Previous Article | Next Article 
Applied and Environmental Microbiology, January 2001, p. 239-244, Vol. 67, No. 1
0099-2240/01/$04.00+0 DOI: 10.1128/AEM.67.1.239-244.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.
Detection of Oxytetracycline Production by Streptomyces
rimosus in Soil Microcosms by Combining Whole-Cell Biosensors
and Flow Cytometry
Lars Hestbjerg
Hansen,1
Belinda
Ferrari,2
Anders Hay
Sørensen,1
Duncan
Veal,2 and
Søren Johannes
Sørensen1,*
Department of General Microbiology,
University of Copenhagen, DK-1307 Copenhagen K,
Denmark,1 and Department of
Biological Sciences, Macquarie University, Sydney, New South Wales
2109, Sydney, Australia2
Received 18 May 2000/Accepted 1 November 2000
 |
ABSTRACT |
Combining the high specificity of bacterial biosensors and the
resolution power of fluorescence-activated cell sorting (FACS) provided
qualitative detection of oxytetracycline production by Streptomyces rimosus in soil microcosms. A plasmid
containing a transcriptional fusion between the
tetR-regulated Ptet promoter from
Tn10 and a FACS-optimized gfp gene was
constructed. When harbored by Escherichia coli, this
plasmid produces large amounts of green fluorescent protein (GFP) in
the presence of tetracycline. This tetracycline biosensor was used to
detect the production of oxytetracycline by S. rimosus
introduced into sterile soil. The tetracycline-induced GFP-producing
biosensors were detected by FACS analysis, enabling the detection of
oxytetracycline encounters by single biosensor cells. This approach can
be used to study interactions between antibiotic producers and their
target organisms in soil.
| |
INTRODUCTION |
The use of bacterial biosensors,
i.e., bacteria giving an easily measurable response upon exposure to a
specific compound or environmental condition, is a promising new
approach in environmental biology. Their use has, however, until now
been limited to measurements in more homogeneous samples such as bulk
water (14, 17, 18) and soil extracts (10).
Soil is a complex matrix of microhabitats conferring highly variable
growth conditions for the microbiota (15). The ability to
gain an understanding of soil microbial ecology and microbial processes
has been severely hampered by the inability to characterize these
microhabitats at a scale or resolution relevant to microbial cells. Due
to their high specificity, sensitivity, and appropriate scale,
whole-cell biosensors offer an approach that could deal with these
issues of spatial resolution.
A major advantage of using biosensor bacteria to detect or quantify
specific compounds in natural samples is that only the fraction
available to the bacteria, the bioavailable fraction, is detected.
Hence, the measurements made are relevant to the effects of compounds
on the microbial community.
An interesting application of bacterial biosensors is the detection of
bioavailable antibiotics in different environments (7).
Detection of antibiotics resulting from anthropogenic or natural
production is crucial for our understanding of the evolution of
antibiotic resistance.
Whether antibiotics are produced in soil by indigenous soil organisms
has been a scientific dispute for several decades (5, 19).
Most antibiotics are excreted as secondary metabolites when the
producers are grown in rich media. It is not evident that conditions in
natural soil would allow the type of growth required for producing and
excreting the antibiotic. The growth of potential antibiotic producers
such as streptomycetes in soil is thought to be localized in discrete
areas, rather than evenly distributed throughout the soil
(19). Therefore, the production and presence of
antibiotics are likely to be limited to a few microhabitats where
conditions are favorable. This makes the detection of indigenous
antibiotic production by conventional methods difficult.
Streptomyces rimosus is a known industrial producer of
oxytetracycline and was originally isolated from soil (4).
Indeed, most known microbial producers of the different tetracyclines are bacteria native to soil. Actinomycetes are usually present in large
numbers in soil, and they constitute about 10% of the culturable
microbial population, exceeding 1 million CFU/g of soil
(5). Furthermore, large numbers of tetracycline resistance determinants are often found in soil samples (2, 13, 16). This has led to speculations that the tetracycline resistance genes are
present in soil because tetracyclines are produced there. However, the
production of tetracyclines in soil has previously never been shown by
direct detection, due to the lack of detection methods with the
necessary specificity and resolution power.
Most studies, aiming to examine antibiotic production in soil, have
employed extraction of the antibiotic from soil prior to analysis
(19). However, this method does not take the spatial distribution as well as the bioavailability of the compounds into account. If antibiotics are produced only in small amounts in a few
local microhabitas, this amount will be highly diluted during extraction. This dilution could completely mask the presence of the
compound, resulting in false-negative results.
