Rapid genetic erosion in pollutant

advertisement
Environmental Pollution 157 (2009) 881–886
Contents lists available at ScienceDirect
Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol
Rapid genetic erosion in pollutant-exposed
experimental chironomid populations
Carsten Nowak a, *, Christian Vogt b, Markus Pfenninger a,
Klaus Schwenk a, Jörg Oehlmann b, Bruno Streit a, Matthias Oetken b
a
Abteilung Ökologie und Evolution, Institut für Ökologie, Evolution und Diversität, J. W. Goethe-Universität Frankfurt am Main,
Siesmayerstrasse 70, 60054 Frankfurt am Main, Germany
Abteilung Aquatische Ökotoxikologie, Institut für Ökologie, Evolution und Diversität, J. W. Goethe-Universität Frankfurt am Main,
Siesmayerstrasse 70, 60054 Frankfurt am Main, Germany
b
Chronic TBT exposure reduces allelic variation at five variable microsatellite loci in experimental populations of Chironomus riparius.
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 14 May 2008
Received in revised form
4 November 2008
Accepted 5 November 2008
Few studies have evaluated how effectively environmental contamination may reduce genetic diversity
of a population. Here, we chose a laboratory approach in order to test if tributyltin (TBT) exposure at
environmentally relevant concentrations leads to reduced genetic variation in the midge Chironomus
riparius. Two TBT-exposed and two unexposed experimental populations were reared simultaneously in
the laboratory for 12 generations. We recorded several life-history traits in each generation and monitored genetic variation over time using five variable microsatellite markers. TBT-exposed strains showed
increased larval mortality (treatments: 43.8%; controls: 27.8%), slightly reduced reproductive output, and
delayed larval development. Reduction of genetic variation was strongest and only significant in the TBTexposed strains (treatments: 45.9%, controls: 24.4% of initial heterozygosity) after 12 generations. Our
findings document that chemical pollution may lead to a rapid decrease in genetic diversity, which has
important implications for conservation strategies and ecological management in polluted
environments.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Allelic variation
Chemical pollution
Evolutionary ecotoxicology
Chironomidae
Microsatellites
1. Introduction
Numerous studies have documented negative effects of environmental pollutants on life-history traits of ecotoxicological
model species in the laboratory (Walker et al., 2001). Although
laboratory test systems provide a powerful tool for toxicity
assessment, they mostly consider acute or single-generation effects
on life-history responses towards chemical exposure (Vogt et al.,
2007a). In contrast to single-generation experiments, natural
populations are chronically exposed to pollutants over multiple
generations and numerous factors influence the long-term
response to chemical exposure in the field. For instance, populations may adapt to polluted environments (Gillis et al., 2002;
* Corresponding author. Present address: Department of Limnology and
Conservation, Senckenberg Research Institute, Clamecystrasse 12, 63571 Gelnhausen, Germany. Tel.: þ49 6051 61954 3122; fax: þ49 6051 61954 3118.
E-mail addresses: [email protected], [email protected] (C. Nowak), [email protected]
bio.uni-frankfurt.de (C. Vogt), [email protected] (M. Pfenninger),
[email protected] (K. Schwenk), [email protected]
(J. Oehlmann), [email protected] (B. Streit), [email protected]
(M. Oetken).
0269-7491/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.envpol.2008.11.005
Vogt et al., 2007a). Furthermore, chemical exposure has frequently
been predicted to reduce genetic variation in natural populations
(Bickham et al., 2000; Van Straalen and Timmermans, 2002). This
effect, named genetic erosion, has been reported in field surveys for
various animal taxa, like fish (Murdoch and Hebert, 1994), crustaceans (Krane et al., 1999) and marine gastropods (Kim et al., 2003).
However, most of these studies have provided little evidence for
erosion of genetic variation in response to pollution stress (Brown
et al., 2007; Chen et al., 2003; Nadig et al., 1998). Thus, the question
if genetic erosion threatens populations in polluted habitats
remains unanswered (Van Straalen and Timmermans, 2002).
Patterns of genetic variation within and among natural populations
are influenced by various historical and recent factors, like past and
present selection processes, migration and chance events (Avise,
2004). Most field studies with regard to the effects of pollution on
genetic variation, however, do not consider those natural processes
adequately (Bickham et al., 2000; Whitehead et al., 2003), and
consequently fail to distinguish natural ‘‘background noise’’ from
anthropogenic impacts. In order to exclude factors like gene flow
among populations and to control fluctuations in population size,
we chose an experimental approach here. The experiment was
designed to investigate if pollutants at environmentally relevant
882
C. Nowak et al. / Environmental Pollution 157 (2009) 881–886
concentrations reduce genetic variation in populations of an ecotoxicological model organism, the midge Chironomus riparius
Meigen (Diptera: Chironomidae). C. riparius is widely distributed in
small streams, ditches, ponds and puddles throughout the
holarctic. In organically polluted and muddy habitats, the species
may reach high densities and it plays a key role in freshwater
ecosystems for nutrient cycling and energy flux (Armitage et al.,
1995). As detritus feeders living in the upper sediment layers,
chironomids come into contact with sediment-associated toxicants
like heavy metals, PCBs or organotin compounds (Vogt et al.,
2007b).
