Environmental Pollution 207 (2015) 176e182

Contents lists available at ScienceDirect

Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol

Multigenerational contaminant exposures produce non-monotonic,

transgenerational responses in Daphnia magna
David A. Kimberly a, b, *, Christopher J. Salice a, c
a
Department of Environmental Toxicology, The Institute of Environmental and Human Health, Texas Tech University, 1207 Gilbert Drive, Lubbock, TX 79416,
USA
b
Department of Biology, Westminster College, 442 Meldrum Science Center, Salt Lake City, UT 84105, USA
c
Environmental Science and Studies Program, Towson University, 8000 York Rd., Towson, MD 21252, USA

a r t i c l e i n f o a b s t r a c t

Article history: Generally, ecotoxicologists rely on short-term tests that assume populations to be static. Conversely,
Received 4 February 2015 natural populations may be exposed to the same stressors for many generations, which can alter
Received in revised form tolerance to the same (or other) stressors. The objective of this study was to improve our understanding
27 August 2015
of how multigenerational stressors alter life history traits and stressor tolerance. After continuously
Accepted 6 September 2015
Available online 15 September 2015
exposing Daphnia magna to cadmium for 120 days, we assessed life history traits and conducted a
challenge at higher temperature and cadmium concentrations. Predictably, individuals exposed to cad-
mium showed an overall decrease in reproductive output compared to controls. Interestingly, control
Keywords:
Non-monotonic response
D. magna were the most cadmium tolerant to novel cadmium, followed by those exposed to high cad-
Transgenerational effects mium. Our data suggest that long-term exposure to cadmium alter tolerance traits in a non-monotonic
Tolerance way. Because we observed effects after one-generation removal from cadmium, transgenerational effects
Cadmium may be possible as a result of multigenerational exposure.
Daphnia magna © 2015 Elsevier Ltd. All rights reserved.
Population growth rate
Life history traits
Aquatic ecotoxicology

1. Introduction
A primary goal in ecotoxicology is to understand and ultimately
predict effects of chemical stress on ecological systems (Walker,
2006). Because the vast majority of ecotoxicity studies are single-
species toxicity tests, populations are the level of biological orga-
nization most immediately tractable to experimentation. A key
challenge then lies in relating effects observed in standardized
toxicity tests to changes in population size or dynamics (Forbes
et al., 2008; Barnthouse et al., 2007). The reliance of ecotoxicol-
ogy on relatively short-duration toxicity studies implicitly assumes
that traits of individuals comprising an exposed population are
static. That is, the population response we observe at one time is
the same response we'll predict at another time (given the same
conditions). This is inherently problematic because not only might
populations change within ecological timeframes (e.g., few
* Corresponding author. Department of Biology, Westminster College, Salt Lake
City, USA.
E-mail address: dkimberly@westminstercollege.edu (D.A. Kimberly).
generations; Carrol et al., 2007; Hairston et al., 2005; Salice et al.,
2010), but for many contaminants (e.g., persistent chemicals), ex-
posures are likely to continue beyond the duration of many toxicity
tests. Surprisingly few studies have explicitly compared the results
and inferences derived from standardized representations of indi-
vidual effect to actual effects at the population level for exposures
to persistent chemicals that can occur in nature over one entire or
even multiple generations.
The changes populations experience as a result of longer dura-
tion contaminant exposure are disparate and may come in the form
of decreased sensitivity or tolerance (e.g., Clements, 1999; Klerks
and Weis, 1987) or increased sensitivity (e.g., Kashian et al., 2007;
Koivisto et al., 1992). Both outcomes have important management
implications and neither may be indicated by short duration
toxicity tests. Indeed, within the same study on multigenerational
exposure to cadmium, persistence at higher concentrations and
extinction at intermediate concentrations has been observed in
snails (Salice et al., 2009) and chironomids (Postma et al., 1995).
With regard to increased tolerance, Klerks and Weis (1987) pro-
posed two categories of tolerance that aquatic organisms may
achieve after exposure to environmental contaminants. The first is