We present here the construction of a biosensor bacterium which is
induced to express green fluorescent protein (GFP) upon exposure to
tetracycline. These experiments used soil microcosms where the
oxytetracycline producer S. rimosus was inoculated with the
biosensor. The aim of this research is to use biosensors that are
distributed in the soil at high density to detect highly localized production of oxytetracycline. To detect the potentially small subpopulation of biosensor cells expressing GFP, the bacteria were
extracted from the soil and analyzed using flow cytometry. Flow
cytometry enables the detection of low numbers of fluorescing cells
against a high background of nonfluorescent cells. To confirm the
flow cytometry results, induced biosensor bacteria were isolated using
fluorescence-activated cell sorting (FACS) and examined by
epifluorescence microscopy. In this study, we combine the high specificity of bacterial biosensors and the resolution power of FACS
analysis. We demonstrate this method's potential for addressing environmental problems that were hitherto difficult or impossible to analyze.
| |
MATERIALS AND METHODS |
Strains and plasmids.
The bacterial strains and plasmids
used in this study are shown in Table 1.
Media and culture conditions.
Both Escherichia
coli strains used were grown in a modified Luria-Bertani broth
(11) containing only 4 g of NaCl per liter (LB4). LB4
with 15 g of agar per liter was used for plate counts. S. rimosus was continuously maintained on Trypticase Soy broth (TSB)
(DSMZ medium 545) plates. S. rimosus spores were
pregerminated in pregermination buffer (8). Antibiotics
used were ampicillin and tetracycline, added to culture media to final
concentrations of 100 and 10 µg/ml, respectively. Oxytetracycline at
various concentrations was added to media when induction of
Ptet was examined. All recombinant DNA
manipulations were carried out by standard methods (11),
except where otherwise stated. All enzymes used were purchased from
Boehringer, Mannheim, Germany, and used according to the
manufacturer's guidelines. Antibiotics were purchased from Sigma.
Cloning of tetracycline biosensors.
Plasmids pTGFP1 and
pTGFP2, which both contain transcriptional fusions between the
tetracycline promoter Ptet and a FACS-optimized
gfp gene from pJBA27 (9), were constructed using PCR. The tetracycline repressor tetR (a negative
regulator of Ptet [1]) and
Ptet were amplified from pUTtetgfp (7) using three different primers. Primer set 1 (used to
construct pTGFP1) consisted of the primers
5'-AAAAGAATTCGCTGCTTTTAAGACCCAC-3' and
5'-CTATGCATGCCACTTTTCTCTATCACTG-3', giving a 732-bp fragment containing tetR and Ptet, as well as
an EcoRI site upstream and an SphI site
downstream of this regulatory region. Following digestion with
EcoRI and SphI, the PCR product was directionally ligated to EcoRI- and SphI-digested pJBA27 using
T4 DNA ligase. The ligation mix was then transformed into competent
DH5
cells (11), and transformants were screened for
tetracycline-inducible GFP production in an epifluorescence microscope
(model Axioskop 2; Carl Zeiss, Sydney, Australia). We thereby replaced
the already existing promoter region in pJBA27 (PA1-04/03)
with the tetR-regulated promoter
Ptet. In this plasmid, the ATG codon that would normally (in Tn10) be the start codon in the tetA
gene (encoding a tetracycline resistance efflux pump) was now the start
codon of the FACS-optimized gfp gene.
Primer set 2 (used to construct pTGFP2) consisted of the same forward
primer as in primer set 1 and 5'-CCTTTACGCATGCTGAGTCTCCAG-3'.
Amplification from pUT
tetgfp with these primers gave a
928-bp
fragment that included not only
tetR and
P
tet but also
the highly efficient
atpE transcriptional initiation region from
plasmid
pUT
tetgfp. This PCR fragment was also digested with
EcoRI
and
SphI, ligated into pJBA27, transformed
into DH5
, and screened
for tetracycline-inducible GFP production.
Plasmid pTGFP2 is shown
in Fig.
1. After
verification by electrophoresis, the plasmids
were transformed into
E. coli strain MC4100. The recombinant strains
were then
examined for their GFP levels in response to oxytetracycline.
Aliquots
of 50 µl from exponentially growing cultures of
E. coli MC4100 harboring either pTGFP1 or pTGFP2 were inoculated into
test
tubes containing 5 ml of LB4 with different concentrations
of
oxytetracycline. Three parallel samples were incubated for
each
concentration. Cultures were incubated at 30°C with shaking
for
16 h. After incubation, 0.5 ml of culture from each concentration
was centrifuged (6,000 ×
g, 2 min), washed once and
resuspended
in 3 ml of 0.9% NaCl, and transferred to cuvettes.
Fluorescence
(from GFP) was measured in a Perkin-Elmer LS50
luminescence spectrometer
(Perkin-Elmer, Beaconsfield, Buckinghamshire,
England). The excitation
wavelength used was 488 nm, and the emission
wavelength was 511
nm, both with a slit width of 2.5 nm. Optical
densities of cultures
were measured at 600 nm (OD
600) in an
Ultrospec 2000 Spectrophotometer
(Pharmacia Biotech, Cambridge, United
Kingdom). Relative fluorescence
units per OD
600 was then
plotted against the oxytetracycline concentration
(Fig.
