University of Bielefeld
Faculty of Science
Department of Animal Ecology
Academic year 2010-2012
Effects of the antibiotic Tetracycline: sublethal nematode toxicity tests
Matthew Vangheel
1
Promoter: Prof. Dr. Walter TraunspurgerCo-promoter: Dr. Nicole Spann
Thesis submitted to obtain the degree of European Master of Science in Nematology
Effects of the antibiotic Tetracycline: sublethal nematode toxicity tests
Matthew VANGHEEL1,2*
1 University Gent, Department of Biology, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium
2 University Bielefeld, Department of Animal Ecology, Morgenbreede 45, 33615 Bielefeld, Germany
* Corresponding author email: [email protected]
2
Summary - Modified ISO tests were used to examine the effects of the antibiotic
Tetracycline on the reproduction, growth and population growth rate of the free-living
nematode Caerorhabditis elegans. Tetracycline has been observed in open waters and
soils today, as a consequence of inappropriate usage. In a comparison of two different
media, one liquid, one more viscous (gellan gum), C. elegans was found to be more
sensitive in the latter. This was determined from a decrease in the animalsʼ reproduction
over a range of different concentrations. In the liquid medium, growth was found to be
more tolerant to the toxic effects of the antibiotic than reproduction. Lower inhibition
concentrations (IC) were recorded for reproduction. The nematodeʼs intrinsic growth rate
was estimated at the end of a prolonged ISO test, conducted in the more viscous medium.
After 15 days, the population growth rate of C. elegans was estimated and a decrease was
noted correlated with increasing concentrations of Tetracycline. For concentrations higher
than 3mg/L, the carrying capacity was reached 3 days later than seen in the control.
A food attraction experiment to examine the nematodeʼs appeal to Tetracycline
contaminated food was unsuccessful in drawing unambiguous conclusions. Finally, it was
obvious that the Lowest Observed Effect Concentrations (LOEC) for different endpoints,
examined in both media, were far beyond those concentrations found in the environment.
Keywords - Caerorhabditis elegans, ISO test, intrinsic growth rate, ecotoxicology,
veterinary medicine, risk assessment, CNGG
3
Todayʼs livestock farming requires the usage of many pharmaceuticals, in order to
keep the animals sound and to increase the production rate. Products are used to
stimulate growth, to induce ovulation, but more important in prevention or treatment of
parasites and bacterial diseases (Boxall et al., 2003). These farm animals are often bred
on a large scale and as a consequence animal medicine or other drugs are applied in
large quantities. These amounts are catching up with those of typical agrochemicals
(Jones et al., 2001; Benbrook, 2002). Only in some situations the dosage is specifically
adapted to the needs of one or a small group of animals. In most cases the entire livestock
is treated even if the majority does not show any symptoms. Unadjusted dosages lead to
incomplete metabolisation and not metabolised and degradation products are excreted by
the animal (Boxall et al., 2003). The excessive use of pharmaceuticals has led to an
increase in concentrations of these potentially harmful compounds in natural soils and
aquatic systems (Hamscher et al., 2002; Kolpin et al., 2002). Active components can form
a serious risk for both public health and natural ecosystems (Boxall et al., 2003;
Kümmerer, 2008; Sibley & Hanson, 2011).
Luckily degradation of pharmaceuticals can occur under natural conditions, resulting
in abiotic products (Oka & Ikai, 1989). Once parental compounds and their derivates (biotic
or abiotic degradation products) are leached into the environment, they can disperse in
many ways. During their journey they are subjected to different environmental conditions,
possibly enhancing the degradation process. As noted by Boxall et al. (2003), animal
manure is often collected and stored for a long period before application. Storage keeps
the pharmaceuticals out of the environment, until the manure is used to fertilise the fields.
This extra storage time may contribute in decreasing the effectiveness of the
pharmaceutical, but there is a lot of variation in degradation times between chemical
compounds (Dolliver et al., 2008). Furthermore, under the right conditions storage of
manure may lead to regenerated parental compounds (Langhammer & Buening-Pfaue,
4
1989) and degenerated products are not always less harmful (Boxall et al., 2003). The
activity of these compounds depends a lot on their chemical structure and the surrounding
physical conditions. Variations in pH may alter binding properties and isoelectric points.
Hydrophobic structures can reduce their solubility. Cation exchange capacity may result in
captured molecules. All these physico-chemical properties may reduce the chemicalʼs bio-
availability (Le-Minh et al., 2010; Yang et al., 2010; Zhou et al., 2011). Sorption to soils or
sediments may even completely inhibit any effect. Although usually only found in very low
concentrations, the continuous supply through various ways makes pharmaceuticals and
their degradation products pseudo-persistent in the environment (Koschorreck & de Kecht,
2008) and therefore still potentially dangerous.
Antibiotics and other pharmaceuticals bring forth biological effects, even at low
concentrations. They are specifically constructed for this purpose, making them potentially
more harmful than other chemical compounds (Williams, 2005). According to Boatman
(1998) more than 2500 tons of the antibiotic tetracyclines are used every year in veterinary
therapy, of which tetracycline (TC), chlortetracycline (CTC), and oxytetracycline (OTC) are
applied the most. This group of antibiotics are broad-spectrum agents, active against a
wide range of microorganisms. For example, different types of gram-positive and species
of gram-negative bacteria, atypical organisms such as chlamydiae, mycoplasmas,
rickettsiae and protozoan parasites are affected. The octahydronaphthalene ring, which is
the backbone structure of this group of antibiotics, binds to 30S ribosomal RNA. This
binding prevents translation and hence inhibits the synthesis of bacterial proteins (Chopra
& Roberts, 2001). As mentioned above, some of the antibiotic may be excreted by the
animal after treatment. Between 50 - 80% (of the applied dosage) of not metabolised
material can be found in animal urine. The variance is due to different factors such as age,
route of administration, pH of urine, glomerular filtration rate and the particular tetracycline
used (Aiello, 1998).
5
Table 1 - Concentrations TC found in the environment, as reported in recent literature
Environment Concentration Reference
Soil fertilised with animal manure:
Swine slurry 295 ± 128a1 Hamscher et al. (2005)
Livestock 20.9 - 105a2 Hu et al. (2010)
Swine 90.0 ± 14.7a1 Qiao et al. (2012)
Fish 86.2a Hamscher et al. (2002)
Dairy cattle 55a Watanabe et al. (2010)
Swine 9a Zilles et al. (2005)
River sediment Pearl Rivers (CHN)c 40.9a Yang et al. (2010)
Cache la Poudre River (USA) 32.8 ± 0.3a1 Kim & Carlson (2007)
Surface water Jiangsu Province (CHN) 0.81b Wei et al. (2011)
Queensland (AUS) 0.080b Watkinson et al. (2009)
Jiulongjiang River, Fujian (CHN) 0.050b Zhang et al. (2012)
Haihe River (CHN) 0.027b Luo et al. (2011)
Cache la Poudre River (USA) 0.02 ± 0.01b1 Kim & Carlson (2007)
Mess River (LUX) 0.017 ± 0.003b1 Meyer et al. (2011)
Atibaia River (BRA) 0.011b Locatelli et al. (2011)
Choptank River (USA) 0.005b Arikan et al. (2008)
WWTPd effluent (PRT) ≤ 6b Cavnati et al. (2012)
effluent (CAN) 1.0b Boussu et al. (2007) Miao et al. (2004)
influent (USA) 0.1 - 0.6b2 Kim et al. (2005)
influent (NLD) < 0.1b Schrap et al. (2003)
1values represent the means ± standard deviations2values represent a range a µg/kgb µg/Lc ISO 3166 Codes (Countries)d wastewater treatment plants
6
A high quantity of tetracyclines leach into the environment, forming a potential danger
for all sorts of different ecosystems. Hamscher et al. (2005) sampled farmed soils that
were fertilised using animal manure and found a maximum concentration of 0.30 mg
Tetracycline (TC) per kilogram of soil. Table 1 summarises concentrations found in recent
literature. A clear distinction between aquatic and sediment samples is due to sorption
onto sediment particles (Le-Minh et al., 2010; Yang et al., 2010).
Because of its abundance in the environment, researchers have been aware of the
potential problem and tried to asses the environmental risks. Many different approaches,
using all sorts of organisms, help us to evaluate the problem. Different tests have been
done, mostly on bacteria -as they are the main target of the antibiotic (Chopra & Roberts,
2001; Christensen et al., 2006; Halling-Sørensen et al., 2002)- but also on phytoplankton
(Wilson et al., 2004), periphyton (Quinlan et al., 2011) and water plants (Brain et al. 2005;
Pomati et al., 2004). Araujo & McNair (2007) tested the effects of TC on the reproduction
and lifespan of two species of rotifers: Brachionus calyciflorus Pallas (freshwater) and B.
plicatilis Müller (brackish water). In general reproduction was reduced and shifted to an
earlier age. Furthermore the lifespan was shortened. The 1% Inhibition Concentration (IC)
for lifespan ranged between 31.5 and 32.43 mg/L for B. calyciflorus and for B. plicatilis
between 28.11 and 28.80 mg/L. The IC50 was noted between 171.21-172.74 mg/L and
329.69-331-31 mg/L respectively. Considering the effects on life time reproduction IC1-
values were much lower: between 3.53-4.11 mg/L (B. calyciflorus) and 8.60-9.07 mg/L (B.
plicatilis). The IC50 was estimated between 78.81 - 79.56 mg/L and 119.41-121.47 mg/L
respectively. Wollenberger et al. (2000) assessed the acute and chronic toxicity on
Daphnia magna. Acute effects were only recorded in dosages far beyond ecological
relevance (no effect was shown below a 100 mg/L concentration). The IC10 of
reproduction was estimated at 29.4 (24.6-35.2) mg/L.