In this study, C. riparius larvae were exposed to sediments
contaminated with the highly effective biocide tributyltin (TBT).
We used a TBT concentration (measured time-weighted mean
concentration ¼ 8.93 mg as Sn kg1 sediment dry weight ¼ dw),
which is environmentally relevant (Wade et al., 2004) and has
previously been shown to affect life-history traits of C. riparius
(Vogt et al., 2007b). The exposure was maintained for 12 generations and genetic variation as well as life-history traits were
monitored over time.
In particular, we focused on two main questions:
(i) Does TBT exposure at an environmentally relevant concentration cause a measurable reduction of genetic variation at
neutral loci within only a few generations?
(ii) Can we observe changes in life-history responses to TBT
exposure over time?
2. Materials and methods
2.1. Experimental procedure and TBT exposure
For the experiments we used a laboratory strain of C. riparius originating from
a cross-breeding of 11 laboratory strains in 2004 (GENþ; described more in detail in
Nowak et al., 2007a). Four experimental populations (¼groups) were separated from
the GENþ source strain and reared in the laboratory for twelve consecutive generations. Two groups were kept under control conditions (solvent control ¼ SC), while
larvae of the two remaining groups were chronically exposed to sediment-bound
TBT as described below. To start the experiment, 42 egg masses were taken from the
source strain, and hatched larvae were carefully mixed in a Petri dish. Three hundred
and fifty larvae were randomly chosen as start generation for each group. Larvae
were transferred in seven 2 l glass vessels per treatment containing 1 l reconstituted
water and 100 g artificial sediment (pure quartz sand; for details see Vogt et al.,
2007b). Fifty larvae were dispersed into each replicate. For the TBT treatments,
sediment was spiked with nominally 160 mg as Sn kg1 sediment dw (390 mg as
TBT kg1 sediment dw) using pure ethanol as solvent. After larval development,
emerged imagines were transferred into 10 l reproduction containers
(30 20 20 cm) containing a square water-filled box (11.5 11.5 5.5 cm,
400 ml) for egg laying. Egg masses were collected from the reproduction container
and placed into 24-microwell plates for hatching. After hatching, 350 L1 larvae were
taken from the two days with the highest egg mass production and mixed in
a vessel. Subsequently, larvae were randomly distributed over the replicates to start
the next generation. The following life-history parameters were measured in each
generation: larval mortality, mean emergence time (EmT50), produced egg masses
per female and hatchability of egg masses. Furthermore, daily population growth
rate (PGR; calculated according to Vogt et al., 2007b) was used as an integrating
endpoint in the experiments.
For the genetic analyses, 200 larvae of each experimental group were taken from
the same clutches as those used for the multi-generation study and reared in two
separate 10 l aquaria. Rearing conditions were exactly the same in these aquaria
compared to the test vessels, including the sediment contamination in the TBT
exposed treatments. All resulting L4 larvae were frozen in liquid nitrogen and stored
at 80 C for further analyses.
The experiment was conducted at 20 C (0.5 C) under an 8:16 h dark/light
cycle following the OECD guideline 218 for Chironomus sediment tests (OECD, 2004).
For further details regarding the experimental procedure see (Vogt et al., 2007a). The
experiments were stopped after the 12th generation.
2.2. Acute tests
In order to detect alterations in TBT susceptibility of C. riparius larvae, acute tests
were performed with L1 larvae of all four groups in generations 4, 8 and 12. Larvae
were exposed to nominal TBT concentrations of 2, 6.3, 20, 63 and 200 mg as Sn l1 via
water, including a solvent control (10 ml EtOH l1, equals 0.17 mM). After 24 h,
survival was determined using a stereo microscope.