http://dx.doi.org/10.1016/j.envpol.2015.09.020
0269-7491/© 2015 Elsevier Ltd. All rights reserved.
 D.A. Kimberly, C.J. Salice / Environmental Pollution 207 (2015) 176e182
tolerance due to acclimation, where an organism experiences a
physiological change that leads to higher tolerance. Another is
tolerance due to adaptation, where organisms achieve genetically
based tolerance that can be inherited by offspring. Increased
sensitivity resulting from continuous exposure has received less
explicit experimental attention but presumably results from an
accumulation of stress that may include contaminant bio-
accumulation and/or transgenerational decrements in fitness (e.g.,
Kimberly and Salice, 2014).
In the freshwater cladoceran Daphnia magna, toxicity of metals
has been widely investigated (Guan and Wang, 2006; De
Schamphelaere and Janssen, 2002). Many single- and several
multi-generational studies have been conducted on the effects of
metals on life history traits and changes in tolerance to novel
stressors in D. magna (Biesinger and Christensen, 1972; Lopes et al.,
2004; Stuhlbacher et al., 1992; Münzinger and Monicelli, 1992;
Guan and Wang, 2006). LeBlanc (1982) demonstrated that the
progeny of D. magna exposed to copper had higher survival rates
than those progeny from unexposed parents. Muyssen and Janssen
(2004) described an increase in acute tolerance after D. magna were
exposed to cadmium for three generations. Further, chromium
tolerant D. magna showed increased tolerance for novel chromium
and nickel exposures when compared to stock populations. How-
ever, those same chromium tolerant D. magna where shown to be
significantly more sensitive to novel zinc exposures (Munzinger,
1994; Münzinger and Monicelli, 1992). Additionally, D. magna
acclimated to cadmium showed increased sensitivity, as seen in an
overall reduction in population growth rate, when exposed to
cadmium at novel temperature and nutritional states (Heugens
et al., 2006). These studies suggest that a change in tolerance/
sensitivity to novel stress can result from long term exposure to
contaminants, specifically trace metals. However, there is still a
paucity of data on this topic and emerging patterns are not uni-
versally conclusive.
The goal of this study was to evaluate the effects of continuous
multigenerational cadmium exposure on life history traits and
novel stressor tolerance. More specifically, we were interested in
whether there were persistent, transgenerational effects of cad-
mium even after the long-term exposure had ceased. We hypoth-
esize that D. magna would show decreased fitness (estimated as
population growth rate) and decreased cadmium tolerance when
exposed to higher cadmium concentrations for multiple genera-
tions. Additionally, we hypothesized that adverse effects of long-
term exposure would persist and manifest as increased sensitivity
to higher cadmium and novel temperature exposures. To test these
hypotheses, we employed a longitudinal, continuous multigener-
ational design where 21-day life history assessments were con-
ducted after 120 days of exposure to cadmium. A novel cadmium
and temperature challenge was also conducted on populations
cultured for one generation in cadmium free conditions.
2. Methods
2.1. Study species
D. magna used in this study were derived from a stock colony
that was initiated from few individuals and has been cultured at the
Institute of Environmental and Human Health since 2007. Stock
populations had been reared at 25
C and 12:12 light:dark cycle in
moderately hard freshwater (60 mg/L CaSO4, 60 mg/L MgSO4, 4 mg/L
KCl, 98 mg/L NaHCO3; hereafter called lab water) and fed ad libitum
quantities of Scenedesmus acutus. Abundance of the laboratory
stock populations was not rigorously controlled although care was
taken to avoid unusually high densities. Experimental organisms
were obtained from laboratory stocks by separating adult D. magna
177
into individual beakers and removing neonates from the fourth
brood. This process was used twice because 4th brood neonates
were used to produce both the adults and juveniles that constituted
the initial longitudinal exposure populations. Adults for the study
were first taken as neonates from the 4th brood of adults from the
stock colony, housed together in conditions described above, and
allowed to grow to adulthood. Juveniles (<24-h old) from the 4th
brood of additional adults from the stock colony were obtained and
were added to experimental populations at the same time as adults
from the first collection of neonates.
2.2. Longitudinal exposure phase
In the longitudinal phase, D. magna were exposed to cadmium
(low or high) for 120 days or roughly 14 generations. This phase
consisted of three treatments and four replicates per treatment.
Treatments included a control (no added cadmium), 0.5 mg/L Cd
(low), and 2.0 mg/L Cd (high) (Fig. 1). These concentrations were
shown in pilot studies to elicit a range of sublethal responses and
no or limited mortality. Each replicate was maintained in an incu-
bator at 25
C and consisted of a plastic BPA-free container filled
with 750 mL of lab water. Each replicate was initiated with 16 ju-
veniles and 8 adults. Subsequently, life stages were defined by size
and observation of embryos in the brood chamber. Juveniles were
defined as newly hatched neonates <0.5 mm in length. Animals
were considered adult if they were greater than 4 mm or were
observed with embryos in the brood chamber. Rarely were
D. magna less than 4 mm in size observed with embryos in the
brood chamber. Replicates were fed daily with 2.4 mL of S. acutus
(3.0  107 cell/mL) and every other day with a 2.4 mL YCT (yeast,
cerophyll, and trout chow) mixture as described by USEPA (2002).
Algal density was verified weekly by hemocytometer and in all
cases, measured density was within 5% of target density. The vol-
ume of 2.4 mL corresponds to 0.1 mL diet per individual. As pop-
ulation abundance increased or decreased diet was modified after
each life stage observation and counting (every three days) to
maintain a consistent ratio of food to abundance. It should be noted
that no individuals from any life stage were removed from con-
tainers, except those used for life history assessments (described
below). Data on population abundance through time were collected
but are reported elsewhere (Kimberly and Salice, in preparation).
All cadmium (Cd2þ) test solutions were created by serial dilution
in lab water from a 100 mg/L stock solution made by dissolving
Fig. 1. Experimental design schematic. Schematic shows the exposure regime and
timeline. D. magna were exposed to one of three treatments (control, low, or high
cadmium) for 120 days. They were then moved to cadmium free conditions for one
generation. Offspring reared in the no-cadmium conditions were then subjected to a
novel cadmium and temperature challenge with survival taken over a 14-day period.
178 D.A. Kimberly, C.J. Salice / Environmental Pollution 207 (2015) 176e182