7
Within this research nematodes will be used in the experimental setup, because they
are ubiquitous and highly abundant animals (Bongers & Bongers, 1998) and according to
Höss et al. (2009) essential in ecotoxicological assessments. The soil free-living nematode
Caenorhabditis elegans Maupas (1899), was proposed by Brenner (1974) as an “easy-to -
work-with-organism”. Its ease of cultivation in the lab, short generation times and
hermaphrodite reproduction made it excellent for genetic testing. Ever since Brenner
suggested C. elegans, many other scientists have seen its potential in many different
fields. The nematode has crawled into the world of ecotoxicology and has been used in
numerous experiments. Höss & Williams (2009) summarised the benefits of using
nematodes and in particular C. elegans, in assessing the effects of potential toxins on the
environment.
To assess the possible environmental impacts of specific chemicals, an integration of
different disciplines is absolutely essential (Caswell, 1996; Van den Brink, 2008). Only by
combining the data and findings of different experimental approaches, effects can be
evaluated. A simple standard test may yield some results and be the foundation of more
in-depth research, such as life cycle, behaviour, mesocosm studies and eventually
molecular research. Although more relevant than biochemical, molecular or cellular testing,
the life cycle tests of individuals are still far off from the environmental effects.
Assessments of changes in demographic structures within a population or micro- and
mesocosm experiments (that involve different organisms), are more pertinent to the
investigation, but naturally have a high complexity (Höss & Williams, 2009). The ultimate
experimental setup with the highest relevance are of course field experiments, conducted
in situ. However, it is hard to standardise or eliminate parameters, leaving it very difficult to
interpret the results. This trade-off forces the investigators to integrate different
approaches, in order to make a good environmental assessment (Van den Brink, 2008;
Sibley & Hanson, 2011). In this research, it was chosen to test the effects on a single
8
organism in vitro. This approach can be found somewhere in the middle between clear
results and ecological relevance (Höss & Williams, 2009).
To test the effects of Tetracycline on C. elegans, different experimental endpoints
were taken into account. Alterations in growth (Kim et al., 2012; Dhawan et al., 1999),
reproduction (Kim et al., 2012; Anderson et al., 2001) and population growth rate (Kim et
al., 2012 Ohba & Ishibashi, 1984) were examined. These parameters are of great
ecological relevance and can clearly indicate how an organism responds to a potential
toxin. The aim of this study is to assess the toxicity of Tetracycline, by observing the effect
on those different endpoints. Possible alterations of reproduction were examined in two
different media: K-medium and CNGG (Brinke, Heininger & Traunspurger, 2011). Both
differ in their viscosity and resemble different environments in which C. elegans can
confront the pharmaceutical. Furthermore, behavioural effects were examined by tracking
the nematodeʼs movement towards a food source (Dhawan et al., 2000; Boyd & Williams,
2003; Ma et al., 2009). If hampered, the animal may not find enough nutrients to
reproduce or grow, indicating a secondary effect of the antibiotic. Expectations were that
Tetracycline will affect all endpoints negatively. The question however was to what extent.
9
Material & Methods
Preparation of the food source
The bacterium Escherichia coli strain OP50 was chosen as the food source in all the
conducted experiments. Stock cultures were prepared and stored in agreement with ISO
10872:2010. Prior to experimental setup, 25mL of autoclaved LB-medium (10g/L NaCl,
10g/L peptone, 5g/L yeast extract) was inoculated with 30µL of thawed stock culture in an
Erlenmeyer flask. After 17h in an incubator (37°C) on a shaker (200rpm), the medium was
washed (using K-medium: 3,1g/L NaCl, 2,4 g/L KCl) in order to remove excretion and
waste products. Therefore the medium was first centrifuged (2000g for 20min) without
addition of K-medium. After removal of the supernatants the bacterial pellet was
resuspended in K-medium and centrifuged again. The resulting much clearer supernatant
was again discarded and the bacterial pellet resuspended in K-medium in order to
measure the bacterial density (photo-spectrometer: 600nm). Bacterial densities were
different for each experimental setup and will therefore be discussed further in detail
below.
Test organism
Caenorhabditis elegans Maupas (1899) strain N2 (Bristol, UK) was selected to be an
appropriate test organism. It was cultured in agreement with ISO 10872:2010 on nutrient
agar plates (15g/L agar, 0,5 g/L peptone, 0,3g/L yeast extract, 0,1g/L NaCl and 5mg/L
cholesterol). Cultures were prepared 4 to 7 days before experimental setup. To obtain first
stage juveniles (J1), the plates were rinsed (K-medium) above stacked nylon mesh sieves
(10µm and 5µm pore size) and captured in a beaker.
10
Tetracycline
The Tetracycline hydrochloride (TC) (Molecular Weight 480.90; PubChem Substance
ID 24900468; ≥95% CELLPURE®) was purchased from Roth as a powder. Stock solutions
of 4g/L were prepared, by dissolving the powder in de-ionised water. They were kept in
glass flasks, wrapped in tinfoil at 7°C for no longer than 12 days.
ISO test (K-medium)
The ISO test was conducted according to ISO 10872:2010, with only few
modifications. The bacterial density was adjusted to (1000±50) FAU (ISO 7027) and
cholesterol stock solution was added (0,2% of the total volume of the bacterial suspension
in K-medium). From the TC stock solution, a series of concentrations was prepared by
dilution with K-medium (0,1,5,10,20,50,100,150,200 mg/L). Five replicas of each
concentration (plus K-medium as control), were set up in 12-well plates. Therefore 0,5 mL
of TC-solution (K-medium for control) was transferred to each well and in addition, 0,5 mL
of well-stirred homogenised bacterial suspension, resulting in 45 wells of different TC
concentrations and a bacterial density of 500 FAU. Ten moving J1 juveniles were
transferred to each well, with a glass pasture pipet from which the tip had been narrowed
down. Furthermore 30 J1 juveniles, were transferred to a K-medium solution in a Petri
dish, stained using Rose Bengal stock solution and put in an drying oven for 15 min at
80°C. The 12-well plates were sealed using Parafilm® and incubated in the dark (at 20°C
± 0,5) for 96h. Since the generation time of C. elegans is around 2,5 days, a 4 day runtime
would yield 10 adults and their progeny. To determine the test, Rose Bengal stock solution
was added to each well and the plates were placed in a drying oven for 20 min at 80°C.
Samples that were not analysed immediately, were stored in a fridge (app. 7°C).
To analyse the samples, the content of each well was transferred to a counting dish.
The well and pipet were both checked thoroughly after transition to make sure all
11
nematodes were transferred. Adults were picked out using a metal fishing needle and
stored in K-medium. Juveniles (progeny) were counted under a dissecting microscope
(Leica S6E) at 10x3,2 magnification. The effects of different concentrations TC on
reproduction were assessed by dividing the total number of juveniles by the total number
of adult females. Samples containing adult males were checked if they were in agreement
with the restrictions found in ISO 10872:2010.
The 30 juveniles that were isolated before the start of the experiment, were
transferred to a slide and their length was measured (10x10 Zeiss A-plan; Microscope:
AX10). The adults were also transferred to slides and measured (10x5 Zeiss N-achroplan;
Microscope: AX10). To determine growth, the average length of 30 juveniles was
subtracted form the measured length of the adult females.
Inhibition of reproduction and growth was calculated in reference to the average
number of juveniles per females and the average growth of the control samples. The
yielded percentages were then fed into the statistical program R (R Core Team, 2012) and
dose-response curves (non-linear regression: 1/{1+(x/α2)α1} with α1 and α2 as parameters)
were calculated using the package ʻdrcʼ, self starter ʻLL.2ʼ (Ritz & Streibig, 2005).
Furthermore, estimates of Inhibition Concentrations (IC) were calculated.
The data were checked for significant differences between treatments by running an
ANOVA in R. To satisfy the assumptions, all data (reproduction and growth) were log(1+x)
transformed. Subsequently the significance level between control and treatments were
checked by a post hoc Dunnett-test, with the ʻmultcompʼ package (Hothorn, Bretz &
Westfall, 2008) in R.
ISO test (CNGG)
The ISO test was conducted according to ISO 10872:2010, with only few
modifications. The bacterial density was set to 5E9 cells/mL, therefore a corresponding
12
amount of suspended bacterial solution was transferred to 15mL plastic centrifugal tubes.
After centrifugation (2000g for 20min) the supernatant was removed and 8mL of sterile
Gelrite® solution (1,876g/L Gelrite®, 1,250mL/L cholesterol) was added to each tube. In
total 9 tubes were prepared: control plus 8 different concentrations (1,5,10,20,50,
100,150,200 mg/L) of TC. To each tube, 1mL of corresponding concentration TC (diluted
from stock using sterile de-ionised water) and 1mL of sterile de-ionised water for the
control was added, followed by 1mL of sterile salt solution (1M CaCL2, 1M MgSO4). After
each addition the tubes were mixed thoroughly (Vortex) to guarantee equal distribution.