2). The performance of
E. coli
strains containing pUT
tetgfp,
pTGFP1, and pTGFP2 was tested
for applicability in flow cytometry
after induction of the strains with
50 ng of oxytetracycline per
ml (see below).
View larger version (8K):
[in this window]
[in a new window]
|
FIG. 1.
Composition of pTGFP2. A928-bp
EcoRI-SphI fragment encompassing tetR,
Ptet, and the E. coli atpE
translation initiation region (in black), all obtained by PCR, was
inserted into pJBA27. The bla gene encodes  -lactamase
conferring resistance to ampicillin. ori colE1, origin of
replication originating from pUC18-NotI; lacZ',
partially deleted lacZ gene; gfpmut3,
FACS-optimized gfp gene encoding GFP. The two T's are
transcriptional terminator sequences from pJBA27.
Ptet is the tet promoter and
tetR encodes the tet repressor protein, both
originating from Tn10.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
FIG. 2.
GFP in cultures of MC4100 harboring pTGFP1 or
pTGFP2. Strains were grown overnight in LB4 containing increasing
concentrations of oxytetracycline. Diamonds represent MC4100/pTGFP1,
and squares represent MC4100/pTGFP2. Vertical bars show the standard
deviations (n = 3).
|
|
Microcosm experiment.
Samples of 2.8 g of dried soil
from Sturt National Park, far northwestern New South Wales, Australia
(a gift from Andrew J. Holmes, Macquarie University, Sydney,
Australia), including two barley leaves (1.5 by 0.5 cm), were
distributed into 15-ml polypropylene conical Falcon tubes (Becton
Dickinson, North Ryde, New South Wales, Australia). This soil was
chosen due to its high content of 16S ribosomal DNA sequences related
to actinomycetes (data not shown). The barley leaves were intended for
microscopic examination following the incubation. The microcosms were
then sterilized by exposing the tubes to a dose of 8,000 Gy from a
gamma source (Macquarie University cobalt-60 gamma source). From an
overnight culture of biosensor E. coli MC4100/pTGFP2, 0.5 ml
was reinoculated into 50 ml of LB4 medium. Cells were grown until
OD600 reached 0.8. Thereafter, the biosensor cells were
washed twice in phosphate-buffered saline (PBS) and resuspended in 20 ml of PBS. An aliquot (350 µl) of this suspension was then added to
each tube (1.7 × 107 CFU as tested on LB4 plates
containing ampicillin). Spores of S. rimosus were harvested
from 6-day-old TSB plates, by adding 10 ml of sterile double-distilled
H2O and scraping the spore layer of the colonies with a
sterile inoculation loop. The spore suspension was collected and
filtered twice through a syringe filled with sterile, nonabsorbent
cotton wool (8). After filtration, the spores were
centrifuged and resuspended in 10 ml of sterile double-distilled H2O. Following incubation of the spores at 50°C for 10 min, an equal volume of 2× pregermination buffer (8) was
added and the spores were incubated at 37°C with shaking for 3 h. The
spores were then washed twice in PBS and resuspended in 10 ml of PBS (to a final concentration of 3.2 × 105 CFU/550 µl
as tested on TSB plates containing tetracycline). Nine soil microcosms
were inoculated with 550 µl of undiluted spore suspension (series A,
8.6 × 104 spores/g of wet soil), nine soil microcosms
were inoculated with 550 µl of a 10
2 dilution of spores
(series B, 8.6 × 102 spores/g of wet soil), nine soil
microcosms were inoculated with 550 µl of a 104 dilution
of spores (series C, 8.6 spores/g of wet soil), and nine control
microcosms were supplemented with 550 µl of PBS (series D, no
spores). Thus, all the microcosms contained 900 µl of liquid. In
order to verify that the S. rimosus spore inoculum did not contain any oxytetracycline, exponentially growing cells of E. coli MC4100/pTGFP2 were inoculated with a dilution series of
sonicated (5 min in a 250/450 Sonifier [Branson Ultrasonics Corp.,
Danbury, Conn.]) and sterile-filtered S. rimosus inoculum
and shaken at 37°C for 16 h. A tube containing E. coli MC4100/pTGFP2 plus PBS and a tube containing MC4100/pTGFP2
plus 50 ng of oxytetracycline per ml were included. The cultures were
washed twice in PBS, and fluorescence was determined using an LS 50B
luminescence spectrometer (Perkin-Elmer). A scan of light emission from
495 to 525 nm was recorded when samples were excited at 488 nm. The
FACS-optimized GFP has an emission peak at 511 nm.
