 20 to 60%. The
samples with a PCR product at the same position as the
inoculated bacteria were considered positive. Di¡erences
in the amount of the PCR products between the samples
were estimated visually and scored from 0 to 4.
2.5. Chemical analyses
PCBs were extracted from 20-g soil samples that were
taken at each soil depth and from the lower uncontami-
nated soil layer that was used to collect water leachate.
The samples were placed in 40-ml vials and extracted by
the addition of 10 ml hexane:acetone (9:1 v/v) and 5 ml of
3% sodium dodecyl sulfate solution. The vials were sealed
with Te£on tape and shaken on a horizontal shaker for 24
h. The soils were then centrifuged (10 000Ug for 20 min),
after which they were frozen at 320‡C to separate the
water from the organic supernatant. The organic super-
natant was removed and analyzed by gas chromatography
using £ame ionization detection (GC-FID) (Hewlett-Pack-
ard Co., Palo Alto, CA, USA) as previously described
[13].
2.6. Statistical analyses
Statistical analyses of the di¡erences in PCB concentra-
tion were carried out by analysis of variance (ANOVA)
with repeated measure design using SAS 8.0 for Windows
(SAS Institute, Inc., Cary, NC, USA). Least signi¢cance
194 E. Luepromchai et al. / FEMS Microbiology Ecology 41 (2002) 191^197

Table 1
Population sizes of biphenyl-degrading bacteria in treated soil (U107
CFU g31 )a
Depths Soil columns
No earthworm Earthworm
b c
BIO MS BIO MS
0^3 cm 24.3aA 4.0aB 140aC 31aA
3^9 cm 6.7bA 1.3bA 7.5bA 7.9bA
9^20 cm 0.6cAC 0.2cA 5.2bB 0.7cC
a
Comparisons within a soil column between depths are signi¢cantly dif-
ferent (LSD, P 6 0.05) if marked with di¡erent lowercase letters. Com-
parisons within a depth between soil columns are signi¢cantly di¡erent
(LSD, P 6 0.05) if marked with di¡erent uppercase letters.
b
BIO: Bioaugmentation treatment.
c
MS: MS medium treatment.

di¡erence (LSD) tests were performed to determine signi¢-

cant di¡erences between the means (P = 0.05) of the treat-
ments. The cell densities of the biphenyl-utilizing bacteria,
the colony count data were log-transformed prior to sta-
tistical analysis.
3. Results
3.1. Biphenyl-utilizing bacterial population densities
The cell densities of culturable, biphenyl-utilizing bacte-
Fig. 1. PCR products ampli¢ed from bph gene speci¢c primers on aga-
rose gels. Template DNA used in the ampli¢cation was as follows : lane
1, Burkhoderia sp. LB400; lane 2, Rhodococcus globerulus P6; lane 3,
Ralstonia eutrophus H850; lane 4, Rhodococcus sp. ACS; lane 5, non-
sterile soil amended with Ralstonia eutrophus H850; lane 6, non-sterile
unamended soil ; and lane 7, no template. Bands corresponding to bphA
and bphC genes were 310 bp and 182 pb long, respectively. The lower
bands correspond to the primers used in the ampli¢cation.

ria and their distribution in soil columns were enumerated

using agar plate assays that do not di¡erentiate between
bioaugmented bacteria and the indigenous biphenyl-de-
grading bacteria. Total cell densities of biphenyl-degrading
bacteria were a¡ected by the presence of earthworms and
bioaugmentation, but only in the top 3 cm of the soil.
Bioaugmentation alone increased the number of biphenyl
utilizers 5-fold in the 0^3 cm depth, such that biphenyl
utilizers had an average population size of 4U107 CFU
g31 in non-bioaugmented columns and 24U107 CFU g31
in the bioaugmented columns (Table 1). The addition of
earthworms signi¢cantly increased the number of biphenyl
utilizers in the top 3 cm by 6- and 8-fold in bioaugmented
and non-bioaugmented columns, respectively. In soil be-
low this zone at the 3^9 cm depth, the population size of
biphenyl-degrading bacteria ranged from 1.3 to 7.9U107
CFU g31 soil, and there were no signi¢cant di¡erences
between treatments. At the 9^20 cm depth, the population
densities decreased another order of magnitude in all of
the treatments with the exception of the bioaugmented
columns with earthworms, which maintained 5U107 bi-
phenyl degraders per g of soil. The densities of biphenyl
utilizers at the 9^20 cm depth in the non-bioaugmented
soils and treatments without earthworms ranged from
0.2U107 to 0.7U107 CFU g31 soil.
3.2. Detection of the bphA and bphC genes in soil
The upper pathway of aerobic PCB degradation in-
volves four enzymes that are encoded by the bphABCD
genes, which transform individual PCB congeners to their