1mL of each concentration (5 replicas) was pipetted into a well of a 12-well plate.
Afterwards 10 J1 juveniles were transferred to each well using a glass pipet. The 12-well
plates were sealed using Parafilm® and incubated in the dark (at 20°C ± 0,5) for 96h. To
determine the test, 4 mL EDTA-Rose Bengal solution (0,04M EDTA, Rose Bengal stock
solution) was added to each well and the plates were placed in a drying oven for 20 min at
80°C. Samples that were not analysed immediately, were stored in a fridge (approx. 7°C).
Sample processing was similar to the K-medium ISO test described above, with the
exception of measurements being taken. Inhibition of reproduction was calculated in
reference to the average number of juveniles per female of the control samples. The data
were analysed using the statistical program R (ʻdrcʼ package, non-linear regression: 1/{1+
(x/α2)α1} with α1 and α2 as parameters), yielding dose-response curves and IC-estimates.
An ANOVA was ran to check significant differences between treatments. The assumptions
were met after log(1+x) transforming all data. Furthermore the significance level between
control and treatments were checked by a post hoc Dunnett-test, with the ʻmultcompʼ
package in R.
13
Population growth rate
The CNGG ISO testʼs yielded EC-values were the starting point for the population
growth experiment, which is in many ways similar to the ISO test (CNGG) but for an
extended period of time. Seven concentrations (5 replicas) of TC were tested (0, 1, 3, 7,
17, 68 mg/L), corresponding with control, IC10, IC30, IC50, IC70 and IC90 of the ISO test
performed in CNGG medium. At the start of the experiment, 1mL of each solution was
again transferred to a well within a 12-well plate and stored in the dark at 20°C ± 0,5. After
3 days a subsample was taken from each well, by pipetting 1/5 of the total volume (200µL)
into a clean Petri dish and adding 3,8 mL of EDTA (0,02 M) and Rose Bengal. These Petri
dishes were placed in a drying oven for 20 min at 80°C and stored in a fridge (approx.
7°C) until examination. Individual nematodes (adults and juveniles) were counted in each
Petri dish, using a dissection microscope. The wells were refilled up to 1mL by adding
200µl of corresponding solution (correct concentration of TC and bacteria), which was
prepared the same day from resuspended bacterial pellets. By adding fresh medium after
3 days, the degradation of TC and loss of bacteria through consumption was partially
countered. The procedure was repeated every 3 days and after 15 days the experiment
was determined.
To estimate the total number present in each replica at each sampling day, the
counted numbers of individuals from the subsample were multiplied by 5.
The intrinsic population growth rate (PGR, r), defined as the growth rate of an
increasing population (Lotka, 1913), was estimated for each replica and each
concentration (grouped replicas) in SigmaPlot v11.0 by means of a non-linear regression:
14
Nt: number of individuals per dayN0: initial number of individualsr: population growth ratet: time (day)K: carrying capacity
This formula generated an approximation of the different PGRs under test conditions.
However these PGRs have to be corrected because of the interference each 3 days
(subsampling), which leads to an artificially higher mortality rate, estimated by:
Therefore the corrected growth rate (rʼ) is calculated by adding r and m. rʼ reflects the
PGR under undisturbed conditions.
To check if there were significant differences in PGRs between treatments, an
ANOVA was run in R after the data had been log(1+x) transformed, followed by a post hoc
Dunnett-test. In order to construct a dose-response curve in R (ʻdrcʼ package, non-linear
regression: 1/{1+(x/α2)α1} with α1 and α2 as parameters), percentages of inhibition were
calculated in reference to the average PGR under control conditions. Furthermore
estimates of IC-values were calculated.
Food attraction
To see if food attraction is hampered by TC, an additional experiment was set up.
Firstly Volvic agar (Volvic water, 3g/L NaCl, 17g/L agar) was prepared, of which 25mL was
transferred to 8.5cm Petri dishes after sterilisation.
After solidification, the agar was perforated using a
specially designed stencil, leaving 8 similar sized
holes located at the same distance form the centre of
the dish, approx. 3cm (figure 1). The holes were
refilled with 150µL of Volvic agar, forming cavities in
which the test solutions could be pipetted. All
procedures were conducted under sterile conditions.15
m: artificial mortalityn: times sampledTn: last sample day (days)T1: first sample day (days)
Figure 1: Food attraction test plates
Three different experimental setups (8 replicas) were carried out. In the first one 4
cavities were filled with a TC-contaminated bacterial solution (5E9 cells/mL, K-medium,
38mg/L TC) and 4 with K-medium. The cavities were filled with 350 µL of a solution in an
alternating pattern, so that each contaminated sample was surrounded by 2 non-
contaminated samples. Twenty mobile adult female nematodes were transferred to the
centre of each plate (minimum 2 cm away from the cavities). They were allocated from a
7-days-old culture plate using an eyelash mounted on a toothpick or glass pipet. Every 30
min the dishes were checked using a dissecting microscope, to see if the nematodes had
moved and the observations were noted down. This procedure was repeated 6 times and
once more after 24h, assuming that most worms had made their final decision after one
day. The experiment was repeated twice more, with different solutions: contaminated
bacterial solution (5E9 cells/mL) versus uncontaminated bacterial solution (5E9 cells/mL)
and contaminated bacterial solution (5E9 cells/mL) versus uncontaminated low density
bacterial solution (2E8 cells/mL).
The data were combined per replica for each time interval, thereby neglecting intra-
replica variation, leading to only 2 data points per replica per time. To see if these were
significantly different, a likelihood G-test was run in R (Deducer package; Fellows, 2012).
A difference would indicate a preferred food choice.
16
Results
ISO test (K-medium)
A decrease in both growth and reproduction (juveniles per female) corresponding to
increased concentrations of TC is described in figure 2. A clear effect on reproduction
(ANOVA: F-value: 472.77; p < 2.2e-16) and on growth (ANOVA: F-value:245.64; p <
2.2e-16) is proven to be significant.
!"
!!!"
!!!"!!!" !!!"
#"$#"%#"&#"'#"(#")#"*#"+#",#"$##"
#" $" (" $#" %#" (#" $##" $(#" %##"
!"#$%&'$()*
)+$,-'$)
./)0,1*23)45)
!!" !!!" !!!"
!!!"
!!!"!!!"
!!!"
#$%"
#$&"
#$'"
($("
($)"
($%"
($&"
#" (" %" (#" *#" %#" (##" (%#" *##"
!"#$
%&'()
)*'
+,'()-./*'01'
Figure 2: Test conducted in K-medium, 96h, N=5. 2A: representation of the average reproduction (juveniles/female) per concentration of TC. Error bars indicate standard deviation. 2B: representation of the average growth (mm) per concentration of TC. Error bars indicate standard deviation. Significant differences between treatment and control are indicated: ʻ*ʼ : p <0.05 ; ʻ**ʼ : p <0.01; ʻ***ʼ : p <0.001
17
Figure 3: Dose-response curves, K-medium, 96h, N=5. Replicas are displayed as ʻ°ʼ . Inhibition is plotted in ratios (e.g. 0.5 = 50% inhibition) The concentrations of TC are plotted on a logarithmic scale, hence the broken axis for the control (0mg). 3A: Inhibition of reproduction (in reference to the control average). 3B: Inhibition of growth (in reference to the control average).
18
The decrease in progeny (figure 2A) starts to differ significantly from the control from
20mg/L TC onwards. 20mg/L can therefore be referred to as the Lowest Observed Effect
Concentration (LOEC), considering reproduction in K-medium. However, the lowest
observed effect of TC on growth is significantly different from the control at a much lower
concentration of 5 mg/L. This indicates that the LOECs are dissimilar for different
parameters. Nevertheless both reproduction and growth show a negative trend with
increasing concentration higher than 1mg/L.
The dose-response curves (figure 3) plot the inhibition ratios in relation to the
concentrations of TC. It is found that the inter-replica variation is small and extreme
measurements are absent. In both fig. 3A and fig 3B, the data points approach the plotted
curve within acceptable distances.
Table 2 - Calculated Inhibition Concentrations (mg/L) of TC
K-medium CNNG
Reproduction Growth Reproduction PGR
IC10 8.7 ± 1.1 13.6 ± 1.3 0.8 ± 0.2 3.0 ± 0.8
IC30 19.5 ± 1.5 81.1 ± 3.0 3.1 ± 0.5 20.5 ± 2.5
IC50 32.4 ± 1.9 249.0 ± 13.4 § 7.4 ± 0.8 68.8 ± 11.1
IC70 53.8 ± 3.2 764.1 ± 74.7 § 17.4 ± 2.0 231.2 ± 62.1 §
IC90 120.6 ± 12.1 4561.7 ± 798.9 § 67.9 ± 14.4 1596.1 ± 744.5 §
Values represent arithmetic means ± SD.
ʻ§ʼ Indicates concentrations beyond the scope of the experiment and are therefore not considered to be of
any value.