Microcosms were incubated at room temperature, and triplicate samples
were taken at given times. At each sampling, 10 ml of
PBS was added to
each microcosm sampled, the tubes were vortexed
for 1 min, and the soil
slurry was allowed to settle for 1 h to
avoid background fluorescence
from soil particles. Subsamples
(2 ml) of the supernatant were filtered
through a 38-µm-pore-size
stainless steel mesh (using Swinnex filter
holders [Millipore];
13-mm diameter). A dilution series was made in
PBS for plate counts
of both
S. rimosus (on TSB containing
tetracycline) and
E. coli MT4100/pTGFP2 (on LB4 containing
ampicillin). The undiluted filtrate
was analyzed using a Becton
Dickinson FACScalibur flow cytometer
(Becton Dickinson, North Ryde, New
South Wales, Australia) to
detect and enumerate biosensor bacteria
which had been exposed
to oxytetracycline (see
below).
Flow cytometry.
Flow cytometry analyses were performed using
a FACScalibur flow cytometer equipped with an argon ion laser (488 nm)
capable of GFP excitation. Voltages were set at 380 V for side scatter (SSC), 600 V for detector FL1, and 650 V for detectors FL2 and FL3
(fluorescence detectors). Sheath fluid consisted of undiluted Osmosol
(Lab Aids Pty Ltd., Narrabeen, New South Wales, Australia). Fluorebrite
beads (6 µm; Polysciences, Warrington, Pa.) were run to check
instrument performance before analysis of samples. This setup was used
to test pure cultures.
For analysis of soil samples, the instrument was set up for
environmental analysis (AusFlow protocol) by swapping FL1 and
FL2
detectors. Compensation was set at FL2
%FL1 = 99.9% and
FL2
%FL3 = 99.9% to reduce natural autofluorescence
found in
environmental samples. Samples were analysed in triplicate for
every condition, and the instrument was decontaminated by running
10%
(wt/vol) sodium hypochlorite solution followed by distilled
water
between each sample. Positive control samples consisted
of soil
slurries (extracted as described above) with biosensor
bacterium
E. coli MC4100/ pTGFP2, which had been induced with
50 ng of
oxytetracycline per ml prior to mixing with the soil.
These samples
were analyzed first, and the threshold was set,
using the green
fluorescence detector (FL2), to just above the
majority of the
background of fluorescent particles (Fig.
3).
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 3.
Flow cytometric analysis of E. coli
MC4100/pTGFP2 cells (induced with 50 ng of oxytetracycline per ml)
within soil. Region 1 defines where bacteria lie according to size
(left). The second dot plot (right) is gated on bacteria (region 1 [R1]). Region 2 (R2) defines the population of GFP-expressing
bacteria.
|
|
A polygonal gate (R1) was defined around the population of positive
control bacteria on a bivariate dot plot of log SSC versus
log forward
scatter (Fig.
3). A second dot plot of log SSC versus
log green
fluorescence was then analyzed by gating on R1, and
an ellipse region
(R2) was defined around the bacteria expressing
GFP. Negative samples
without fluorescent cells were analyzed
to determine the level of
background fluorescence occurring within
R2.
Sample filtrates from microcosms (100 µl) were added to 100 µl of
PBS in 5-ml Falcon tubes (BD Biosciences, San Jose, Calif.)
and were
vortexed prior to FACS analysis. Samples were analyzed
on medium flow
rate for 3 min (equivalent to 40 µl of filtrate
sample). All events
satisfying threshold requirements were collected
and saved into data
files. All data analyses were carried out
using Cellquest software (BD
Biosciences). The software WinMDI
obtained from Joseph Trotter (Salk
Institute for Biological Studies,
La Jolla, Calif.) was used for
graphic presentation of data. Verification
of the GFP-expressing
population (Fig.
3) (R2) was carried out
using FACS of GFP-induced
cultures. A FACScalibur flow cytometer
modified for environmental
analysis was used to select target
cells, allowing the confirmation of
these by epifluorescence microscopy
(Axioskop 2; Carl Zeiss.
Confirmation of GFP-expressing bacteria
was carried out using ×40 and
×100 (oil immersion) objectives.
Excitation of GFP was through a 100-W
mercury vapor arc lamp with
a filter block that excited GFP between 450 and 490 nm, allowing
visualization at 520
nm.
Statistics.
Cell numbers found on different days were
compared by the use of Student's t test. Probabilities of
less than 0.05 were considered significant.
| |
RESULTS AND DISCUSSION |
Biosensor construction.
Detection of tetracycline in soil by
FACS analysis was in this study achieved by the construction of a
bacterial biosensor which contained the GFP gene under the regulation
of the tetracycline-responsive promoter Ptet.
An existing tetracycline-induced GFP-producing biosensor construct
(
7) did not produce sufficient fluorescence for detection
in a FACS sorter (data not shown). The GFP produced from this
construct
has an optimal excitation wavelength of 395 nm compared
to the FACS,
which excites using a 488-nm laser beam. It was especially
inadequate
when soil samples were analyzed, since the Sturt National
Park soil
used in this experiment contained many weakly fluorescent
particles
giving forward scatter and SSC signals in the same region
as the
E. coli MC4100 cells (data not
shown).