Table 2
PCR detection of bphA and bphC gene in treated soila
Treatment Mineral salts medium Bioaugmentation
0^3 cm 3^9 cm 9^20 cm 0^3 cm 3^9 cm 9^20 cm
Earthworm
bphA gene 3 3 3 + +++ ++
bphC gene +++ +++ ++++ + ++ +++
No earthworm
bphA gene 3 3 3 ++ +++ 3
bphC gene 3 3 3 ++ ++ +++
a
3, no ampli¢cation; + to ++++, ampli¢cation, with the strength of the ampli¢cation signal increasing from + to ++++.

FEMSEC 1379 7-8-02

 E. Luepromchai et al. / FEMS Microbiology Ecology 41 (2002) 191^197 195

respective chlorobenzoic acids [17]. We selected the bphA
and bphC genes to monitor the inoculated and indigenous
PCB-degrading bacteria because these genes exhibit high
homology among many known PCB degraders [18]. PCR
fragments corresponding to the bphA and bphC genes were
ampli¢ed from genomic DNA of known PCB-degrading
bacteria (i.e. Burkhoderia sp. LB400, R. globerulus P6, R.
eutrophus H850, and Rhodococcus sp. ACS) and from soil
DNA that was inoculated with R. eutrophus H850 (Fig. 1).
The bphA gene PCR products were further analyzed by
DGGE. All DGGE bands from the PCB degraders had
the same relative position at about 50% denaturant (data
not shown).
At the end of the experiment, the bphA gene was de-
tected only in soil DNA samples from bioaugmented soil
(Table 2). Within the bioaugmented treatments, distribu-
tion of the bphA gene at di¡erent soil depths was in£u-
enced by the presence of earthworms. The bphA gene
product was produced by PCR ampli¢cation of DNA ex-
tracted from soil at all depths in the bioaugmented treat-
ments with earthworms, and in the bioaugmented soil col-
umns at 0^3 and 3^9 cm depths, but not at the 9^20 cm
depth in bioaugmented columns without earthworms. The
bphC gene product distribution did not correspond with
the distribution of the bphA gene, and was detected at high
levels throughout the soil pro¢le in non-bioaugmented
soils with earthworms. Lesser quantities of the bphC
gene were detected in the bioaugmented soils, irrespective
of the presence of earthworms. As with the bphA gene, the
bphC gene was not detected in control columns without
earthworms.
3.3. PCB recovery
ment employing bioaugmentation with PCB-degrading
bacteria and earthworms resulted in signi¢cant disappear-
ance of PCBs at all depths (23%, 52%, and 61% recovery
at 0^3 cm, 3^9 cm, and 9^20 cm, respectively). In contrast,
PCB disappearance in the bioaugmented soil columns
without earthworms was signi¢cantly less as compared
to bioaugmented columns with earthworms. In the treat-
ment employing bioaugmentation with PCB-degrading
bacteria without earthworms, approximately 50% of the
PCBs were recovered in the top 3-cm layer of soils, which
increased to 79 and 71% PCB recovery in the 3^9 and 9^20
cm depths, respectively. These values at the soil depths
below 3 cm were not signi¢cantly di¡erent from those
obtained in the non-bioaugmented control treatment with-
out earthworms.
Earthworms alone without bioaugmentation promoted
PCB disappearance in the upper 3 cm, although not to the
same extent as the combined treatment using both bioaug-
mentation and earthworms (Table 3). PCB recovery in
non-bioaugmented columns with earthworms was 53% at
the 0^3 cm depth, but increased to 76 and 72% at the 3^9,
and 9^20 cm depths, respectively. In the control treatment
without bioaugmentation or earthworms, the di¡erences in
PCB recovery between depths were not signi¢cant. PCB
recoveries in these columns ranged from 78 to 80%. A
consistently small amount of PCBs were recovered from
the pristine, leachate collection soil, and averaged 12.5
ppm across all treatments. Calculations of the mass of
PCBs that leached from the upper soil pro¢le into the
uncontaminated soil showed that there was an approxi-
mate 3% loss of the PCBs from the soil columns due to
leaching.