Although not found significant, a 10% inhibition of reproduction is seen in figure 3A
between the 5mg/L and 10mg/L concentration data points. Reproduction is inhibited
almost entirely at the highest concentrations, where growth is not. Reproduction is
inhibited by 50% in K-medium at a calculated concentration of 32.391 ± 1.8520 mg/L TC.
Growth however, is not inhibited by 50% or more in any of the samples. Estimations of 50-
percent inhibitory concentrations for growth, are much higher than the IC50 of reproduction 19
(Table 2). Figure 3B does not show a sigmoid curve, because absolute inhibition of growth
is not reached given the test concentrations. Although shown significantly that growth is
affected by TC, it appears not to be inhibited as much as reproduction.
ISO test (CNGG)
Figure 4 shows a clear effect of TC on the reproduction in CNGG (F-value: 337.22; p
< 2.2e-16). The average progeny per female is affected negatively with increasing
concentrations. The treatments with concentrations of 5mg/L and higher are significantly
different from the control.
Remarkably the effect of TC on reproduction seems to be more pronounced in
CNGG than in K-medium (lower IC values; table 2). The IC50 in K-medium is around 4
times higher, than in CNGG, even though the bacterial concentration is 25 times higher in
CNGG. The higher concentration is found to be necessary in CNGG to allow C. elegans to
reproduce, after preliminary experiments (data not shown). Other than the bacterial density
!!!"
!!!"!!!"
!!!"!!!" !!!"
#"
$#"
%#"
&#"
'#"
(#"
)#"
#" $" (" $#" %#" (#" $##" $(#"
!"#$%&'$()*
)+$,-'$)
./)0,1*23)
Figure 4: Test conducted in CNGG, 96h, N=5. Representation of the average reproduction (juveniles/female) per concentration of TC. Error bars indicate standard deviation. Significant differences treatment control are indicated: ʻ*ʼ : p <0.05 ; ʻ**ʼ : p <0.01; ʻ***ʼ : p <0.001
20
and the composition of the two different media, the experimental setups do not differ. It
was thrived to keep differences to the minimum, to make comparison possible.
The variance between replicas is larger in figure 5 at low concentration, but still within
acceptable ranges. This variance decreases with increasing concentration. The data
points line up with the plotted curve and no outliers were found.
Population Growth Rate (PGR)
Figure 6 shows the increase in absolute numbers over a period of time (15d). Due to
limited resources, a carrying capacity (K) was reached after 9 days in the control and 1mg/
L setup. It took 3 days longer in the 3, 7 and 17mg/L treatments to reach K and for the
68mg/L treatment K was barely reached after 15 days. TC seemed to effect the growing
abilities of a C. elegans population negatively, requiring more generations to reach K.
Figure 5: Dose-response curve, CNGG, 96h, N=5. Replicas are displayed as ʻ°ʼ . Inhibition is plotted in ratios (e.g. 0.5 = 50% inhibition) The concentrations of TC are plotted on a logarithmic scale, hence the broken axis for the control (0mg). Inhibition of reproduction (in reference to the control average).
21
Especially in the control and 1mg/L samples, the carrying capacity was surpassed
after day 9, followed by a collapse in numbers (recorded at day 12). The same event was
seen in the 3, 7 and 17mg/L treatments, were it 3 days later. No such high peaks could be
confirmed in the 68mg/L samples, given the huge variation between them. However, the
average values never seemed to transcend the plotted curve.
!"
#!!!"
$!!!!"
$#!!!"
%!!!!"
%#!!!"
&!!!!"
&#!!!"
!" %" '" (" )" $!" $%" $'" $("
!"#$%&'&%(
)*+""
,)-"
./"*+,-.+/"
$01"
&01"
201"
*+,-.+/3"
$013"
&013"
2013"
!"
#!!!"
$!!!!"
$#!!!"
%!!!!"
%#!!!"
&!!!!"
&#!!!"
!" %" '" (" )" $!" $%" $'" $("
!"#$%&'&%(
)*+"
,)-"
./"
*+,-.+/"
$01"
$201"
()01"
*+,-.+/3"
$013"
$2013"
()013"
Figure 6: Population Growth Rate. Calculated non-linear regression curve. CNGG, 15 days, N=5. Data indicated with an ʻAʼ refers to the average number of individuals counted per sampling day. Error bars indicate standard deviation. Data points are slightly shifted to the right (in reference to the control data points) for reading convenience. Both control and 1mg-treatment data points are displayed in figure 6A and 6B, to make comparison between figures possible.
22
The calculated PGR (r) was corrected and is shown in figure 7 as rʼ. This rʼ-value is
an estimation of the actual PGR under unlimited conditions. The PGR differs significantly
from the control in all treatments with exception of the first, indicating an effect of TC on C.
elegansʼs growth rate in CNGG. The effect seemed to be negative for concentrations
higher than 1mg/L. One exception was noted for 17 mg/L, which seemed to indicate a
higher growth rate than the 7mg/L samples, but the difference between these treatments
was found not to be significant (t-value: 0.663; p-value: 0.738), nor was there any
difference between 3mg/L and 17mg/L treatments (t-value: -1.407: p-value: 0.384).
The dose-responsive curve (figure 8), shows an incomplete sigmoid curve. The used
concentrations seem not to effect the PGR more than 50%. In comparison to the
reproduction (table 2), PGR seems to be a more rigid parameter (e.g. IC50 is 10 times
higher).
!!"!!!" !!!"
!!!"
#$%"
#$&"
#$'"
#$("
#$)"
*"
*$*"
*$+"
*$,"
*$-"
#" *" ," '" *'" &("
!"#$%&'($
)*$%+,-.($
Figure 7: Test conducted in CNGG, 15 days, N=5. Representation of the corrected calculated population growth rate r ̓ per concentration of TC. Error bars indicate standard deviation. Significant differences between treatment and control are indicated: ʻ*ʼ : p <0.05 ; ʻ** ̓ : p <0.01; ʻ*** ̓ : p <0.001. Test concentration are related to IC-values of the ISO-test (CNGG): IC10:1mg; IC30:3mg; IC50:7mg; IC70:17mg; IC90:68mg
23
Food attraction
The checks after 24h yielded the most nematodes on average in each experimental
setup (figure 9). Numbers increased over time, indicating that once a spot was reached
nematodes probably stayed and only in few occasions a second choice was made.
Remarkable is that in none of the replicas (data not shown), the total number (summation
of both treatments) exceeded 15. On average not even 10 of the 20 nematodes
transferred to the plates, made a final choice after one day. Most of them were located in
the centre of the Petri dishes, but at a certain distance from where they were initially
placed. These low numbers indicate a low attractiveness in general. Comparing
treatments, figure 9A shows a small preference of contaminated food over the nutrition low
K-medium after 24h. However, this difference is not significant (table 3). Figure 9B shows
Figure 8: Inhibition of PGR (r) (in reference to the control average). Dose-response curve, CNGG, 15days, N=5. Replicas are displayed as ʻ°ʼ . Inhibition is plotted in ratios (e.g. 0.5 = 50% inhibition).The concentrations of TC are plotted on a logarithmic scale, hence the broken axis for the control (0mg).
24
fluctuating preferences over time, but a higher appetite for the uncontaminated food
source after one day. The food preference was nonetheless not found to be significant
(table 3). In the third experimental setup, both food sources were contaminated but
treatments differed in their bacterial concentration. On average 7 nematodes made a
choice in each replica (figure 9C), but without any clear fondness for one of the food
sources (table 3). In summary, there seems to be no significant preference in any of the
setups after 24h.
%
Figure 9: Food attraction, 24h, agar, N=8. The average number (grouping intra- and inter-replica data) of nematodes counted in each treatment at a certain time point. Error bars indicate standard deviation. 9A: TC-contaminated food sourcea vs. K-medium. 9B: TC-contaminated food sourcea vs. Uncontaminated food sourcea. 9C: TC-contaminated food sourcea vs. TC-contaminated food sourceba bacterial density: 5e9 cells/mLb bacterial density: 2e8 cells/mL
25
!"
#"
$"
%"
&"
'"
("
)"
*"
+"
+!,-." #*!,-." $&/"
!"#01.23,-.3245"
67,45-8,"
!"
#"
$"
%"
&"
'"
("
)"
*"
+!,-." #*!,-." $&/"
!"#01.23,-.3245"
6.71.23,-.3245"
!"
#"
$"
%"
&"
'"
("
)!*+," #-!*+," $&."
!"# /+0." 123"
Table 3 - Summarised results of the food attraction experiment after 24h
Medium G-value DFa p-value
TC-contaminated food sourceb vs. K-medium 13.9 12 3.10
TC-contaminated food sourceb vs. Uncontaminatedb 11.8 10 0.30
TC-contaminated food sourceb vs. TC-contaminated food sourcec 18.4 20 0.56
A G-test was used to asses likelihood between experimental setups.a degrees of freedomb bacterial density: 5e9 cells/mLc bacterial density: 2e8 cells/mL
26
Discussion
Material and Methods
The salt solution (K-medium) used in the liquid ISO tests, was found to be an easy to
work with medium. It requires little preparation time and is easy to transfer. Furthermore it
is a very clear solution, making observations possible. On the downside, the lack of an
internal structure (matrix) or viscosity restricts the free movement of the nematodes in all
dimensions. As a consequence all action happens at the floor of the wells as both bacteria
and nematodes sink to the bottom, where gas exchange can be limited. This may induce
stress, leading to a higher energy requirement (Williams & Dusenbery, 1990). However,
given the small volume (1mL) and the relatively large surface area, gas exchange was not
considered to be a problem. The competition for space however was. Given the short
runtime of the experiment (4 days) and only 10 nematodes in each well at the start, the
effects of space limitations could be cancelled out by the results of the control.