CdCl2 (cadmium chloride, 99.00 þ %, SigmaeAldrich, St. Louis, MO)

in 3% nitric acid (for stabilization) and lab water. Cadmium con-
centrations in natural waters are typically <1 mg/L, but can range up
to 10 mg/L (World Bank Group, 1999) and can reach as high as
>40 mg/L in heavily contaminated sites (Environmental Integrity
Project and Earthjustice, 2010). Therefore, the concentrations
used in this experiment represent a highly likely exposure scenario
in cadmium-contaminated environments. Exposures followed a
static-renewal design with 100% renewal of solutions every six days
following a full water change. After and before water changes,
water samples were collected for cadmium concentration verifi-
cation by atomic absorption spectrophotometry (Solaar-AA,
Thermo Scientific) using flame ionization. Internal standards were
used for determining cadmium concentrations. Cadmium concen-
trations in water samples taken before and after water changes
were not significantly different from each other. Measured cad-
mium concentrations were within 10% of nominal, thus we refer to
exposure concentrations using the nominal concentration values.

2.3. Life history assessments

Twenty-one-day chronic toxicity life cycle tests were performed

using <24-h D. magna juveniles from the longitudinal exposures.
The purpose of these tests was to determine, explicitly, how the
populations might have been altered by longitudinal exposure to
cadmium. Ten juveniles were randomly selected from among each
of the four replicates of each treatment (2 or 3 per replicate). Tests
were conducted in 30-mL plastic cups with one juvenile/cup. Each
cup was filled with 25 ml lab water and juveniles were fed 0.1 mL
S. acutus daily. Cups were randomly assigned to 1 of 30 positions in
an incubator at 25
C. In the life history assessments, daphnids were
exposed to the same cadmium concentrations as during the
multigenerational exposures. For example, juveniles exposed to
low cadmium longitudinally were exposed to low cadmium during
the life history assay. Endpoints of the assay included survival, time
to first reproduction, number of total offspring, and number of
broods produced during the assay. A cadmium free 21-day assay
was also conducted for every treatment (control conditions for all
animals). The goal of the cadmium-free assay was to test for
possible transgenerational effects of long-term cadmium exposure.
As an integrative measure of fitness, Population Growth Rate (l)
was calculated for each individual using an age-based matrix model
(Sibly and Hone, 2002). The model was a 21  21 matrix repre-
senting each day of the 21-day experimental generation and sur-
vival and reproduction data from the assays was incorporated. The
time step was a single day, and survival per day was entered on the
subdiagonal, whereas fecundity was entered on the first row.
Population growth rate was obtained as the dominant eigenvalue of
the resulting matrix model (Caswell, 2001).