Two new biosensor cassettes were therefore cloned using a
FACS-optimized
gfp gene (
gfpmut3) fused to
the
tet repressor gene
tetR and the
tet promoter P
tet. GFP produced from
this
gene has a 21-fold increase in fluorescence intensity when excited
at 488 nm compared with the wild-type GFP (
3). This makes
it
more suitable for FACS analysis. Two versions of this biosensor
construct were made. Plasmid pTGFP1 contains
tetR-P
tet, the "natural"
translation initiation region from the
tetA gene,
fused to
the FACS-optimized
gfp gene. Plasmid pTGFP2 (Fig.
1)
contained, in addition to the components of pTGFP1, the highly
efficient translation initiation region of the
E. coli atpE gene.
The fluorescence produced from pTGFP1 was much lower than the
fluorescence from pTGFP2 (Fig.
2). This was true both for basal-level
GFP expression (at 0 µg of oxytetracycline per ml) and at higher
concentrations.
E. coli MC4100 containing either plasmid
showed
an increase of fluorescence in response to increased
oxytetracycline
concentrations. Both cultures induced with 50 ng of
oxytetracycline
per ml were then tested in the FACS sorter. At this
oxytetracycline
concentration, clear induction was seen in
E. coli MC4100/pTGFP2
(Fig.
2). The two induced cultures were added
to the soil which
was to be used in the microcosm experiments,
extracted with PBS,
filtered, and run through the FACS sorter. Only the
induced culture
of
E. coli MC4100/pTGFP2 was readily
distinguishable from the
other particles in the soil. The result is
shown in Fig.
3.
Microcosms.
Four sets of sterile soil microcosms containing
high numbers of cells of the tetracycline biosensor bacterium E. coli MC4100/pTGFP2 were set up. To each set, a decreasing number
of S. rimosus spores were added. Following incubation,
microcosms were sacrificed and the content of induced biosensor
bacteria was determined by FACS analysis.
No fluorescent bacteria (in the R2 region [Fig.
3]) were detected
above the background level on day 0 (Fig.
4). On day 2,
however, there was a
fluorescent population in the series A samples
(highest spore inoculum)
of 1.24 × 10
4 R2 counts per ml of extract
corresponding to 28% of the total
biosensor population calculated from
the plate counts. The number
of fluorescent bacteria in the series A
samples remained at approximately
the same high level on day 5 (9.64 × 10
3 R2 counts/ml of extract). Due to the
overall growth of the biosensor
in these samples on day 5, the
fluorescent population had dropped
to constitute only 2.4% of the
total number of biosensors. Figure
5
shows the distribution of fluorescence counts in a typical series
A
sample from day 0 and day 2. Only counts in the R2 region are
summarized in Fig.
4. The increase in the numbers of fluorescent
biosensors is most likely caused by oxytetracycline production
in the
soil microcosms. Cell sorting, followed by microscopy,
confirmed that
the particles counted by FACS indeed were induced
biosensor cells.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 4.
Enumeration of induced biosensor bacteria (E. coli MC4100/pTGFP2). Soil microcosms were extracted with 10 ml of
PBS, sedimented for 1 h, and filtered through a 38-µm-pore-Size
mesh. Soil extracts were then analyzed on a FACScalibur flow cytometer,
and only bacteria lying in the R2 region (shown and defined in Fig. 3)
are counted as positive. Values and standard deviations in the negative
area are not shown. Series A, B, C, and D were initially inoculated
with 8.6 × 104, 8.6 × 102, 8.6, and
0 spores per g of wet soil, respectively.
|
|
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 5.
Flow cytometric analysis at day 0 (left) and day 2 (right) of soil extract from microcosms inoculated with a high density
of S. rimosus spores (series A).
|
|
In series B samples (samples with a 10
2-diluted
S. rimosus inoculum) from day 2 and day 5, the number of induced
biosensors
appeared to be higher than in both the series C and the
series
D samples. The difference was, however, not statistically
significant
due to large fluctuations within replicates in series B,
day 5
(see error bars in Fig.
4). In any case, it suggests that
tetracycline
production is occurring in this series as well. FACS
signals in
samples from series C and series D remained at the same
background
level throughout the whole
period.
No detectable concentrations of GFP were found in the controls when a
sonicated spore inoculum was examined as described above,
indicating
that no oxytetracycline was added to the microcosms
together with the
S. rimosus spore inoculum (data not
shown).
Background fluorescence stemming from soil particles averaged 192 R2
counts/ml of extract (data not shown). This background
level was
subtracted from the counts of each microcosm harvested.
Only readings
above this background level are shown in Fig.
4.
Selected barley leaves were inspected in the epifluorescence microscope
in order to produce photographic evidence of the interaction
between
Streptomyces hyphae and associated biosensor cells. However,
the concentrations of the organisms were apparently too low, as
no
fluorescing cells were
found.