2.3. Chemical sediment analyses
TBT analyses of the sediments were performed by GALAB Laboratories, Geesthacht, Germany, according to DIN guideline 19744 (2003). Approximately 2 g
freeze-dried sediment were taken from test vessels of the first experimental
generation. The extracted organotin compounds were derivatized with 4% NaBEt4 in
H2O, concentrated into cyclohexane by liquid–liquid extraction, and subsequently
analyzed with gas chromatography tandem atomic emission detector GC-AED (GC
6890 Series, Agilent Technologies; AED: HP G2350A, Hewlett Packard, Waldbronn,
Germany). The PACS-2 reference material (National Research Council Canada,
Ottawa, ON) was analyzed to check extraction efficiency, showing recoveries of 95–
100%. The detection limit was 1 mg kg1 sediment dw. TBT and its metabolites were
analyzed for controls and for TBT treatments after solvent evaporation 1 and 72 h,
and 14 days of test duration. Based on the TBT loss after solvent evaporation and the
measured TBT concentrations after 14 days the time-weighted average concentration was calculated according to OECD guideline 211 (OECD, 1998). For the acute
tests, no TBT analyses were performed, and hence the calculated LC50 values are
based on nominal concentrations.
2.4. Genetic analyses
Genetic diversity was estimated using five variable microsatellite markers
described in Nowak et al. (2006). Frozen L4 larvae were homogenized in 1.5 ml
EppendorfÒ tubes containing 700 ml standard CTAB buffer and 4 ml 20 mM
proteinase K. Tissue was digested in a water bath at 63 C for 1 h followed by
standard chloroform/isoamylalcohol 24:1 treatment. DNA was precipitated in 1 ml
isopropanol 100% for 1 h at -20 C and washed twice with 300 ml ethanol 70%. DNA
was diluted in 600 ml water and stored at 4 C for further treatment. Polymerase
chain reaction was performed on a Tetrad thermocycler (InvitrogenÒ) with cycling
conditions as follows: 1 min 92 C, 1 min 55 C, 1 min 72 C, repeated 36 times.
Reactions were performed in a total volume of 10 ml containing 0.1 ml of 25 mM
dNTPs, 0.5 ml 50 mM MgCl2, 1 ml 10 reaction buffer, 0.2 ml of each primer (10 pmol)
and 0.1 ml 5 U Taq DNA polymerase (Invitrogen). PCR products were loaded on a 1.4%
agarose gel at 120 V and quantified after staining with ethidium bromide under UV
light.
For fragment analysis, 1 ml DNA solution containing w1 ng of microsatellite
fragment was mixed with 5 ml dextran blue solution and 1 ml of internal size marker
lambda DNA. The solution was loaded on a 9% polyacrylamide gel using an ALF
sequencer (PharmaciaÒ). Gels were run for 450 min at 1000 V. Fragment lengths
were scored manually using the ALFWIN 3.1 software.
Genetic diversity was checked for 24 individuals per group and generation (¼96
individuals/generation). Every second generation was included for genetic analysis
following the starting generation (¼generations 0, 2, 4, 6, 8, 10 and 12).
2.5. Statistical analysis
Prior to statistical analysis, all endpoints were tested for normal distribution
with the Kolmogorov–Smirnov normality test. Nested two-way ANOVA with
repeated measurements was used in order to reveal the influence of TBT on variation
in larval mortality and mean emergence time (EmT50) and to test for significant
differences in life-history response between groups and generations. We chose an
ANOVA design in which experimental groups (SC I, SC II, TBT I, TBT II) were nested
into treatments (SC and TBT). The analyses were performed using STATISTICA 7.1
(StatSoft Inc., USA). We tested for the presence of significant linear trends in all lifehistory traits (mortality, EmT50, produced fertile egg ropes per female, population
growth rate) and population genetic parameters (see below) with the Mann–Kendall
test, and calculated Sen’s slope estimate with the Excel-application MAKESENS 1.0,
provided by the Finnish Meteorological Institute. Population genetic parameters
(expected and observed heterozygosity [HE and HO], number of alleles per locus [NA],
number of polymorphic loci [NP] and test for deviations from Hardy–Weinberg
equilibrium) based on the microsatellite data were calculated using GenAlEx 6
software (Peakall and Smouse, 2006).
3. Results
3.1. TBT concentrations in the sediment
No TBT or other organotin compounds were detected in the
control sediment. After solvent evaporation a TBT concentration of
21.7 mg as Sn kg1 sediment dw was determined in the TBT-spiked
sediment, and after 1 h of test duration (containing animals, water
and food) a TBT concentration of 9.8 mg as Sn kg1 sediment dw was
measured. A similar TBT concentration was also measured after
72 h with 8.9 mg as Sn kg1 sediment dw. Prior to larval emergence
C. Nowak et al. / Environmental Pollution 157 (2009) 881–886
(after 14 days of test duration) the TBT concentration decreased to
2.8 mg as Sn kg1 sediment dw. Two TBT metabolites, monobutyltin
(MBT) and dibutyltin (DBT), were detected in low microgram per
kilogram concentrations after 72 h (MBT ¼ 4.3 mg as Sn kg1 and
DBT ¼ 1.6 mg as Sn kg1 sediment dw) and 14 days (MBT ¼ 8.1 mg as
Sn kg1 and DBT ¼ 0.8 mg as Sn kg1 sediment dw). During the
whole test the time weighted mean TBT concentration was 8.93 mg
Sn kg1 sediment dw.