2.4. Cadmium and temperature challenge

The purpose of the challenge experiment was to determine if

there was a signature of effect of the long-term, multi-generational
exposure to sublethal cadmium after a generation removed from
the toxic environment. As a way to assess whether there may have
been a trangenerational (or “carry-over”) effect on cadmium
tolerance, we exposed D. magna from all treatments to the same
high cadmium and high temperature challenge. At the end of the
120-day longitudinal exposure, all animals were transferred to lab
water only (no cadmium added; Fig. 1). After one week in these
control conditions, five adults from each treatment were housed
individually to collect juveniles for use in a life cycle assay
(described above). Twenty juveniles from the third brood were
collected from among the five F0 adults. These F1 juveniles were
Fig. 2. Life history traits of D. magna in cadmium exposed and cadmium free condi-
tions. Twenty-one day life cycle assays were conducted at day 120 in cadmium
exposed and cadmium free conditions. Life history traits observed included time to
first reproduction (A), total offspring produced (B), and number of broods (C).
then exposed individually to a cadmium and temperature chal-
lenge. Juvenile D. magna were housed in 30 mL plastic cups with
25 mL lab water and fed 0.1 mL S. acutus. All juveniles, regardless of
their previous cadmium exposure (Low or high Cd), were exposed
to the higher cadmium concentration of 6 mg/L at 25
C or 30
C.
This produced a total of 6 treatments with 10 animals per treat-
ment. Survival over the next 14 days was observed and recorded.
 D.A. Kimberly, C.J. Salice / Environmental Pollution 207 (2015) 176e182
2.5. Statistical analysis
Results from life history assays at day 120 in cadmium free
conditions were compared to results by one-way ANOVA with
cadmium (longitudinal exposure concentration) as the factor.
Population growth rates of exposed and cadmium free treatments
at day 120 were also compared by one-way ANOVA. Survival ana-
lyses were performed on all mortality data using the SURVIVAL and
KMsurv (Venables and Ripley, 2002) packages for R. Survival ana-
lyses are time-to event regression methods that are well suited to
mortality data (Newman and Aplin, 1992). The logrank test (aka
ManteleCox test) was used to compare survival curves between
treatments, which computes observed and expected number of
events at each time point for each group. The expected values are
subtracted from observed values for each time point and then
summed across all time points. In the analysis, individuals that
survived the duration of the experiment were statistically
accounted for through censoring. Through the 14-day period, in-
dividual observations were censored from all treatments except the
30
C-L and the 25
C-L treatments, where complete mortality was
reached. Results were considered significant at the p  0.05 level
and R 2.12.2 statistical package was used for all analyses (R
Development Core Team, 2010).
3. Results
3.1. Life history assays at day 120
During the 21-day assay we observed the following endpoints:
mortality, time to first reproduction, total offspring, and number of
broods produced over the assay duration (Fig. 2). There was no
mortality in any treatments confirming that these concentrations
were sub-lethal. At day 120, time to reproduction was 14.5, 10.25,
and 8.75 days for high Cd, low Cd, and control, respectively
(Fig. 2A). Total offspring produced were 11, 43, and 52 neonates
over the 21-day duration for high Cd, low Cd, and control, respec-
tively (Fig. 2B). Lastly, the number of broods produced was 5.5, 9.2,
and 10.4 broods for high Cd, low Cd, and control, respectively
(Fig. 2C). For all three endpoints, high cadmium treatments were
Fig. 3. Population growth rate (l) D. magna in cadmium exposed and cadmium free
conditions. Life history traits observed at day 120 were used to understand population
change in cadmium exposed and cadmium free conditions. Cadmium exposed con-
ditions included control, low, and high concentrations.
179
Fig. 4. Survival of D. magna exposed to a novel cadmium concentration and temper-
ature. At day 120, cadmium exposed treatments were transferred to cadmium free
conditions for one generation. Offspring of adults in the cadmium free conditions were
subjected to a novel cadmium and temperature challenge. Survival was recorded over
14 days. In the figure, although every treatment is exposed to the novel cadmium, the
C, L, and H represent the longitudinal exposure while and 25 or 30 represents the
temperature at which they are being exposed to the novel concentration.
significantly different than both control and low cadmium treat-
ments (P ¼ 0.001, for all relationships; denoted by “*” in Fig. 2). To
understand potential carry over or transgenerational effects of
long-term cadmium exposure, 21-day assay results from non
cadmium-exposed D. magna at day 120 were compared to assay
results from cadmium-exposed D. magna (Fig. 2 e each panel).
There was no significant differences found between the two assays