The more viscous CNGG medium brought solution because of its internal structure
and allowed the nematodes to fully benefit from the 1mL volume. However, this meant
increasing the bacterial density 25 times, because the bactaria would also disperse,
leading to a relatively lower density. The medium had proven to be a good alternative
(Muschiol & Traunspurger, 2007) and provides the possibility to accurately subsample.
This was absolute necessity in the population growth experimental setup. However, the
presence of an organic compound (Gelrite®), may have interacted with TC, resulting in
bio-availability differences between the two test media. In a comparison study of both
hydrophobic (ivermectin) and soluble chemicals (aldicarb and cadmium), Brinke, Heininger
and Traunspurger (2011), concluded that the results between experiments conducted in
liquid and CNGG test medium, only differed significantly in the case of cadmium. This
effect was dedicated to the chemical properties of the divalent cation. As a consequence
27
the bio-availability was reduced, as cadmium might have been incorporated into the
medium. CNGG was therefore not recommended for assessing heavy metal toxicity.
ISO test (K-medium)
As seen in figure 2, TC affects both reproduction and growth negatively. The LOECs
(20mg/L and 5mg/L, respectively) however differ between parameters. This is not unusual
as previous studies have shown that the correlation is test specific. Although reproduction
is generally considered to be the more sensitive of the two (Traunspurger et al., 1997;
Höss et al., 2009; Hao et al., 2010), differences are not always as pronounced (Höss et al.,
1999). Although the variance between treatments and control were only to be found
significant at 20mg/L when considering reproduction, the dose-responsive curves and the
calculated IC-values proved differently. The contrasting significant levels, are possibly due
to the high variability in progeny at low concentrations. The variance between growth is
less pronounced. Figure 3 clearly shows how reproduction decreases more than growth
under the experimental conditions. Therefore it can be said that C. elegans subjected to
the test environment, reduces investments in progeny first. The decrease in growth follows
only when the organism is exposed to much higher concentrations of TC (table 2). With
this strategy, C. elegans favours its own survival over the production of progeny, possibly
saving its energy for when conditions are more favourable.
ISO test (CNGG)
Different form the ISO test conducted in K-medium, the LOEC (reproduction) in
CNGG is 5mg/L. A clear decrease in average progeny in reference to the control was
highly significant. The media seems to differ even more in the calculate IC-values (table 2),
leading to the conclusion that C. elegans is far more sensitive to the applied chemical in
CNGG than in K-medium. These findings are in particular remarkable, given the fact that
28
the food available in the former was 25 times higher than in the latter. Furthermore, pH-
tests showed no differences in acidity (both pH=5). It is unlikely that the Gelrite® present in
the CNGG can be held responsible, because interaction between TC and the organic
compound would have led to a smaller bio-availability, hence lower sensitivity. This is even
more the case, because the presence of divalent cations (Mg2+ and Ca2+) could also
reduce the bio-availability of TC. TC has been proven to be less toxic in soils or sediments
with a high concentration of divalent metals, since tetracyclines are able to chelate with
metal cations (Halling-Sørensen et al., 2002). Though presumably present, it is clear that
these interactions were overwhelmed by another event.
Possible flaws in the experimental setup, may explain the apparently larger sensitivity
in CNGG. The order and the recipient in which the liquid test were prepared, may be key
to the noted differences. In preparation of the ISO test (K-medium), a dilution series of
Tetracycline HCl dissolved in K-medium was prepared, of which 0,5mL was pipetted into
plastic 12-well plates. This was followed by 0,5mL of bacteria suspended in K-medium, but
with an inevitable time gap in between. Tetracycline HCl in solution breaks down to
Tetracycline molecules and H+ and Cl- -ions. The hydrophobic TC molecules are likely to
adhere to the plastic walls of the wells during the time gap and furthermore during the
complete run time of the experiment, leaving little of the antibiotic to interact with both
bacteria and nematodes. In preparation of the ISO test (CNGG), the different compounds
were added in a different order. Firstly, 8mL Gelrite® and a bacterial pellet were mixed to a
homogeneous suspension in a small centrifugal tube. Secondly, Tetracycline HCl diluted to
the appropriate concentration with de-ionised water, was added to the suspension,
followed by the salt solution that started the solidification of the CNGG medium. In a final
step 1mL of the prepared CNGG was transferred to each well in a plastic 12-well plate, but
by the time the medium was transferred, a matrix of compounds and bacteria had been
established with more or less a homogeneous distribution. Even though the medium was
29
stirred right before pipetting, causing it to liquify again, the CNGG solidified rapidly
afterwards. The chemical properties of the medium, possible prevented sorption of TC to
the plastic walls to some extent, leaving more to interact with the present organisms. In
addition Chen et al. (2011) concluded that the presence of Mg2+ - ions, may slow down the
normal degradation by the formation of metal-ion complexes, leading to a relatively higher
concentration of TC present in the medium.
At the time, there was no notion that the preparation order and recipients used may
have possibly influenced the results, simply because of alterations in bio-availability.
Similar problems were countered by Brinke, Heininger and Traunspurger (2011) by the
usage of glass wells. Glass or inert recipients and different preparation methods could
reduce the variation in results between the tests run in the two different media.
As stated by Brinke, Heininger and Traunspurger (2011), the sorption of an antibiotic
to food bacteria is most likely to happen, however this must not be considered as an
experimental error. Free-living nematodes, such as C. elegans, can be subjected to
contamination in many ways. Next to dissolved toxins in their surrounding aqueous
environment, particle-bound and bacteria-bound chemicals are plausible to increase
contact with a certain compound (Chandler et al., 1994; Höss et al., 2001; Offerman et al.,
2009).
It is clear that the medium composition influenced both the toxicity and the bio-
availability of TC in the performed tests. Similar problems have been addressed more in
depth by Brinke, Heininger and Traunspurger (2011), Donkin and Williams (1995) and
Khanna et al. (1997).
Population Growth Rate (PGR)
Figure 7 shows a significant decrease in C. elegansʼ growth rate from 3mg/L
onwards. This concentration corresponds with the IC30 (reproduction) found in the ISO
30
test (CNGG). Looking at the dose-responsive curve (figure 8), 3mg/L correlates with a
10% inhibition of the nematodes growth rate. Furthermore, table 2 provides a clear
overview on how IC-values variate between different characteristics. The relatively high
toxic effects of TC seen as low IC-values for reproduction, are not reflected in those for the
PGRs. This is remarkable, because the effect on individual reproduction seems to be
buffered by the population. Especially an estimated IC50-value for PGR 10 times higher
than its counterpart for reproduction, is worth mentioning. These findings might question
the ecological relevance of endpoints such as reproduction and growth, tested in ISO tests
(Forbes & Calow, 2002; Höss & Williams, 2009). Nonetheless, it can be considered a good
thing that the population might survive a greater toxic stress than could be expected from
ISO test derived results. Although, one should be careful in reflecting conclusions drawn
from experiments conducted in the lab to the natural world. As seen in figure 6,
concentrations of 3mg/L and higher, result in a 3 day delay of two events (in reference to
the control), both in reaching the highest peaks as in reaching the carrying capacity. This
may seem insignificant at first, but given the fact that C. elegansʼ generation time is
approximately 2,5 days, a whole generation is lost. After an ecological catastrophe, this
delay may hamper recovery of the system to a great extent.
Furthermore it has to be considered that the concentration during the test was never
consistent. Even though it was thrived to keep the concentration within certain limits by
adding fresh medium every 3 days, this could not counteract degradation of TC.
Tetracyclines tend to degrade easily under intense light (photolysis: Oka et al., 1989, Chen
et al., 2011) and therefore the wells were kept in the dark at a constant temperature.
However, the plates were inevitable exposed to light every 3 days (subsampling). Even
under complete darkness tetracyclines degrade (Mitscher, 1978; Halling-Sørensen et al.,
2002), leaving a lot of degradation products to interact with both nematodes and bacteria.
Given the runtime of the prolonged experiment, it is highly plausible that concentrations of
31
certain compounds at the end were very different from the start of the experiment. Not
knowing the consistence of the medium at given time points, it is hard to make any
conclusion strong. Furthermore, by disturbing the medium every 3 days and thereby
liquifying it, chances are that sorption of TC to the walls of the well was unintentionally
stimulated. Additionally the competition for space in a limited volume (1mL), might have
brought about extra stress and bias the results.
Taking all this into account, it is most likely that the effects of TC on C. elegansʼ PGR
were underestimated, but the continuous supply (addition of TC every 3 days) may reflect
an antibioticʼs pseudo-persistency in the environment (Koschorreck & de Kecht, 2008).
Therefore this experimental setup may be considered ecologically relevant.
The growth rate of C. elegans N2 (Bristol, UK) was estimated by Muschiol et al.