Di¡erences in PCB recovery were used to determine 4. Discussion

PCB losses due to biodegradation and any other loss
mechanisms that may be associated with earthworms This research examined the interactions of earthworms
and bioaugmentation (Table 3). After 9 weeks, the treat- with indigenous and bioaugmented PCB-degrading bacte-
ria. Through the inevitable modi¢cation of the soil envi-
ronment via burrowing and comminution, earthworms ap-
Table 3 peared to increase inoculum dispersal and the population
Percent PCB recovery from soil after bioaugmentation in the presence
size of indigenous PCB-degrading microorganisms, which
or absence of earthwormsa
may have contributed to the observed increase in PCB
Depths Soil columns biodegradation. Earthworm-amended soils that were bio-
No earthworm Earthworm augmented with PCB-degrading bacteria contained higher
BIO b
MS c
BIO MS levels of biphenyl-degrading bacteria in the top 3 cm of the
0^3 cm 51aA 78aB 23aC 53aA soil, and contained increased levels of the bphA and bphC
3^9 cm 79bA 82aA 52bB 76bA genes that could be detected in the top 9 cm of the soil
9^20 cm 71abA 80aA 61cA 72bA using PCR-based methods. The combined treatment with
Leachate soild (ppm PCB) 14A 12A 12A 12A earthworms and bioaugmentation was by far the best
a
Comparisons within the soil columns between depths are signi¢cantly treatment and resulted in signi¢cantly enhanced disappear-
di¡erent (LSD, P 6 0.05) if marked with di¡erent lowercase letters. ance of PCBs, with approximately 50% of the PCBs being
Comparisons within a depth between soil columns are signi¢cantly dif-
removed in the top 9 cm of the soil columns after 9 weeks
ferent (LSD, P 6 0.05) if marked with di¡erent uppercase letters.
b
BIO: Bioaugmentation treatment. of treatment. In comparison, bioaugmentation with PCB-
c
MS: MS medium treatment. degrading bacteria in the absence of earthworms, or the
d
PCB extracted from uncontaminated soil layer at 20^25 cm depth. addition of earthworms without bacterial bioaugmentation