(cadmium-exposed and non-exposed) for any of the three end-
points at any of the three treatment levels (F2,3 ¼ 78.91, P ¼ 0.767)
suggesting that there may be a carry-over effect of long term
exposure even in the absence of the stress.
When the D. magna life history data from the 21-day assays
were combined into an estimate of population growth rate (l),
overall there was a significant effect of treatment (l; Fig. 3;
F2,5 ¼ 10.9, P ¼ 0.022). Lambdas of treatments were compared
within cadmium-exposed and cadmium-free treatments, but also
between cadmium-exposed and cadmium-free treatments. Within
both cadmium-exposed and cadmium-free conditions, a strong
significant difference in l existed between high cadmium treat-
ments and both low cadmium and control treatments (P < 0.001).
Additionally, low cadmium and control treatments did not have
significantly different l (P ¼ 0.244) under either cadmium-exposed
or cadmium-free conditions. Similar to results for individual
reproductive endpoints, when l estimates were compared between
cadmium-exposed and cadmium-free treatments, they did not
significantly differ (P < 0.551).
3.2. Cadmium and temperature challenge
All D. magna in the challenge phase of the experiment were
exposed to 6.0 mg/L Cd. Challenge treatments, however, were
comprised of a combination of temperature (low or high) and
longitudinal cadmium exposure (control, low an high cadmium). To
distinguish the different treatments we used the numeric
180 D.A. Kimberly, C.J. Salice / Environmental Pollution 207 (2015) 176e182
temperature (25 or 30) followed by a letter indicating the longi-
tudinal cadmium exposure (C, L, or H). A treatment denoted as 25L,
for example, refers to D. magna that were exposed previously to low
cadmium (L) during the longitudinal phase but exposed to 6.0 mg/L
Cd at 25
C during the challenge. D. magna in the 30L treatment
were exposed to low cadmium longitudinally but then exposed to
6.0 mg/L Cd at 30
C during the challenge. Hence, the 30
C repre-
sents a temperature challenge concomitant with the high cad-
mium. Overall, there was a significant difference in survival by
challenge treatment (P < 0.001; Fig. 4). Pairwise comparisons also
indicated a number of significant differences between specific
treatments. For example, all D. magna that were exposed to cad-
mium during the longitudinal phase were significantly more sen-
sitive to the cadmium challenge than the control D. magna at both
25
C and 30
C (P < 0.001 for all relationships). The 25L D. magna
were significantly more sensitive to the cadmium challenge con-
centration than the 25H (P ¼ 0.007) but not than 30H treatments
(P ¼ 0.374). The 30L D. magna were similarly more sensitive to the
acute cadmium challenge compared to the 30H D. magna
(P ¼ 0.049). Lastly, survival in the 30L treatments did not differ
significantly from the 25L exposures and nor did the 30
C control
treatment (30C) compared to the 25
C control (25C). To simplify,
the most sensitive to the cadmium challenge were the D. magna
that were exposed to the low cadmium concentration longitudi-
nally, regardless of the temperature during the challenge. And,
generally, the higher temperature during the challenge led to
higher sensitivity when comparing D. magna from the same lon-
gitudinal treatment (e.g., 25L vs 25H).
4. Discussion
We investigated the impact of long-term, multigenerational
cadmium exposure on D. magna life history and stressor tolerance
traits. On the last day of the longitudinal exposure (day 120), and in
addition to the standard assay, we conducted a life history assay in
cadmium-free conditions to investigate whether a persistent,
carry-over effect of the long-term cadmium exposure existed.
Additionally, we were interested in the effects of long-term expo-
sure on novel stressor response traits. We tested this by subjecting
D. magna to a high cadmium and temperature challenge. Overall we
found that D. magna that were exposed to cadmium at sublethal
concentrations but continuously for multiple generations had
lower fitness and decreased stressor tolerance compared to
D. magna controls. The effects of cadmium, however, showed a mix
of monotonic (fitness) and non-monotonic (stressor tolerance)
patterns of response. Within the cadmium treatments, high cad-
mium treated D. magna displayed the longest time to reproduction,
the fewest total offspring produced, and the fewest broods pro-
duced but then displayed increased tolerance to cadmium
compared to D. magna from the low cadmium treatment.
Numerous studies have identified decreases in reproduction as a
result of contaminant exposure. Villarroel et al. (2000) observed
decreased offspring production in F1 D. magna as a result of
multigenerational pesticide exposure. Bervoets et al. (1996) and
Baldwin et al. (1995) both observed decreased reproduction in
second-generation offspring exposed to environmental contami-
nants. Similar to these studies, we identified decreased reproduc-
tion in D. magna as the primary adverse response to low and high