3.2. Life-history traits
In the multi-generation study TBT had a significant effect on
both larval mortality and mean emergence time (EmT50; two way
ANOVA, both p < 0.001; Table 1). Mortalities ranged from 14.2%
(TBT II in generation 12) to 100% (TBT I and II in generation 10) in
the TBT treatments and from 6.3% (SC I in generation 1) to 80.3% (SC
I in generation 6) in the controls (Fig. 1A). Over all generations,
mean mortality was higher in the TBT-treated groups than in the
control groups (43.8% in the treatments compared to 27.8% in the
controls). However, only in four generations both TBT-treated
groups showed higher mortalities than the two respective controls
(generations 1, 5, 9 and 10). Because of the extinction of both TBTexposed groups in generation 10, the study was maintained using
backup groups that had been isolated from both TBT-exposed
groups in the previous generation and being maintained under
control conditions in generation 10.
EmT50 values were higher in both TBT groups compared to the
controls in all generations except of generations 4 and 5 (Fig 1B).
We found considerable variation in mortality and mean emergence
time between groups and generations. In addition, a significant
interaction between treatments and generations as well as
between groups and generations was detected (all p < 0.001;
nested two-way ANOVA, Table 1).
Number of fertile clutches produced per female varied between
0.40 (TBT I, generation 7) and 1.15 (TBT I, generation 3) in the TBT
groups and 0.59 (SC I, generation 7) and 1.4 (SC I, generation 11) in
the controls (Fig. 1C). Over all generations, TBT-exposed females
produced less egg masses than unexposed females (TBT ¼ 0.83
compared to SC ¼ 0.99; Fig. 1C).
TBT-treated groups showed reduced population growth rates
(PGR) compared to the respective control groups in all but two
generations (3 and 4; Fig. 1D). Mean PGR summarized for all
generations was 1.22 for TBT (without generation 10) and 1.26 in
the controls. No experimental group showed a significant time
trend in larval mortality or fertility (Table 2). EmT50 values showed
a clear tendency to increase over time. However, a significant time
trend was revealed only for SC II (p < 0.05). Population growth rates
tended to decrease with increasing generations in all four groups.
Table 1
Influence of the factors treatment, experimental group nested in treatment,
generation and interaction between these factors on mortality and mean emergence
time (EmT50) of Chironomus riparius.
Trait
Cause
df
SS
Mortality
Treatment
Experimental group (treatment)
Generation
Treatment generation
Experimental group generation
1
2
10
10
20
9482.7
9803.1
69185.4
16375.4
18183.5
EmT50
Treatment
Experimental group (treatment)
Generation
Treatment generation
Experimental group generation
1
2
10
10
20
536.8
21.1
1318.4
19.1
8.2
F
p
71.960
37.196
37.719
8.928
4.957
529.9
10.4
96.1
13.9
6.0
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
Shown are degrees of freedom (df), sums of squares (SS), F-values (F) and significance levels (p).
883
Significant effects were observed for SC II (p < 0.05) and TBT I
(p < 0.01).
3.3. Genetic variation
All loci exhibited allelic variation in the source strain prior to the
start of the experiment (generation 0, data not shown). Allelic
richness slightly decreased in the control groups over time.
However, a significant negative trend was only revealed for the
expected heterozygosity in SC II (Table 2). Values for all genetic
parameters recorded in this study decreased more strongly over
time in the presence of TBT (Fig. 2, Table 2).
Number of alleles decreased from 3.0 at the beginning of the
experiment to 2.0 after 12 generations in both stressed treatments.
Observed heterozygosity (HO) decreased significantly in both TBTexposed groups (Table 2), while no significant trend was observed
in the controls. Over time, expected heterozygosity (HE) decreased
significantly only in one exposed group. However, rates of genetic
impoverishment were similar in both TBT-affected groups and HE
values decreased from 0.39 at the beginning to 0.21 (TBT I) and 0.20
(TBT II) in the last generation. In contrast, reduction of heterozygosity was clearly slower in the absence of sediment contamination
(HE ¼ 0.31 for SC I and 0.28 for SC II in generation 12). Under TBT
exposure, all genetic parameters showed a significant (p < 0.05)
decrease or at least a trend (p < 0.1) for decreasing levels of genetic
variation at the microsatellite loci (Table 2). At the end of the
experiments, one control group had become monomorphic at one
locus. In the presence of TBT both groups lost variation at three of
the five loci (data not shown).