FEMSEC 1379 7-8-02

196 E. Luepromchai et al. / FEMS Microbiology Ecology 41 (2002) 191^197
was e¡ective for increasing the removal of PCBs only in
the top 3 cm of the soil columns.
Aerobic mineralization of PCBs requires the activity of
two di¡erent populations of bacterial species, the ¢rst con-
sisting of bacterial species that contain the bph genes en-
coding the upper pathway for degradation of chlorinated
biphenyls to chlorobenzoates. Microorganisms containing
the lower pathway genes may then degrade the resulting
chlorobenzoates, which can be used as carbon substrates
for growth. Alternatively, the chlorobenzoates may be for-
tuitously metabolized during growth-linked metabolism on
other carbon substrates [1,19]. Although most soils con-
tain high population densities of bacteria that carry genes
for the upper pathway encoding the enzymes for biphenyl
degradation, these bacteria are not usually induced to de-
grade PCBs unless they are grown on a cosubstrate that
induces the expression of the bph genes. In addition to
biphenyl, several naturally occurring substrates including
monoterpenes and salicylic acid have been identi¢ed as
potential cometabolites, putting in question whether these
are the ‘natural’ substrates for PCB-degrading enzymes.
The ¢nding that earthworms alone also promote PCB re-
moval is intriguing since no cometabolites were added in
the control treatment. This suggests that earthworm casts
(i.e. egested soil from the earthworm) may contain certain
compounds that serve as cometabolites for induction of
the bph genes. Addition of earthworms alone to the soil
columns was associated with an 8-fold increase in the
population size of culturable, indigenous biphenyl-degrad-
ing bacteria. Compounds originating from earthworm ac-
tivity, such as urea, have been speculated as potential
stimulants of PCB degradation [13], however, they are
unlikely to be related to bph induction.
Biphenyl-degrading bacteria vary greatly in their ability
to degrade di¡erent PCB congeners. Previously, increasing
the population size of biphenyl-utilizing bacteria, either
through bioaugmentation with selected PCB-degrading
bacteria, or by biostimulation with cosubstrates that in-
duce the bph genes has been shown to enhance PCB deg-
radation [11,12,20^24]. In this research, earthworm activ-
ity was associated with an increase in the level of the bphC
gene throughout the entire 20-cm soil pro¢le of columns
without bacterial bioaugmentation. Curiously, this was
not accompanied by a concomitant increase in the level
of detection in the bphA gene, which suggests that the
earthworms a¡ected bacteria other than those carrying
the bph operon that normally contains both genes. The
bphA gene is involved in dioxygenation of the biphenyl
ring at the 2,3 position with formation of dihydrodiol,
whereas the bphC gene codes for a dioxygenase that is
involved in ring ¢ssion of phenylcatechol [25]. The inde-
pendent detection of bphC, but not the bphA gene suggests
that indigenous bacteria may exist that have an alternate
gene coding for the ¢rst step. Alternatively, they may
carry a gene similar to bphC that is contained in bacteria
that are enriched by earthworm activity, but which may or
FEMSEC 1379 7-8-02
may not be involved in PCB degradation. For example,
the increased detection of bph genes may re£ect the pres-
ence of homologous dioxygenase genes that are responsi-
ble for degradation of naphthalene, toluene, and benzoate
[17]. Analyses of the ways in which earthworms a¡ect
microbial consortia that are involved in PCB degradation,
and the distribution of di¡erent genes encoding for PCB-
degrading enzymes will require better knowledge of the
diversity of these enzymes and the development of appro-
priate probes for monitoring their quantity and expression
during bioremediation of contaminated soils.
Acknowledgements
We thank James Borneman for providing valuable ad-
vice in nucleic acid extraction, Sam Alvey for his assis-
tance on data analysis, and Katechan Jampachaisri for
help with the statistical analysis. Funding for Ekawan
Luepromchai was provided by Chulalongkorn University,
Thailand.
References
[1] Robinson, G.K. and Lenn, M.J. (1994) The bioremediation of poly-
chlorinated biphenyls (PCBs): problems and perspectives. Biotechnol.
Genet. Eng. Rev. 12, 139^188.
[2] Head, I.M. (1998) Bioremediation: towards a credible technology.
Microbiology 144, 599^608.
[3] Liu, S. and Su£ita, J.M. (1993) Ecology and evolution of microbial
populations for bioremediation. Trends Biotechnol. 11, 344^352.
[4] Pritchard, P.H., Mueller, J.G., Lantz, S.E. and Santavy, D.L. (1995)
The potential importance of biodiversity in environmental biotech-