cadmium exposures based on life history assays. These results can
potentially be explained in the context of energy allocation.
Under stressful conditions organisms often utilize different
strategies that increase survival and reproductive success. Guan
and Wang (2006) suggested that tradeoffs exist between impor-
tant functions such as growth and reproduction in the presence of
metal stress and are likely a result of energy redistribution (Forbes
and Forbes, 1994). These tradeoffs can serve to increase tolerance to
a specific stressor while maintaining other functions at a basic level.
The underlying explanation may be that organisms devote energy
to ameliorate the effects of stressors and as a result have less energy
available for growth and reproduction. These concepts have been
formalized with the application of Dynamic Energy Budget Theory
to problems in ecotoxicology (e.g., Nisbet et al., 2000; Jager and
Zimmer 2012; Martin et al., 2013). Importantly, the theory would
suggest that when the original stress is reduced or eliminated,
animals would return to their original functional state. Thus, the
energy used to tolerate the stress could be used again for mainte-
nance, growth, or reproduction. In fact, Guan and Wang (2006)
discuss that after a single generational recovery from cadmium
exposure, D. magna achieved total or at least partial recovery.
Both our life history data and population growth rates are
contrary to Guan and Wang's (2006) findings and suggest that ef-
fects of the original stressor may continue (to some degree) after
the stressor has been removed. Operating under the hypothesis
that organisms should return to the original state after the stress is
removed, we would predict that once in non-cadmium conditions
life history traits of longitudinally exposed D. magna would be
similar to those of the control treatment. However, the opposite
occurred. When measured in cadmium-free conditions, D. magna
from experimental populations exposed to cadmium longitudinally
showed similar decrements in reproduction and population growth
rate compared to D. magna from the same populations that were,
however, exposed to cadmium. These results suggest the presence
of a carry over, transgenerational effect of the previous cadmium
exposure. Long lasting effects of contaminants after removal of
animals from exposures have been described before. Physid snails
that were exposed to cadmium during development, followed by
control conditions, displayed decreased reproductive success an
entire generation later (Kimberly and Salice, 2014). These types of
carry-over or post-exposure effects have important implications
that call into question the predictive utility of ecotoxicity data and
models. Additionally, natural populations that experience carry-
over effects are likely to experience greater adverse effects as
well as increased sensitivity to novel, concomitant stressors.
The mixed monotonic and non-monotonic responses seen in
this study were particularly interesting, if not surprising. We hy-
pothesized that D. magna exposed to high cadmium longitudinally
would respond most adversely to the cadmium challenge e they
would be most sensitive to higher temperature and/or higher
cadmium. However, this was only partially true. D. magna exposed
to long-term low cadmium were the most sensitive to the cadmium
and temperature challenge, reaching 100% mortality. Alternatively,
D. magna exposed to the high cadmium treatment longitudinally
never reached 100% mortality but had higher mortality than
D. magna from the longitudinal controls. What seems to be evident
is that high cadmium treatments were more tolerant of the cad-
mium challenge treatment than low cadmium treatments but were
less tolerant than the control D. magna. While the life history (21-
day) assays indicated that D. magna from the low cadmium treat-
ments were actually performing better (as seen in reproductive
data) compared to D. magna from the high cadmium treatments,
once exposed to a more stressful condition (high, acutely toxic
cadmium and stressful temperature) all compensatory responses
were reversed. As might be expected, concomitant exposure to high
temperature increased mortality in all treatments, regardless of the
longitudinal cadmium concentration. Several studies have reported
increased cadmium tolerance as a result of low-level multigener-
ational exposure (Guan and Wang, 2006; Muyssen and Janssen,
2004; Salice et al., 2010) but here we show a bimodel response.
Conversely to those examples, higher exposure levels seemed to
produce D. magna that were more tolerant to a higher cadmium
 D.A. Kimberly, C.J. Salice / Environmental Pollution 207 (2015) 176e182
concentration. In terms of explanation, it may be possible that
D. magna exposed to high cadmium had acquired higher cadmium
tolerance by expending more energy to cadmium tolerance or had
higher expression levels of protective molecules (an acclimation) or
some selection may have occurred (an adaptation) (Klerks and
Weis, 1987). Because D. magna used in these experiments were