3.4. Acute tests
LC50 values based on TBT acute tests ranged from 7.33 mg Sn l1
(SEM 1.05, generation 4) to 35.7 mg Sn l1 (SEM 1.01, generation
8) in the pre-exposed groups (TBT I, TBT II). Values in the control
treatments (SC I, SC II) ranged from 11.1 mg Sn l1 (SEM 1.05,
generation 4) to 25.4 mg Sn l1 (SEM 1.09, generation 12; Table 3).
No significant differences were observed between pre-exposed and
control groups in any test (t-test).
4. Discussion
The results presented in this study provide clear evidence that
chemical exposure may lead to genetic impoverishment in populations. Although significant stressor effects were observed
throughout the study, the impact of TBT on larval mortality, mean
emergence time and reproduction was moderate throughout this
investigation when compared with the controls. However, both
number of alleles and levels of observed and expected heterozygosity decreased over time in the TBT-exposed groups. The unexposed groups showed only a slight decrease of genetic variation and
no significant negative trends were observed for most genetic
parameters.
Chemical exposure may lead to decreased genetic variation in
populations in different ways. First, selection against less stresstolerant genotypes may increase frequencies of alleles (potentially
leading to fixation) that provide a selective advantage under
stressed conditions (Bickham et al., 2000). This kind of selectionbased reduction of genetic variation, however, will solely affect loci
which are under direct selection or physically linked to those.
Neutral genetic variation, which can be visualized by microsatellite
markers, is not directly affected by selection. However, contaminant exposure may lead to increased mortality and reduced fertility
and thus reduce the number of individuals that contribute to the
next generation (expressed as effective population size, NE). The
extent of allelic variation in a population is positively correlated
884
C. Nowak et al. / Environmental Pollution 157 (2009) 881–886
SC I
SC II
A
TBT I
TBT II
B
30
mortality [%]
100
EmT50 [d]
75
50
25
20
25
15
0
1
2
3
4
5
6
7
8
1
9 10 11 12
2
3
4
generation
6
7
8
9 10 11 12
generation
D
population growth rate [d-1]
C
1.5
clutches/female
5
1.0
0.5
0.0
1
2
3
4
5
6
7
8
9 10 11 12
generation
1.4
1.3
1.2
1.1
1
2
3
4
5
6
7
8
9 10 11 12
generation
Fig. 1. Life-history of two TBT exposed (TBT) and two unexposed (SC) Chironomus riparius populations over 12 generations. Shown are mean mortality (in %, SEM; A), Mean
emergence time (EmT50, in days, SEM; B), mean number of fertile egg clutches produced per emerged female (clutches/female; C) and the population growth rate (in days1, SD;
D).
with its effective population size, because random genetic drift is
more effective in small populations (Frankham, 1996). Thus,
chemical stress can reduce neutral genetic variation due to nonselective reduction of NE.
In this study we obtained no evidence for selection processes, as
no significant time trend in any life-history trait was observed, and
L1 larval tolerance towards TBT did not significantly change over
time. However, TBT led to increased mortalities and a reduced
Table 2
Time series analyses of four genetic parameters and five life-history traits of Chironomus riparius
Trait
n
Repl.
SC
p Level
Genetic
variation
Life-history
traits
þ
HO
7
HE
7
NA
7
NP
7
Mortality
12
EmT50
12
Clutches/female
12
Clutch size
12
PGR
12
I
II
I
I
II
I
II
I
II
I
II
I
II
I
II
I
II
*
þ
þ
*
þ
*
TBT
Slope
0.005
0.022
0.009
0.013
0.133
0.100
0.200
0.000
2.285
0.964
0.183
0.273
0.011
0.018
10.18
6.125
0.003
0.007
p Level
Slope
*
*
þ
*
þ
*
þ
þ
0.025
0.028
0.036
0.036
0.200
0.150
0.500
0.333
þ
þ
**
1.946
0.048
0.483
0.290
0.024
0.013
7.825
1.375
0.010
0.004
p < 0.1; *p < 0.05; **p < 0.01.
HO, observed heterozygosity; HE, expected heterozygosity, NA, mean number of
alleles per locus; NP, number of polymorphic loci; EmT50, mean emergence time, n,
number of data points (¼generations) considered in the analysis.
number of fertile clutches in most generations. Consequently,
a lower number of individuals contributed to the next generation.
Our observation of decreasing genetic diversity is therefore most
likely the result of reduced effective population sizes in the exposed
groups, although absolute population sizes were held constant at
the beginning of each generation. In a previous study using
a similar experimental design, but a lower TBT concentration
(measured time-weighted average: 4.46 Sn kg1 sediment dw),
presence of the biocide did not significantly lower fitness traits in
most generations, and had thus no clear effects on NE (Vogt et al.,
2007a). Consequently, no reduction in genetic variation due to
pollution was observed in this study, although some indications for
selection processes were found.