(2009) to be 1.375, which is slightly higher than the estimations of rʼ in this study (1.24 ±
0.075). In a different lab estimates of the population growth rate were around 1.5
(Anderson et al., 2011). Given the variance between labs, probably due to overaged
stocks, the estimated growth rate of this study can be accepted. Because of its close
approximation, the experimental conditions in the control are likely to produce good
estimates of PGR.
Recent studies on Daphnia magna Straus showed a similar negative effect of TC on
the population growth rate (Kim et al., 2012). However, the effect concentrations were
much lower. 0.1mg/L effected the PGR by 30% and concentration up to 5mg/L reduced the
growth rate by 60%. Kim et al. (2012) described a slight increase in body mass, correlated
with the loss of PGR. Even though this correlation was not checked in this study and given
the experimental differences, it can be carefully suggested that C.elegansʼ PGR is less
sensitive to TC than the PGR of D. magna.
32
Food attraction
Although a trend towards uncontaminated food was seen in figure 9, the statistical
analysis showed no significant differences in any of the experimental setups. The
distribution of the nematodes seemed to have occurred at random. This conclusion is
strengthened by the fact that an apparently high number was attracted to the K-medium.
Since attraction towards contaminated food (Kenney et al., 2006) and nematodes feeding
behaviour after exposure to contaminants have been described (Dhawan et al., 1999;
Dhawan et al., 2000; Anderson et al., 2001; Boyd et al., 2003; Anderson et al., 2004), it is
imaginable that the seemingly randomness is a result of a bad experimental setup. Even
though nematodes possess an arsenal of sensors (Lee, 2002), it can be that the presence
of certain chemical compounds limits their abilities to locate food (Riga, 2004; Donkin &
Williams, 2000). In those cases an unambiguous experimental setup is the key to drawing
conclusions. Unfortunately, the high number of cavities (8) in the food attraction test plates
(figure 1) and the limited number of replicas (N=8) may not have been a proper
experimental design. It is plausible that nematodes attracted to a food source of their
choice, fell by accident in the wrong cavity. Furthermore, given the tension forces between
liquid K-medium and solid agar, it might have been hard for the nematodes to escape
these forces. Although a different experimental setup may yield better results, the
experiment is not considered to be a total failure. If TC-contaminated manure enters the
environment of a nematode population, it is likely to be dispersed in a very heterogeneous
manner on micro scale. In this context the TC present may affect the nematodes ability to
track food, to date this has not been proven. Given the fact that the nematodes survived
for 15 days, with only contaminated food at their disposal in the PGR-experiments, it is
likely that C. elegans feeds even upon TC-contaminated E. coli.
33
C. elegans among others
It has been shown that Tetracycline affects non-target organisms to some extend.
Wilson et al., (2004) tested the effects of 4 antibiotics (tetracycline, TC; chlortetracycline,
CTC; oxytetracycline, OTC and doxycycline, DC, mixed in different concentrations) on the
structure, function and interactions of the plankton community. Wilson and his team
recorded a significant reduction in abundance of phytoplankton, even at weighted
concentrations as low as 0.218 μM (24.9µg/L TC, 25.1µg/L OTC, 29.4µg/L, CTC and
24.9µg/L DC). However the communities recovered after the treatment was ended. A more
significant decrease was recorded when the concentrations were higher than 0.662 and
2.29 μM (72.0 µg/L TC, 79.9 µg/L OTC, 75.9 µg/L CTC, 76.1µg/L DC and 260 µg/L TC,
296 µg/L OTC, 208µg/L CTC, 281µg/L DC, respectively). The effects were strong in all
tested groups, but more expressed in the cyanobacteria, cryptophyta and dinophyta
groups. The effect concentrations are relatively low in comparison with those found in this
study. Plankton seems to be affected in a range of concentrations one magnitude lower
than C. elegans, but like C. elegans, plankton may not be at risk, given the relatively low
concentrations found in surface waters today (table 1).
Araujo and McNair (2007) described a reduced lifespan and reproduction among
rotifers. Lifespan as such was not used as an endpoint in this study, but reproduction was.
IC10-values were found between 26.10-28.34mg/L (B. calyciflorus) and 28.89-29.85mg/L
(B. plicatilis) and IC50-values were estimated between 78.81 - 79.56mg/L and 119.41 -
121.47 mg/L, respectively. Given the differences in experimental setups, comparisons
should be made carefully, but both IC-values are much higher than those found for C.
elegans in both test media (table 2). This may indicate a higher tolerance of rotifers for TC-
contamination in comparison with C. elegans. Moreover, rotifers are unlikely to be affected,
because the concentrations discovered so far in their habitats are significantly lower than
the described IC-values.
34
The reproduction of D. magna has been tested by Wollenberger et al. (2000) and
effect concentrations (EC) were calculated. EC10 (29.4 mg/L) and EC50 (44.8 mg/L) are
higher than those found in table 2 for both test media. But again it has to be stressed that
different experimental setups yield results that are hard to compare. Remarkable is the
relatively small difference between EC10 and EC50, which is approximately 15mg/L.
Although both ICs are lower for C. elegans in both experimental setups, the comparative
differences between IC10 and IC50 are bigger. This indicates that D. magna tolerates
relatively high dosages of TC, but once a certain concentration has been reached, its
reproduction is affected strongly. C. elegans is less robust and its reproduction is inhibited
gradually with increasing concentrations. However as discussed above, Kim et al. (2012)
described an earlier effect on D. magnaʼs PGR. The growth rate decreased up to 30% for
concentrations of 0.1 mg/L and up to 60% for 5.0 mg/L TC. In comparison, C. elegansʼ
growth rates are affected less severely. These interesting differences between individual
reproduction and population growth reflect different ecological strategies and make it even
harder to assess possible toxic effect of TC in different environments. It is plausible that D.
magna tolerates TC better than C. elegans if contact with the antibiotic is limited in time.
Nonetheless, populations of D. magna may find it harder to survive if TC is (pseudo)-
persistent in the environment, even at low concentrations.
Although dissimilar inhibition concentrations were found between different media and
endpoints, none of them seemed to be in the range of concentrations found in the
environment today. The apparently high tolerance of C. elegans, may indicate that
accidental leaches of TC in the environment cause little harm. Nevertheless one should
keep in mind that ecosystems are much more complicated than single-species tests,
because of the complex interactions that exist between different organisms. Consequently
the addition of indirect and direct effects of chemical compounds may lead to a far greater
effect. Only through integration of different disciplines and more in depth research, a
35
proper risk assessment can be made. This study is just one of the many pieces of the
puzzle, and in itself it has little meaning.
36
Acknowledgement
First and foremost I want to thank Dr. Nicole Spann for her guidance and support
throughout the entire process. Without her help I would not have been able to complete
this thesis successfully. I express my gratitude to my promotor Prof. Dr. Walter
Traunspurger for his expertise and advise. Special thanks to Mrs. Stefanie Gehner for
providing me with all the means to run my experiments and her assistance in the lab. I
thank Dr. Kai Ristau and Sebastian Weber for helping me with the experimental setup. In
extension, all members of the Department of Animal Ecology (University of Bielefeld),
made me feel welcome and created a nice working atmosphere. Let it be clear that without
the financial support from the European Master of Science in Nematology (EUMAINE)
programme, this work would not have been possible. I express my gratitude for providing
me with this wonderful opportunity. I want to thank all the interesting people I have met
along the way, especially those who provided me with new scientific insights. I extend my
thanks to all professors and people working hard in and around the nematology labs at
Ghent University and the University of Jaén, in particular Mrs. Nicole Smol and Prof. Dr. R.
Peña-Santiago. Last, but certainly not least, I want to thank my parents and girlfriend for
supporting me in all my endeavours.
37
References
AIELLO, S.E. (1998). The Merck veterinary manual, 8th ed. Merck, White-house Station, NJ
ANDERSON, G.L., BOYD, W.A. & WILLIAMS, P.L. (2001). Assessment of sublethal endpoints
for toxicity testing with the nematode Caenorhabditis elegans. Environmental
Toxicology and Chemistry 20(4), 833-838.
ANDERSON, G.L., COLE, R.D. & WILLIAMS, P.L. (2004). Assessing behavioral toxicity with
Caenorhabditis elegans. Environmental toxicology and chemistry 23, 1235-1240
ANDERSON, J.L., ALBERGOTTI, L., ELLEBRACHT, B., HUEY, R.B. & PHILLIPS, P.C. (2011). Does
thermoregulatory behavior maximize reproductive fitness of natural isolates of
Caenorhabditis elegans? BMC evolutionary biology 11,157
ARAUJO, A., & MCNAIR, J. N. (2007). Individual- and population-level effects of antibiotics on
the rotifers, Brachionus calyciflorus and B. plicatilis. Hydrobiologia, 593(1), 185-199.
ARIKAN, O.A., RICE, C., CODLING, E. (2008). Occurrence of antibiotics and hormones in a
major agricultural watershed. Desalination 226:121–133
BENBROOK, C. (2002). Antibiotic drug use in U.S. aquaculture. Institute for Agriculture and
Trade Policy (IATP) Report, 18 pp.
BOATMAN, M. (1998). Survey of antimicrobial usage in animal health in the European
Union. Boatman consulting report for FEDESA
BONGERS, T. & BONGERS, M. (1997). Functional diversity of nematodes. Applied Soil
Ecology 10, 239-251.