nology applications; bioremediation of PAH-contaminated soils and
sediments. In: Microbial Diversity and Ecosystem Function (Allsopp,
D., Hawksworth, D.L. and Colwell, R.R., Eds.), pp. 161^182. CAB
International, Wallingford.
[5] Forsyth, J.V., Tsao, Y.M. and Bleam, R.D. (1995) Bioremediation :
when is augmentation needed? In: B.C. Alleman (Hinchee, R.E. and
Frederickson, J., Eds.), pp. 1^14. Batelle Press, Columbus, OH.
[6] Shannon, M.J.R. and Unterman, R. (1993) Evaluating bioremedia-
tion: distinguishing fact from ¢ction. Annu. Rev. Microbiol. 47, 715^
738.
[7] Heinonsalo, J., Jorgensen, K.S., Haahtela, K. and Sen, R. (2000)
E¡ects of Pinus sylvestris root growth and mycorrhizosphere develop-
ment on bacterial carbon source utilization and hydrocarbon oxida-
tion in forest and petroleum-contaminated soils. Can. J. Microbiol.
46, 451^464.
[8] Holoman, T.R.P., Elberson, M.A., Cutter, L.A., May, H.D. and
Sowers, K.R. (1998) Characterization of a de¢ned 2,3,5,6-tetrachlo-
robiphenyl-ortho-dechlorinating microbial community by compara-
tive sequence analysis of genes coding for 16S rRNA. Appl. Environ.
Microbiol. 64, 3359^3367.
[9] MacNaughton, S.J., Stephen, J.R., Venosa, A.D., Davis, G.A.,
Chang, Y.-J. and White, D.C. (1999) Microbial population changes
during bioremediation of an experimental oil spill. Appl. Environ.
Microbiol. 65, 3566^3574.
[10] Margesin, R., Zimmerbauer, A. and Schinner, F. (2000) Monitoring
of bioremediation by soil biological activities. Chemosphere 40, 339^
346.
[11] Gilbert, E.S. and Crowley, D.E. (1998) Repeated application of car-
 E. Luepromchai et al. / FEMS Microbiology Ecology 41 (2002) 191^197
vone-induced bacteria to enhance biodegradation of polychlorinated
biphenyls in soil. Appl. Microbiol. Biotechnol. 50, 489^494.
[12] Singer, A.C., Gilbert, E.S., Luepromchai, E. and Crowley, D.E.
(2000) Bioremediation of polychlorinated biphenyl-contaminated
soil using carvone and surfactant-grown bacteria. Appl. Microbiol.
Biotechnol. 54, 838^843.
[13] Singer, A.C., William, J., Luepromchai, E., Yahng, C.-S. and Crow-
ley, D.E. (2001) Contribution of earthworms to PCB bioremediation.
Soil Biol. Biochem. 33, 765^776.
[14] Focht, D.D. (1994) Microbiological procedures for biodegradation
research. In: Methods of Soil Analysis. Part 2. Microbiological and
Biochemical Properties (Weaver, R.W., Angle, J.S. and Bottomley,
P.S., Eds.). Soil Science Society of America, Madison, WI.
[15] Borneman, J. (1999) Culture-independent identi¢cation of microor-
ganisms that respond to speci¢ed stimuli. Appl. Environ. Microbiol.
65, 3398^3400.
[16] Park, Y.-I., So, J.-S. and Koh, S.-C. (1999) Induction by carvone of
the polychlorinated biphenyl (PCB)-degradative pathway in Alcali-
genes eutrophus H850 and its molecular monitoring. J. Microbiol.
Biotechnol. 9, 804^810.
[17] Furukawa, K. (1994) Molecular genetics and evolutionary relation-
ship of PCB-degrading bacteria. Biodegradation 5, 289^300.
[18] Kumamaru, T., Suenaga, H., Mitsuoka, M., Watanabe, T. and Fur-
ukawa, K. (1998) Enhanced degradation of polychlorinated biphenyls
by directed evolution of biphenyl dioxygenase. Nat. Biotechnol. 16,
663^666.
FEMSEC 1379 7-8-02
197
[19] Focht, D.D., 1993. Microbial degradation of chlorinated biphenyls.
In: Soil Biochemistry (Bollag, J.M. and Stozky, G., Eds.), vol. 8.
Marcel Dekker, New York.
[20] Barriault, D. and Sylvestre, M. (1993) Factors a¡ecting PCB degra-
dation by an implanted bacterial strain in soil microcosms. Can. J.
Microbiol. 39, 594^602.
[21] Evans, B.S., Dudley, C.A. and Klasson, K.T. (1996) Sequential an-
aerobic-aerobic biodegradation of PCBs in soil slurry microcosms.
Appl. Biochem. Biotechnol. 57/58, 885^894.
[22] Focht, D.D. and Brunner, W. (1985) Kinetics of biphenyl and poly-
chlorinated biphenyl metabolism in soil. Appl. Environ. Microbiol.
50, 1058^1063.
[23] Harkness, M.R., McDermott, J.B., Abramowicz, D.A., Salvo, J.J.,
Flanagan, W.P., Stephens, M.L., Mondello, F.J., May, R.J., Lobos,
J.H., Carroll, K.M., Brennan, M.J., Bracco, A.A., Fish, K.M., Warn-
er, G.L., Wilson, P.R., Dietrich, D.K., Lin, D.T., Morgan, D.B. and
Gately, W.L. (1993) In situ stimulation of aerobic PCB biodegrada-
tion in Hudson River sediments. Science 259, 503^507.
[24] Viney, I. and Bewley, R.J.F. (1990) Preliminary studies on the devel-
opment of a microbiological treatment for polychlorinated biphenyls.
Arch. Environ. Contam. Toxicol. 19, 789^796.
[25] Focht, D.D. and Reineke, W. (2001) Biotransformation of poly-
chlorinated biphenyls. In: Manual of Environmental Microbiology
(Hurst, C.J., Crawford, R.L., Knudsen, G.R., McInerney, M.J. and
Stetzenback, L.D., Eds.). ASM Press, Washington, DC.