The fact that there was no gene flow between the experimental
strains contributed to the fast reduction of genetic variation. Gene
flow among populations restores genetic variation very effectively
(Slatkin, 1987). In our experiment we simulated completely isolated
populations, and thus the obtained results are not easily transferable to natural conditions. However, loss of genetic variability is
rather a problem of small and isolated than of large populations.
Rare and endangered species usually persist in small and isolated
relict populations (Amos and Balmford, 2001). As we showed,
chemical pollution might decrease genetic variation in small and
isolated populations, which are the main targets of conservation
efforts. Hence, our findings provide a functional link between
ecotoxicology and conservation genetics.
Besides the investigation of genetic variation under TBT stress,
high variation in life-history traits was observed not only between
treatments, but also among groups and generations throughout the
multi-generational study. For instance, no stressor effect on larval
mortality was apparent in generations 3 and 8. In contrast, TBT
severely reduced larval survival in other generations (e.g. 5, 9 and
10). These findings document the limited significance of single
generation surveys in standard ecotoxicological studies. Although
C. Nowak et al. / Environmental Pollution 157 (2009) 881–886
SC I
A
SC II
TBT I
B
3.0
2.5
TBT II
3.0
NA
NA
2.5
2.0
2.0
1.5
1.5
0
2
4
6
8
10
0
12
2
4
6
8
10
12
generation
generation
C
D
0.4
0.4
0.3
0.3
HE
HE
885
0.2
0.2
0.1
0.1
0
2
4
6
8
10
12
0
2
generation
4
6
8
10
12
generation
Fig. 2. Genetic variation (A, B ¼ mean number of alleles per locus [NA]; C, D ¼ expected heterozygosity [HE]) of two TBT exposed (TBT) and two unexposed (SC) Chironomus riparius
populations over 12 generations. Lines show combined slope estimates (see Table 2 for values and significance levels).
conditions were kept constant throughout the study and equal
concentrations of TBT were applied in all generations, a comparison
of different generations (e.g. eight and nine) leads to completely
different conclusions concerning the effects of the chosen TBT
concentration on Chironomus larval survival.
There are several ways to explain the large variation in lifehistory traits between experimental groups and generations, like
heterogeneous contamination of the test sediment or slight variability of test parameters. As our data document, however, variation
in the genetic composition between generations is likely to
contribute to the observed life-history variation. For instance,
population growth rates of all four groups investigated tended to
decrease over time (significant for SC II and TBT I) and this reduction was overall slightly stronger in the exposed groups. This
decrease can most likely be explained by reduced fitness due to loss
of genetic variation in the respective groups. Twelve generations
might not be sufficient, though, to detect a significant trend here.
It is likely that genetic effects contributed to extinction of TBTexposed groups after generation 10. Larval development was
already delayed in the two former generations 8 and 9. A further
delay in development might have led to a complete extinction
because of rapidly declining water quality, which can be observed
in most test vessels after approximately 30 days of test duration
(data not shown). High mortalities have also been found in the
previous multi-generational experiment by Vogt et al. (2007a),
Table 3
Acute tests with Chironomus riparius L1 larvae from generations 4, 8, and 12 of the
multi-generation study.
Treatment
Experimental group
Gen. 4
Gen. 8
Gen. 12
SC
I
II
11.1 1.05
17.2 1.11
21.9 1.04
21.5 1.04
25.4 1.09
17.1 1.02
TBT
I
II
7.33 1.05
20.4 1.08
35.7 1.01
23.4 1.07
22.4 1.07
20.7 4.92
Shown are calculated LC50 values (in mg TBT-Sn l1 SEM).
probably indicating the presence of survival-limiting factors that
can hardly be controlled in the laboratory. However, we decided to
continue experiments after the breakdown using backup groups of
the respective populations taken from the previous generation.
These groups were kept for one generation under unstressed
conditions, which should not severely affect experimental
outcomes. However, even when considering only the first 10
generations, significant reduction of genetic variation was still
obvious for most parameters (data not shown).
The reduction of genetic variation, however, does not provide
sufficient explanation for the extinction event in generation 10,
because the backup strains showed considerable survival rates in
the last two generations of the experiment. These backup strains
derive from the same egg masses as the strains used in the lifehistory experiments. It would thus be reasonable to expect similar
reduced survival in the backup strains when exposed to TBT in
generations 11 and 12. Continuing the experiments for two additional generations after the breakdown did prove that other
explanations, such as seasonal effects or variations in exposure
conditions, likely contributed to the observed high variation in life
history among generations.