BOUSSU, K., KINDTS, C., VANDECASTEELE, C., VAN DER BRUGGEN, B. (2007). Surfactant
fouling of nanofiltrationmembranes: measurements and mechanisms.
Chemphyschem 8 (12), 1836 -1845.
38
BOXALL, A., KOLPIN, D., HOLLING-SORENSEN, B., & TOLLS, J. (2003). Are Veterinary
Medicines Causing Environmental Risks? Environmental Science & Technology (37)
286-294.
BOYD, W. A, & WILLIAMS, P. L. (2003). Comparison of the sensitivity of three nematode
species to copper and their utility in aquatic and soil toxicity tests. Environmental
toxicology and chemistry, 22(11), 2768-74.
BRAIN, R. A, WILSON, C. J., JOHNSON, D. J., SANDERSON, H., BESTARI, K. J., HANSON, M. L.,
SIBLEY, P. K., ET AL. (2005). Effects of a mixture of tetracyclines to Lemna gibba and
Myriophyllum sibiricum evaluated in aquatic microcosms. Environmental pollution
(Barking, Essex : 1987), 138(3), 425-42.
BRENNERS, S. (1974). The Genetics of Caenorhabditis elegans. Genetics, 77 (1), 71-94
BRINKE, M., HEININGER, P., & TRAUNSPURGER, W. (2011). A semi-fluid gellan gum medium
improves nematode toxicity testing. Ecotoxicology and Environmental Safety. 74,
1824-1831.
CASWELL, H. Demography meets ecotoxicology: untangling the population – level effects of
toxic substances. In: Newman MC, Jagoe CH, Ed. Ecotoxicology: A Hierarchical
Treatment. New York: CRC Press; 1996. pp. 255-292.
CHANDLER, G.T., COULL, B.C., DAVIS, J.C., (1994). Sediment- and aqueous-phase
fenvalerate effects on meiobenthos: implications for sediment quality criteria
development. Mar. Environ. Res. 37, 313–327.
CHEN, Y., LI, H., WANG, Z., TA, T. & HU, C. (2011). Photoproducts of tetracycline and
oxytetracycline involving self-sensitized oxidation in aqueous solutions: Effects of
Ca2+ and Mg2+. Journal of Environmental Sciences 23(10) 1634–1639.
CHOPRA, I., & ROBERTS, M. (2001). Tetracycline Antibiotics: Mode of Action, Applications,
Molecular Biology, and Epidemiology of Bacterial Resistance. Microbiology and
Molecular Biology Reviews, 65(2), 232–260.
39
CHRISTENSEN, A. M., INGERSLEV, F., & BAUN, A. (2006). Ecotoxicity of mixtures of antibiotics
used in aquacultures. Environmental toxicology and chemistry / SETAC, 25(8),
2208-15.
DHAWAN, R., DUSENBERY, D.B. & WILLIAMS, P.L. (1999). Comparison of lethality, reproduc-
tion, and behavior as toxicological endpoints in the nematode Caenorhabditis
elegans. Journal of toxicology and environmental health 58, 451-462.
DHAWAN, R., DUSENBERY, D.B. & WILLIAMS, P.L. (2000). A comparison of metal-induced
lethality and behavioral responses in the nematode Caenorhabditis elegans.
Environmental Toxicology and Chemistry 19(12), 3061-3067.
DOLLIVER, H., GUPTA, S., & NOLL, S. (2005). Antibiotic degradation during manure
composting. Journal of environmental quality, 37(3), 1245-53.
DONKIN, S.G., WILLIAMS, P.L. (1995). Influence of developmental stage, salts and food
presence on various endpoints using Caenorhabditis elegans for aquatic toxicity
testing. Environ. Toxicol. Chem. 14, 2139–2147
DONKIN, S.G. & WILLIAMS, P.L. (2000). Neurotoxicity: Toxic responses of the nervous
system. In: Williams, P.L., James, R.C., and Roberts, S.M. (Eds.) Principles of
Toxicology: Environmental and Industrial Application, 2nd Edition. New York: John
Wiley & Sons, Inc., pp. 145-155.
FELLOWS, I. (2012). Deducer: Deducer. R package version 0.6-3.. http://cran.r-project.org/
web/packages/Deducer/index.html
FORBES, V.E. & CALOW, P. (2002). Population growth rate as a basic for ecological risk
assessment of toxic chemicals. Philosophical Transactions of the Royal Society B:
Biological Sciences 357, 1299-1306.
HALLING-SØRENSEN, B., SENGELØV, G., & TJØRNELUND, J. (2002). Toxicity of tetracyclines
and tetracycline degradation products to environmentally relevant bacteria, including
40
selected tetracycline-resistant bacteria. Archives of environmental contamination and
toxicology, 42(3), 263-71.
HAMSCHER, G., SCZESNY, S., HÖPER, H., & NAU, H. (2002). Determination of persistent
tetracycline residues in soil fertilized with liquid manure by high performance liquid
chromatography with electrospray ionization tandem mass spectrometry. Analytical
Chemisty (74) 509–1518.
HAMSCHER G., PAWELZICK H.T., HÖPER H., NAU H. (2005). Different behaviour of
tetracyclines and sulfonamides in sandy soils after repeated fertilization with liquid
manure. Environ Toxicol Chem 24, 861–868.
HOTHORN, T., BRETZ, F. & WESTFALL, P. (2008). Simultaneous Inference
in General Parametric Models. Biometrical Journal 50(3), 346--363.
HÖSS, S., HAITZER, M., TRAUNSPURGER, W. & STEINBERG, C.E.W. (1999). Growth and fertility
of Caenorhabditis elegans (Nematoda) in unpolluted freshwater sediments:
Response to particle size distribution and organic content. Environmental Toxicology
and Chemistry 18, 2921-2925.
HÖSS, S., HENSCHEL, T., HAITZER, M., TRAUNSPURGER, W., STEINBERG, C.E.W. (2001).
Toxicity of cadmium to Caenorhabditis elegans (Nematoda) in whole sediment and
pore water—the ambiguous role of organic matter. Environ. Toxicol. Chem. 20, 2794–
2801
HÖSS, S., JÄNSCH, S., MOSER, T., JUNKER, T., & RÖMBKE, J. (2009). Assessing the toxicity of
contaminated soils using the nematode Caenorhabditis elegans as test organism.
Ecotoxicology and environmental safety, 72(7), 1811-8.
HÖSS, S. & WILLIAMS, P.L. (2009). Ecotoxicity Testing with Nematodes. In: Wilson, M.J. &
Kakouli-Duarte, T. (Eds). Nematodes as environmental indicators. Oxfordshire, UK,
CABI Publishing, pp. 208-244.
41
HU, X., ZHOU, Q. & LUO Y. (2010). Occurrence and source analysis of typical veterinary
antibiotics in manure, soil, vegetables and groundwater from organic vegetable
bases, northern China. Environ Pollut. 158, 2992–8.
ISO (2010) Water quality - Determination of the toxic effect of sediment and soil samples
on growth, fertility and reproduction of Caenorhabditis elegans (Nematode). ISO
10872:2010 (E). International Organization for Standardization, Geneva
JONES, O.A.H., VOULVOULIS, N., LESTER J.N. (2001). Human pharmaceuticals in the aquatic
environment: a review. Environ Technol 22, 1383-1394.
KHANNA, N., CRESSMAN, C.P., TATARA, C.P. & WILLIAMS, P.L. (1997). Tolerance of the
nematode Caenorhabditis elegans to pH, salinity, and hardness in aquatic media.
Arch. Environ. Contam. Toxicol. 32, 110–114
KIM, H.Y., LEE, M.J., YU, S.H. & KIM, S.D. (2012). The individual and population effects of
tetracycline on Daphnia magna in multigenerational exposure. Ecotoxicology 21,
993–1002.
KIM, S., EICHHORN, P., JENSEN, J.N., WEBBER, A.S., AGA, D. (2005). Removal of antibiotics in
wastewater: effect of hydraulic and solid retention times on the fate of tetracycline in
the activated sludge process. Environmental Science Technology 39 (15), 5816 -
5823.
KIM, S.-C., & CARLSON, K. (2007). Quantification of human and veterinary antibiotics in
water and sediment using SPE/LC/MS/MS. Analytical and Bioanalytical Chemistry
387, 1301–1315.
KENNEY S.J., ANDERSON G.L, WILLIAMS P.L., MILLNER P.D. & BEUCHAT L.R. (2006). Migration
of Caenorhabditis elegans to manure and manure compost and potential for transport
of Salmonella newport to fruits and vegetables. International Journal of Food
Microbiology 106, 61–68
42
KOLPIN, D.W., FURLONG, E.T., MEYER, M.T., THURMAN, E.M., ZAUGG, S.D., BARBER, L.B., &
BUXTON, H.T. (2002). Pharmaceuticals, hormones, and other organic wastewater
contaminants in U.S. streams, 1999- 2000: A national reconnaissance:
Environmental Science & Technology 36 (6), 1202-1211.