Cell growth and inhibition in the microcosms.
Bacterial
numbers showed that neither growth nor decline of the biosensor
bacteria was taking place during incubation in sterile soil (Fig.
6). On day 5, however, the numbers of
biosensors had increased significantly (10-fold) in the samples with
the highest S. rimosus spore inoculum. This could be due to
excretion of metabolites from S. rimosus as it grows and
degrades the soil polymer components such as chitin and hemicellulose
or the death of S. rimosus hyphae and subsequent release of
nutrients into the soil matrix. The production of oxytetracycline in
the experiment indicated that S. rimosus was metabolically
active in the soil microcosms. On day 5, the R2 numbers remained high
(see above), but the percentage of biosensors which were induced had
dropped from 28 to 2.4%. This is not an unexpected development. If the
induced biosensors were exposed to tetracycline in concentrations that
slow down their growth, it would give a growth advantage to the
bacteria that are situated in areas without tetracycline. An overall
growth in biosensor numbers would therefore favor the uninduced
bacteria, hence the lower percentage. The induced biosensor cells used
for defining the R2 region were grown in the presence of 50 ng of oxytetracycline per ml. This concentration was chosen since it gives an
induction easily distinguishable from soil particles in the FACS.
Concentrations of oxytetracycline above 50 ng/ml have in our hands
slowed down growth of E. coli MC4100. It is therefore likely
that tetracycline was produced by S. rimosus locally in soil
in high enough concentrations to inhibit bacterial growth in the
vicinity of the producer. This indicates that antibiotic production can
give the producer a selective advantage under natural conditions. It
also suggests that indigenous actinomycetes may provide selective
conditions for antibiotic-resistant bacteria in natural soil. However,
since growth conditions in culture cannot be compared to those in soil
microcosms, the response of the biosensor is qualitative rather than
quantitative in this experiment. The number of biosensor bacteria
determined by plate counts accounted for only a fraction of the
biosensor bacteria added. Between 1.7 × 104 and
7.3 × 104 CFU of E. coli MC4100/pTGFP2 per
ml of extract was recovered from the microcosms on day 0 (1 h after
inoculation). This is equivalent to a recovery of the biosensor of
between 1 and 5% of the number of biosensors inoculated into the soil.
The low recovery could be due to the long (1-h) sedimentation period. Apparently, the bacteria precipitated rapidly together with the soil
particles in this type of soil. As a control, samples were taken after
10 min of sedimentation on day 2 and plated onto selective media. The
CFU counts after 10 min were 5- to 10-fold higher than in the same
samples after 1 h of sedimentation (data not shown). The 10-min samples
were, however, not suitable for analysis in the FACS, because of a high
level of particles interfering with the detection of fluorescent
bacteria.
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 6.
Enumeration of biosensor bacteria (E. coli MC4100/ pTGFP2) on LB4 containing ampicillin. Open
diamonds, microcosms containing undiluted S. rimosus spore
inoculum (A series); open squares, 100×-diluted spore inoculum
(B series); open triangles, 10,000×-diluted spore inoculum (C series);
multiplication signs, no spores added (D series).
|
|
Likewise,
S. rimosus numbers in the extract were low
compared to the number of CFU added.
S. rimosus could
be detected by
plating only in the series with the highest
inoculum density.
The counts were 3.5 × 10
2 on day 0, 9.8 × 10
1 on day 2, and
4.4 × 10
2 CFU/ml of sample on day 5. The recovery in
the sample with the
highest inoculum density was approximately 1%.
Samples from the
10-min sedimentation on day 2 (see above) showed the
same sedimentation
for
S. rimosus as for
E. coli MC4100/pTGFP2. The growth or decline
of
S. rimosus CFU could not be detected via plate counts in this
experiment.
Application potential.
We have introduced a new approach in
environmental microbiology by combining the high specificity of
bacterial biosensors and the resolution power of a
fluorescence-activated cell sorter. Application of these techniques
provided detection of oxytetracycline production by S. rimosus in soil microcosms. This has not been possible with
traditional methods, possibly due to the very low oxytetracycline
concentrations in the bulk soil.
Analyzing the samples by FACS has the major advantage that fluorescence
is detected per individual bacterial cell. This allowed
the detection
of a few fluorescent cells exposed to inducing concentrations
of
oxytetracycline among a larger number of nonfluorescing
bacteria.
Moreover, the use of a whole-cell biosensor permits detection of only
the bioavailable fraction of tetracycline. This gives
the opportunity
to study biological effects of antibiotics produced
in the environment
or introduced as pollutants in manure and sewage
sludge.
We believe that an improved method of extracting biosensor cells from
the soil can give higher recovery of biosensor cells
and less variable
detection of oxytetracycline production. This
will provide the means
for tetracycline detection at lower densities
of
S. rimosus
cells. We are currently working on applying this
approach in nonsterile
and unamended soil using different gram-negative
bacterial hosts for
the biosensor
construct.