5. Conclusions
We showed that toxic substances have the potential to reduce
genetic variation before severe population declines due to high
mortality rates or low fertility occurs. While life-history responses to
contamination can remain approximately constant over time,
resulting in stable population sizes, genetic variation may constantly
decrease further over time. The consequences of genetic impoverishment on fitness and tolerance towards chemical exposure and
other stress factors may only be visible over longer time periods.
Therefore, loss of genetic variation has to be considered as an additional threat to populations subjected to chronic environmental
pollution (Theodorakis, 2001; Van Straalen and Timmermans, 2002).
It is also important to stress the relevance of genetic variation in
886
C. Nowak et al. / Environmental Pollution 157 (2009) 881–886
populations for the survival in contaminated environments. Nowak
et al. (2007b), for instance, have shown that C. riparius groups with
low degrees of allelic variation display a significantly lowered
cadmium tolerance compared to more diverse groups. Genetic
impoverishment might therefore have severe long-term consequences for populations in human-affected environments when
stressors change over time. Even if selection may allow for local
adaptation, its effectiveness relies on the presence of a sufficient
degree of genetic variation (Frankham et al., 2002). The loss of
intraspecific diversity might therefore lead to an increased susceptibility to environmental stress, accompanied by a lowered ability to
adapt to stressful environmental conditions. In order to ensure longterm viability of populations in polluted environments, it is therefore
crucial to maintain sufficiently high population sizes and to ensure
gene flow between populations, keeping genetic variation high and
preventing genetic erosion. In addition, rare and isolated populations
should be kept from pollution more rigidly than large populations,
because they might lose variation more rapidly, have lowered
potential for genetic adaptation and might display enhanced stress
susceptibility, as documented for C. riparius in laboratory tests
(Nowak et al., 2007b, 2008; Vogt et al., 2007c).
Investigating the influence of pollution on patterns of genetic
variation represents an important future challenge for ecotoxicologists and conservation geneticists (Belfiore, 2001). Only
comprehensive field studies, which consider the high complexity
of factors impacting the genetic structure of natural populations,
will show the relevance of genetic erosion in polluted environments. Linking ecological and population genetic processes will
eventually allow us to predict the long-term fate of natural
populations and to assess the consequences of human made
changes to natural communities.
Acknowledgments
We appreciate the kind assistance of several undergraduate
students working on this project in the Departments ‘‘Aquatic
Ecotoxicology’’ and ‘‘Ecology & Evolution’’. This work was financially supported by the ‘‘Programm Lebensgrundlage Umwelt und
ihre Sicherung (BWPLUS)’’ of the federal state Baden-Württemberg
(contract number BWR 22018).
References
Amos, W., Balmford, A., 2001. When does conservation genetics matter? Heredity
23, 257–265.
Armitage, P., Cranston, P.S., Pinder, C.V., 1995. The Chironomidae: Biology and
Ecology of Non-Biting Midges. Chapman & Hall, London.
Avise, J., 2004. Molecular Markers, Natural History, and Evolution. Sinauer Associates, Sunderland.
Belfiore, N.M., 2001. Effects of contaminants on genetic patterns in aquatic organisms: a review. Mutation Research-Reviews in Mutation Research 489, 97–122.
Bickham, J.W., Sandhu, S., Hebert, P.D.N., Chikhi, L., Athwal, R., 2000. Effects of
chemical contaminants on genetic diversity in natural populations: implications for biomonitoring and ecotoxicology. Mutation Research-Reviews in
Mutation Research 463, 33–51.
Brown, J.W., Van Coeverden de Groot, P.J., Birt, T.P., Seutin, G., Boag, P.T., Friesen, V.L.,
2007. Appraisal of the consequences of the DDT-induced bottleneck on the level
and geographic distribution of neutral genetic variation in Canadian peregrine
falcons, Falco peregrinus. Molecular Ecology 16, 327–343.
Chen, X.Y., Li, N., Shen, L., Li, Y.Y., 2003. Genetic structure along a gaseous organic
pollution gradient: a case study with Poa annua L. Environmental Pollution 124,
449–455.
Frankham, R., 1996. Relationship of genetic variation to population size in wildlife.
Conservation Biology 10, 1500–1508.
Frankham, R., Ballou, J.D., Briscoe, D.A., 2002. Introduction to Conservation Genetics.
Cambridge University Press, Cambridge.
Gillis, P., Diener, L.C., Reynoldson, T.B., Dixon, D.G., 2002. Cadmium-induced
production of a metallothionein like protein in Tubifex tubifex (Oligochaeta) and
Chironomus riparius (Diptera): correlation with reproduction and growth.
Environmental Toxicology and Chemistry 21, 1836–1844.