KOSCHORRECK, J. & DE KECHT, J. Environmental risk assessment of pharmaceuticals in the
EU. In: Kümmerer K Ed. Pharmaceuticals in the Environment: Sources, Fate, Effects
and Risks. 2nd edition. Berlin: Springer; 2008. pp. 289-310
KÜMMERER, K. (2009). Antibiotics in the environment: a review – Part I. Chemosphere 75,
417-434
LANGHAMMER, J.P. & BUENING-PFAUE, H. (1989). Bewertung von Arzneistoff- Rückständen
aus der Gülle in Boden. Lebensmittelchem. Wissenschaft Umwelt 10, 14–20.
LE-MINH, N., KHANA, S.J., DREWESA, J.E. & STUETZ, R.M. (2010). Fate of antibiotics during
municipal water recycling treatment processes. Water research (44) 4295 - 4323
LEE, D.L. (2002). Behaviour. In: Lee, D.L. (Ed.) The biology of nematodes. London, Taylor
& Lewis, pp. 369-387.
LOCATELLI, M.A.F., JARDIM, W.F. & SODRE, F.F. (2011). Determination of Antibiotics in
Brazilian Surface Waters Using Liquid Chromatography–Electrospray Tandem Mass
Spectrometry. Arch Environ Contam Toxicol 60, 385–393.
LOTKA, A.J. (1913). A natural population norm. J Washington Acad Sci 3, 241–248.
Luo, Y., Xu, L., Rysz, M., Wang, Y., Zhang, H. & Alvarez, P.J. (2011). Occurrence and
transport of tetracycline, sulfonamide, quinolone, and macrolide antibiotics in the
Haihe River Basin, China. Environ Sci Technol 45, 1827–1833.
MA, H., BERTSCH, P.M., GLENN, T.C., KABENGI, N.J. & WILLIAMS, P.L. (2009). Toxicity of
manufactured Zinc oxide nanoparticles in the Nematode Caenorhabditis elegans.
Environmental Toxicology and Chemistry 28(6), 1324-1330.
43
MEYER, B., PAILLER, J.-Y., GUIGNARD, C., HOFFMAN, L. & KREIN, A. (2011). Concentrations of
dissolved herbicides and pharmaceuticals in a small river in Luxembourg. Environ
Monit Assess 180, 127–146.
MIAO, X.S., BISHAY, F., CHEN, M., METCALFE, C.D. (2004). Occurrence of antimicrobials in
the final effluents of wastewater treatment plants in Canada. Environmental Science
Technology 38 (13), 3533 - 3541.
MITSCHER, L.A. (1978). The chemistry of the tetracycline antibiotics. Medicinal research
series 9, Marcel Dekker, New York
MUSCHIOL, D. & TRAUNSPURGER, W. (2007). Life cycle and calculation of the intrinsic rate of
natural increase of two bacterivorous nematodes, Panagrolaimus sp. and
Poikilolaimus sp. from chemoautotrophic Movile Cave, Romania. Nematology, 9(2),
271-284.
MUSCHIOL, D., SCHROEDER, F., & TRAUNSPURGER, W. (2009). Life cycle and population
growth rate of Caenorhabditis elegans studied by a new method. BMC ecology 9,
14.
OFFERMANN, K., MATTHAI, A. & AHLF, W. (2009). Assessing the importance of dietborne
cadmium and particle characteristics on bioavailability and bioaccumulation in the
nematode Caenorhabditis elegans. Environ. Toxicol. Chem. 28, 1149–1158.
OHBA, K. & ISHIBASHI, N. (1984). A Nematode, Caenorhabditis elegans, as Test Organism
for Nematicide Evaluation. J. Pesticide Sci. 9, 91-96.
OKA, H., IKAI Y., KAWAMURA N., YAMADA M., HARADA K., ITO S. & SUZUKI M. (1989).
Photodecomposition products of tetracyclines in aqueous solution. J Agric Food
Chem 37, 226–231.
POMATI, F., NETTING, A. G., CALAMARI, D., & NEILAN, B. A. (2004). Effects of erythromycin,
tetracycline and ibuprofen on the growth of Synechocystis sp. and Lemna minor.
Aquatic toxicology 67(4), 387-96.
44
QIAO, M., CHEN, W., SU, J., ZHANG, B. & ZHANG, C. (2012). Fate of tetracyclines in swine
manure of three selected swine farms in China. Journal of Environmental Sciences
24(6) 1047–1052
QUINLAN, E. L., NIETCH, C. T., BLOCKSOM, K., LAZORCHAK, J. M., BATT, A. L., GRIFFITHS, R., &
KLEMM, D. J. (2011). Temporal dynamics of periphyton exposed to tetracycline in
stream mesocosms. Environmental science & technology, 45(24)
R CORE TEAM (2012). R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL
http://www.R-project.org/.
RIGA, E. (2004). Orientation Behavior. In: Gaugler, R. & Bilgrami, A.L. (Eds.). Nematode
Behaviour. Oxfordshire, UK, CABI Publishing, pp. 63-90.
RITZ, C. & STREIBIG, J. C. (2005). Bioassay Analysis using R. J. Statist. Software, (12) 5.
SCHRAP, S.M., RIJS, G.B.J., BEEK, M.A., MAASKANT, J.F.N., STAEB, J., STROOMBERG, G.,
TIESNITSCH, J. (2003). Human and veterinary pharmaceuticals in Dutch surface water
and wastewater In Dutch: Humane an veterinaire geneesmiddelen in Nederlands
oppervlaktewater en afvalwater. RIZA rapport.
SIBLEY, P. K., & HANSON, M. L. (2011). Ecological Impacts of Organic Chemicals on
Freshwater Ecosystems. Ecotoxicology, 138-164.
TRAUNSPURGER, W., HAITZER, M. HÖSS, S., BEIER, S., AHLF, W.& STEINBERG, C. (1997).
Ecotoxicological Assessment of Aquatic Sediments With Caenorhabditis elegans
(Nematoda)—a Method for Testing Liquid Medium and Whole-Sediment Samples.
Environmental Toxicology and Chemistry, 16 (2) pp. 245–250.
VAN DEN BRINK, P., SIBLEY P.K., RATTE H.T., ET AL. Extrapolation of effect measures across
levels of biological organization in ecological risk assessment. In: Solomon KR, Brock
TCM, et al. Eds. Extrapolation Practice for Ecotoxicological Effect Characterization of
Chemicals. New York: CRC Press; 2008. pp. 105-134.
45
WATANABE, N., BERGAMASCHI B.A., LOFTIN K.A., MEYER M.T., HARTER T. (2010). Use and
environmental occurrence of antibiotics in freestall dairy farms with manured forage
fields. Environ Sci Technol 44(17), 6591–600.
WATKINSON, A.J., MURBY, E.J. & KOLPIN, D.W. (2009). The occurrence of antibiotics in an
urban watershed: from wastewater to drinking water. Sci Total Environ 407, 2711–
2723.
WEI, R., GE, F., HUANGA, S., CHENA, M. & WANG, R. (2011). Occurrence of veterinary
antibiotics in animal wastewater and surface water around farms in Jiangsu Province,
China. Chemosphere 82, 1408–1414.
WILLIAMS, P.L. & DUSENBERY, D.B. (1990). Aquatic toxicity testing using the nematode,
Caenorhabditis elegans. Environmental Toxicology and Chemistry 9, 1285-1290.
WILLIAMS, R.T. Human health pharmaceuticals in the environment – an introduction. In:
Williams RT, Ed. Human Pharmaceuticals: Assessing the Impacts on Aquatic
Ecosystems. Pensacola: SETAC Press; 2005. pp. 1-45.
WILSON, C. J., BRAIN, R. A, SANDERSON, H., JOHNSON, D. J., BESTARI, K. T., SIBLEY, P. K., &
SOLOMON, K. R. (2004). Structural and functional responses of plankton to a mixture
of four tetracyclines in aquatic microcosms. Environmental science & technology, 38
(23), 6430-9.
WOLLENBERGER, L., HALLING-SØRENSEN, B., & KUSK, K. O. (2000). Acute and chronic
toxicity of veterinary antibiotics to Daphnia magna. Chemosphere, 40(7), 723-30.
YANG, J.-F., YING, G.-G., ZHAO, J.-L., TAO, R., SU, H.-C. & CHEN, F. (2010). Simultaneous
determination of four classes of antibiotics in sediments of the Pearl Rivers using
RRLC–MS/MS. Science of the Total Environment 408, 3424–3432.
ZHANG, X., ZHANG, D., ZHANG, H., LUO, Z. & YAN, C. (2012). Occurrence, distribution, and
seasonal variation of estrogenic compounds and antibiotic residues in Jiulongjiang
River, South China. Environ Sci Pollut Res 19, 1392–1404.
46
ZHOU., Y., GENG, J. & WANG, X. (2011). Tetracycline adsorption on kaolinite: pH, metal
cations and humic acid effects. Ecotoxicology 20, 1141–1147
ZILLES, J., SHIMADA, T., JINDAL, A., ROBERT, M. & RASKIN, L. (2005). Presence of macro-
lidelincosamidestreptogramin B and tetracycline antimicrobials in swine waste
treatment processes and amended soil. Water Environ Res 77, 57–62.
47
DeclarationI, Vangheel Matthew, hereby confirm that I have independently composed this Master thesis and that no other than the indicated aid and sources have been used. This work has not been presented to any other examination board.
Bielefeld, 20th August, 2012
Vangheel Matthew
48