from the same culture initiated with few individuals and because
exposure duration was still relatively short, we suspect that adap-
tation by selection may have been unlikely (although see Salice
et al., 2010). Ward and Robinson (2004) showed that selection for
cadmium tolerance in D. magna was possible in eight generations
(or fewer) but the starting population of D. magna was a combi-
nation from eight different sources and selection was evident based
on increased cadmium tolerance and decreased genetic diversity.
Investigations into more plastic responses to cadmium induced
stress as seen in this study would seem warranted and could
include both energetic and biochemical mechanisms.
A number of mechanisms can be involved in increased metal
tolerance. The binding of cadmium (and other metals) by the
sulphur rich metallothionein protein is a mechanism widely stud-
ied and observed in numerous taxa. Metallothionein (MT) can in-
crease tolerance to metals when the protein is up-regulated
following exposure to certain metals including cadmium (Roesijadi,
1992). The up-regulation of MT has been observed in a wide variety
of species including the least killifish (Xie and Klerks, 2003) and
chironomids exposed to cadmium (Postma et al., 1996). Addition-
ally, the resistance of D. magna to sublethal concentrations of
cadmium within only one generation has been attributed to the up-
regulation of metallothionein-like proteins (Bodar et al., 1990).
While single genes have been found to be responsible for the
increased resistance mechanism (Salice and Roesijadi, 2002), a
battery of genes may be responsible for explaining the entire
response observed (Salice et al., 2010). Certainly an improved un-
derstanding of adaptive responses and costs will come from studies
that include both natural history responses and molecular
mechanisms.
In conclusion, the interaction between environmental stressors
and population-level responses should be considered in long term,
multigenerational designs. These designs, paired with short-term
studies, will improve our understanding of the range of responses
that organisms my display when faced with contaminants under
long-term exposure conditions. Using such methods we found that
exposure to a novel stressor environment after long-term exposure
to cadmium may produce non-monotonic results; mortality of
those animals exposed to high stressor levels longitudinally was
lower than mortality of animals exposed to lower stressor levels
longitudinally. Moreover, it was observed that both life history as-
says and l with animals exposed to cadmium longitudinally dis-
played decreased reproduction and l compared to controls.
Therefore, the adverse effect of cadmium exposure was shown to
persist, at least lasting to the next generation.
Acknowledgements
We thank Texas Tech University and The Institute of Environ-
mental and Human Health (TIEHH) for previous funding for D.A.
Kimberly. This research was also supported by a grant from the
Conchologists of America. The manuscript was improved by helpful
comments from anonymous reviewers.
References
Baldwin, W.S., Milam, D.L., Leblanc, G.A., 1995. Physiological and biochemical per-
turbations in Daphnia magna following exposure to the model environmental
estrogen diethylstilbestrol. Environ. Toxicol. Chem. 14, 945e952.
181
Barnthouse, L.W., Munns, W.R., Sorensen, M.T., 2007. Population-level Ecological
Risk Assessment. CRC Press.
Bervoets, L., Baillieul, M., Blust, R., Verheyen, R., 1996. Evaluation of effluent toxicity
and ambient toxicity in a polluted lowland river. Environ. Pollut. 91, 333e341.
Biesinger, K.E., Christensen, G.M., 1972. Effects of various metals on survival, growth,
reproduction, and metabolism of Daphnia magna. J. Fish. Board Can. 29,
1691e1700.
Bodar, C.W.M., Van der Sluis, J.C., Voogt, P.A., Zandee, D.I., 1990. Cadmium resistance
in Daphnia magna. Aquat. Toxicol. 16, 33e39.
Carrol, S.P., Hendry, A.P., Reznick, D.N., Fox, C.W., 2007. Evolution on ecological time-
scales. Funct. Ecol. 21, 387e393.
Caswell, H., 2001. Matrix Population Models, second ed. Sinauer Associates,
Sunderland.
Clements, W.H., 1999. Metal tolerance and predator-prey interactions in benthic
macroinvertebrate stream communities. Ecol. App. 9, 1073e1084.
De Schamphelaere, K.C., Janssen, C.R., 2002. A biotic ligand model predicting acute
copper toxicity for Daphni magna: the effects of calcium, magnesium, sodium,
potassium, and pH. Environ Sci. Technol. 36, 48e54.
Environmental Integrity Project and Earthjustice, 2010. Out of Control: Mounting
Damages from Coal Ash Waste Sites. Washington, DC.
Forbes, V.E., Forbes, T.L., 1994. Ecotoxicology in Theory and Practice. Chapman and
Hall, London.
Forbes, V.E., Calow, P., Sibly, R.M., 2008. The extrapolation problem and how pop-
ulation modeling can help. Environ. Toxicol. Chem. 27, 1987e1994.
Guan, R., Wang, W.X., 2006. Comparison between two clones of Daphnia magna:
effects of multigenerational cadmium exposure on toxicity, individual fitness,
and biokinetics. Aquat. Toxicol. 76, 217e229.