Kim, S.J., Rodriguez-Lanetty, M., Suh, J.H., Song, J.I., 2003. Emergent effects of heavy
metal pollution at a population level: Littorina brevicula as a study case. Marine
Pollution Bulletin 46, 74–80.
Krane, D.E., Sternberg, D.C., Burton, G.A., 1999. Randomly amplified polymorphic
DNA profile-based measures of genetic diversity in crayfish correlated with
environmental impacts. Environmental Toxicology and Chemistry 18, 504–508.
Murdoch, M.H., Hebert, P.D.M., 1994. Mitochondrial L-DNA diversity of brown
bullhead from contaminated and relatively pristine sites in the great-lakes.
Environmental Toxicology and Chemistry 13, 1281–1289.
Nadig, S.G., Lee, K.L., Adams, S.M., 1998. Evaluating alterations of genetic diversity in
sunfish populations exposed to contaminants using RAPD assay. Aquatic Toxicology 43, 163–178.
Nowak, C., Hankeln, T., Schmidt, E.R., Schwenk, K., 2006. Development and localization of microsatellite markers for the sibling species Chironomus riparius and
Chironomus piger (Diptera: Chironomidae). Molecular Ecology Notes 6, 915–917.
Nowak, C., Vogt, C., Barateiro, J., Schwenk, K., 2007a. Genetic impoverishment in
laboratory cultures of the test organism Chironomus riparius. Environmental
Toxicology and Chemistry 26, 118–122.
Nowak, C., Jost, D., Vogt, C., Oetken, M., Schwenk, K., Oehlmann, J., 2007b. Effects of
inbreeding and reduced genetic variation on tolerance to cadmium stress in the
midge Chironomus riparius. Aquatic Toxicology 85, 278–284.
Nowak, C., Czeikowitz, A., Vogt, C., Oetken, M., Streit, B., Schwenk, K., 2008. Variation in tolerance to cadmium exposure among genetically characterized
laboratory populations of the midge Chironomus riparius (Diptera: Chironomidae). Chemosphere 71, 1950–1956.
OECD, 1998. Guideline for testing of chemicals no. 211. Daphnia magna reproduction test, adopted September 1998. Organization for Economic Development
and Cooperation, Paris.
OECD, 2004. Sediment-water chironomid toxicity test using spiked sediment. OECD
guidelines for the testing of chemicals. (Original guideline 218, adopted 13th
April 2004). Organization for Economic Development and Cooperation, Paris.
Peakall, R., Smouse, P.E., 2006. GenAlEx 6: Genetic analysis in Excel. Population
genetic software for teaching and research. Molecular Ecology Notes 6, 288–
295.
Slatkin, M., 1987. Gene flow and the geographic structure of natural populations.
Science 236, 787–792.
Theodorakis, C.W., 2001. Integration of genotoxic and population genetic endpoints
in biomonitoring and risk assessment. Ecotoxicology 10, 245–256.
Van Straalen, N.M., Timmermans, M., 2002. Genetic variation in toxicant-stressed
populations: An evaluation of the ‘‘genetic erosion’’ hypothesis. Human and
Ecological Risk Assessment 8, 983–1002.
Vogt, C., Nowak, C., Barateiro Diogo, J., Oetken, M., Schwenk, K., Oehlmann, J., 2007a.
Multi-generation studies with Chironomus riparius - Effects of low tributyltin
concentrations on life-history parameters and genetic diversity. Chemosphere
67, 2192–2200.
Vogt, C., Belz, D., Galluba, S., Nowak, C., Oetken, M., Oehlmann, J., 2007b. Effects of
cadmium and tributyltin on development and reproduction of the non-biting
midge Chironomus riparius (Diptera) - baseline experiments for future multigeneration studies. Journal of Environmental Science and Health-Part A 42, 1–9.
Vogt, C., Pupp, A., Nowak, C., Jagodzinski, L.S., Baumann, J., Jost, D., Oetken, M.,
Oehlmann, J., 2007c. Interaction between genetic diversity and temperature
stress on life-cycle parameters and genetic variability of Chironomus riparius
populations. Climate Research 33, 207–214.
Wade, T.L., Sweet, S.T., Quinn, J.G., Cairns, R.W., King, J.W., 2004. Tributyltin in
environmental samples from the former Derecktor Shipyard, Coddington Cove,
Newport RI. Environmental Pollution 129, 315–320.
Walker, C.H., Hopkin, S.P., Sibly, R.M., Peakall, D.B., 2001. Principles of Ecotoxicology.
Taylor & Francis, London.
Whitehead, A., Anderson, S.L., Kuivila, K.M., Roach, J.L., May, B., 2003. Genetic
variation among interconnected populations of Catostomus occidentalis: implications for distinguishing impacts of contaminants from biogeographical
structuring. Molecular Ecology 12, 2817–2833.
Download