A combination of this setup and confocal scanning laser microscopy
could give additional information on the spatial diffusion
of
antibiotics and its effect on the surrounding microflora, in
situ.
| |
ACKNOWLEDGMENTS |
We acknowledge the Department of Biological Sciences, Macquarie
University, Sydney, Australia, for patient help and constructive advice, especially Jan Gebicki for providing access to the gamma source
and Andrew J. Holmes for providing the soil used in microcosms.
This work was supported partly by the Danish Ministry of Food,
Agriculture and Fisheries, projects MIL 96-2 and MIL 96-3.
| |
FOOTNOTES |
*
Corresponding author. Mailing address: Department of
General Microbiology, University of Copenhagen, Sølvgade 83 H, DK-1307 Copenhagen K, Denmark. Phone: 45 35 32 20 53. Fax: 45 35 32 20 40. E-mail: sjs{at}mermaid.molbio.ku.dk.
| |
REFERENCES |
| 1.
|
Bertrand, K. P.,
K. Postle,
L. V. Wray, Jr., and W. S. Reznikoff.
1984.
Construction of a single-copy promoter vector and its use in analysis of regulation of the transposon Tn10 tetracycline resistance determinant.
J. Bacteriol.
158:910-919[Abstract/Free Full Text].
|
| 2.
|
Bronstad, K.,
K. Dronen,
L. Ovreas, and V. Torsvik.
1996.
Phenotypic diversity and antibiotic resistance in soil bacterial communities.
J. Ind. Microbiol.
17:253-259[CrossRef].
|
| 3.
|
Cormack, B. P.,
R. H. Valdivia, and S. Falkow.
1996.
FACS-optimized mutants of the green fluorescent protein (GFP).
Gene
173:33-38[CrossRef][Medline].
|
| 4.
|
Finley, A. C.,
G. L. Hobby,
S. Y. P'an,
P. P. Regna,
J. B. Routien,
D. B. Seeley,
G. M. Schull,
B. A. Sobin,
I. A. Solomons,
J. W. Vinson, and J. H. Kane.
1950.
Terramycin, a new antibiotic.
Science
111:85[Free Full Text].
|
| 5.
|
Gottlieb, D.
1976.
The production and role of antibiotics in soil.
J. Antibiot. (Tokyo)
XXIX:987-999.
|
| 6.
|
Hanahan, D.
1983.
Studies on transformation of Escherichia coli with plasmids.
J. Mol. Biol.
166:557-580[Medline].
|
| 7.
|
Hansen, L. H., and S. J. Sørensen.
2000.
Detection and quantification of tetracyclines by whole cell biosensors.
FEMS Microbiol. Lett.
190:273-278[CrossRef][Medline].
|
| 8.
|
Hopwood, D. A.,
M. J. Bibb,
K. F. Chater,
T. Kieser,
C. J. Bruton,
H. M. Kieser,
D. J. Lydiate,
C. P. Smith,
J. M. Ward, and H. Schrempf.
1985.
Genetic manipulation of Streptomyces: a laboratory manual.
The John Innes Foundation, Norwich, United Kingdom.
|
| 9.
|
Normander, B.,
B. B. Christensen,
S. Molin, and N. Kroer.
1998.
Effect of bacterial distribution and activity on conjugal gene transfer on the phylloplane of the bush bean (Phaseolus vulgaris).
Appl. Environ. Microbiol.
64:1902-1909[Abstract/Free Full Text].
|
| 10.
|
Rasmussen, L. D.,
S. J. Sørensen,
R. R. Turner, and T. Barkay.
2000.
Application of a mer-lux biosensor for estimating bioavailable mercury in soil.
Soil Biol. Biochem.
32:639-646[CrossRef].
|
| 11.
|
Sambrook, J.,
E. F. Fritsch, and T. Maniatis.
1989.
Molecular cloning: a laboratory manual, 2nd ed.
Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
|
| 12.
|
Silhavy, T. J.,
M. L. Berman, and L. W. Enquist.
1984.
Experiments with gene fusions.
Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
|
| 13.
|
Smit, E.,
A. Wolters, and J. D. van Elsas.
1998.
Self-transmissible mercury resistance plasmids with gene-mobilizing capacity in soil bacterial populations: influence of wheat roots and mercury addition.
Appl. Environ. Microbiol.
64:1210-1219[Abstract/Free Full Text].
|
| 14.
|
Sticher, P.,
M. C. M. Jaspers,
K. Stemmler,
H. Harms,
A. J. B. Zehnder, and J. Roelof Van Der Meer.
1997.
Development and characterization of a whole-cell bioluminescent sensor for bioavailable middle-chain alkanes in contaminated groundwater samples.
Appl. Environ. Microbiol.
63:4053-4060[Abstract].
|
| 15.
|
Stotzky, G., and H. Babich.
1986.
Survival of, and genetic transfer by, genetically engineered bacteria in natural environments.
Adv. Appl. Microbiol.
31:93-138[Medline].
|
| 16.
|
Sundin, G. W.,
D. E. Monks, and C. L. Bender.
1995.
Distribution of the streptomycin-resistance transposon Tn5393 among phylloplane and soil bacteria from managed agricultural habitats.
Can. J. Microbiol.
41:792-799[Medline].
|
| 17.
|
Vollmer, A. C.,
S. Belkin,
D. R. Smulski,
T. K. Van Dyk, and R. A. Larossa.
1997.
Detection of DNA damage by use of Escherichia coli carrying recA'::lux, uvrA'::lux, or alkA'::lux reporter plasmids.
Appl. Environ. Microbiol.
63:2566-2571[Abstract].
|
| 18.
|
Willardson, B. M.,
J. F. Wilkins,
T. A. Rand,
J. M. Schupp,
K. K. Hill,
P. Keim, and P. J. Jackson.
1998.
Development and testing of a bacterial biosensor for toluene-based environmental contaminants.
Appl. Environ. Microbiol.
64:1006-1012[Abstract/Free Full Text].
|
| 19.
|
Williams, S. T.
1982.
Are antibiotics produced in soil?
Pedobiologia
23:427-435.
|
Applied and Environmental Microbiology, January 2001, p. 239-244, Vol. 67, No. 1
0099-2240/01/$04.00+0 DOI: 10.1128/AEM.67.1.239-244.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Demaneche, S., Sanguin, H., Pote, J., Navarro, E., Bernillon, D., Mavingui, P., Wildi, W., Vogel, T. M., Simonet, P.
(2008). Antibiotic-resistant soil bacteria in transgenic plant fields. Proc. Natl. Acad. Sci. USA
105: 3957-3962
[Abstract]
[Full Text]
-
de Werra, P., Baehler, E., Huser, A., Keel, C., Maurhofer, M.
(2008). Detection of Plant-Modulated Alterations in Antifungal Gene Expression in Pseudomonas fluorescens CHA0 on Roots by Flow Cytometry. Appl. Environ. Microbiol.
74: 1339-1349
[Abstract]
[Full Text]
-
Urban, A., Eckermann, S., Fast, B., Metzger, S., Gehling, M., Ziegelbauer, K., Rubsamen-Waigmann, H., Freiberg, C.
(2007). Novel Whole-Cell Antibiotic Biosensors for Compound Discovery. Appl. Environ. Microbiol.
73: 6436-6443
[Abstract]
[Full Text]
-
Burmolle, M., Webb, J. S., Rao, D., Hansen, L. H., Sorensen, S. J., Kjelleberg, S.
(2006). Enhanced biofilm formation and increased resistance to antimicrobial agents and bacterial invasion are caused by synergistic interactions in multispecies biofilms.. Appl. Environ. Microbiol.
72: 3916-3923
[Abstract]
[Full Text]
-
Coleman, N. V., Holmes, A. J.
(2005). The native Pseudomonas stutzeri strain Q chromosomal integron can capture and express cassette-associated genes. Microbiology
151: 1853-1864
[Abstract]
[Full Text]
-
Norman, A., Hestbjerg Hansen, L., Sorensen, S. J.
(2005). Construction of a ColD cda Promoter-Based SOS-Green Fluorescent Protein Whole-Cell Biosensor with Higher Sensitivity toward Genotoxic Compounds than Constructs Based on recA, umuDC, or sulA Promoters. Appl. Environ. Microbiol.
71: 2338-2346
[Abstract]
[Full Text]
-
Anukool, U., Gaze, W. H., Wellington, E. M. H.
(2004). In Situ Monitoring of Streptothricin Production by Streptomyces rochei F20 in Soil and Rhizosphere. Appl. Environ. Microbiol.
70: 5222-5228
[Abstract]
[Full Text]
-
Bahl, M. I., Hansen, L. H., Licht, T. R., Sorensen, S. J.
(2004). In Vivo Detection and Quantification of Tetracycline by Use of a Whole-Cell Biosensor in the Rat Intestine. Antimicrob. Agents Chemother.
48: 1112-1117
[Abstract]
[Full Text]
-
Bahl, M. I., Sorensen, S. J., Hansen, L. H., Licht, T. R.
(2004). Effect of Tetracycline on Transfer and Establishment of the Tetracycline-Inducible Conjugative Transposon Tn916 in the Guts of Gnotobiotic Rats. Appl. Environ. Microbiol.
70: 758-764
[Abstract]
[Full Text]
-
Leveau, J. H. J., Lindow, S. E.
(2001). Predictive and Interpretive Simulation of Green Fluorescent Protein Expression in Reporter Bacteria. J. Bacteriol.
183: 6752-6762
[Abstract]
[Full Text]
Top
Abstract
Introduction