Hairston, N.G., Ellner, S.P., Geber, M.A., Yoshida, T., Fox, J.A., 2005. Rapid evolution
and the convergence of ecological and evolutionary time. Ecol. Lett. 8,
1114e1127.
Heugens, E.H., Tokkie, L.T., Kraak, M.H., Hendriks, A.J., van Straalen, N.M.,
Admiraal, W., 2006. Population growth of Daphnia magna under multiple stress
conditions: joint effects of temperature, food, and cadmium. Environ. Toxicol.
Chem. 25, 1399e1407.
Jager, T., Zimmer, E.I, 2012. Simplified dynamic energy budget model for analysing
ecotoxicity data. Ecol. Model 225, 74e81.
Kashian, D.R., Zuellig, R.E., Mitchell, K.A., Clements, W.H., 2007. The cost of toler-
ance: sensitivity of stream benthic communities to UV-B and metals. Ecol. App.
17, 365e375.
Kimberly, D.A., Salice, C.J., 2014. If you could turn back time: understanding
transgenerational latent effects of developmental exposure to contaminants.
Environ. Pollut. 184, 419e425.
Klerks, P.L., Weis, J.S., 1987. Genetic adaptation to heavy metals in aquatic organ-
isms: a review. Environ. Pollut. 45, 173e205.
Koivisto, S., Ketola, M., Walls, M., 1992. Comparison of five cladoceran species in
short-and long-term copper exposure. Hydrobiologia 248, 125e136.
LeBlanc, G.A., 1982. Laboratory investigation into the development of resistance of
Daphnia magna (strauus) to environmental pollutants. Environ. Pollut. Ser. A
Ecol. Biol. 27, 309e322.
Lopes, I., Baird, D.J., Ribeiro, R., 2004. Genetic determination of tolerance to lethal
and sublethal copper concentrations in field populations of Daphnia longispina.
Arch. Environ. Contam. Toxicol. 46, 43e51.
Martin, B.T., Jager, T., Nisbet, R.M., Preuss, T.G., Hammers-Wirtz, M., Grimm, V., 2013.
Extrapolating ecotoxicological effects from individuals to populations: a generic
approach based on Dynamic Energy Budget theory and individual-based
modeling. Ecotoxicology 22, 574e583.
Munzinger, A., 1994. Influence of nickel on population dynamics and on some de-
mographic parameters of Daphnia magna. Hydrobiologia 277, 107e120.
Münzinger, A., Monicelli, F., 1992. Heavy metal co-tolerance in a chromium tolerant
strain of Daphnia magna. Aquat. Toxicol. 23, 203e216.
Muyssen, B.A., Janssen, C.R., 2004. Multi-generation cadmium acclimation and
tolerance in Daphnia magna Strauus. Environ. Pollut. 130, 309e316.
Newman, M.C., Aplin, M.S., 1992. Enhancing toxicity data interpretation and pre-
diction of ecological risk with survival time modeling: an illustration using
sodium chloride toxicity to mosquito fish (Gambusia holbrooki). Aquat. Toxicol.
23, 85e96.
Nisbet, R.M., Muller, E.B., Lika, K., Kooijman, S.A.L.M., 2000. From molecules to
ecosystems through dynamic energy budget models. J. Animal Ecol. 69,
913e926.
Postma, J.F., Van Kleunen, A., Admiraal, W., 1995. Alterations in life-history traits of
Chironomus riparius (Diptera) obtained from metal contaminated rivers. Arch.
Enviro. Contam. Tox. 29, 469e475.
Postma, J.F., van Nugteren, P., Buckert-De Jong, M.B., 1996. Increased cadmium
excretion in metal-adapted populations of the midge chironomus riparius
(Dipter). Environ. Toxicol. Chem. 15, 332e339.
R Development Core Team, 2010. R: A Language and Environment for Statistical
Computing. R Foundation for Statistical Computing, Vienna, Austria.
Roesijadi, G., 1992. Metallothioneins in metal regulation and toxicity in aquatic
animals. Aquat. Toxicol. 22, 81e113.
Salice, C.J., Anderson, T.A., Roseijadi, G., 2010. Adaptive responses and latent costs of
multigeneration cadmium exposure in parasite resistant and susceptible strains
of a freshwater snail. Ecotoxicology 19, 1466e1475.
Salice, C.J., Miller, T.J., Roesijadi, G., 2009. Demographic responses to multi-
generation cadmium exposure in two strains of the freshwater gastropod,
Biomphalaria glabrata. Arch. Environ. Contam. Toxicol. 56, 785e795.
182 D.A. Kimberly, C.J. Salice / Environmental Pollution 207 (2015) 176e182
Salice, C.J., Roesijadi, G., 2002. Resistance to cadmium and parasite infection are
inversely related in two strains of a freshwater gastropod. Environ. Toxicol.
Chem. 21, 1398e1403.
Sibly, R.M., Hone, J., 2002. Population growth rate and its determinants: an over-
view. Philo. Trans. Roy. Soc. B: Biol. Sci. 357, 1153e1170.
Stuhlbacher, A., Bradley, M.C., Naylor, C., Calow, P., 1992. Induction of cadmium
tolerance in two clones of Daphnia magna straus. Comp. Biochem. Physiol. Part
C Comp. Pharmacol. 101, 571e577.
United States Environmental Protection Agency (USEPA) document EPA-821-R-02-
012, 2002. United States Environmental Protection Agency. Washington, DC.
Venables, W.N., Ripley, B.D., 2002. Modern Applied Statistics with S. Springer, New
York.
Villarroel, M.J., Ferrando, M.D., Sancho, E., Andreu, E., 2000. Effects of tetradifon on
Daphnia magna during chronic exposure and alterations in the toxicity to
generations pre-exposed to the pesticide. Aquat. Toxicol. 49, 39e47.
Walker, C.H., 2006. Principles of Ecotoxicology. CRC Press, Boca Raton, FL.
Ward, T.J., Robinson, W.E., 2004. Evolution of cadmium resistance in Daphnia
magna. Environ. Toxicol. Chem. 24, 2341e2349.
World Bank Group, 1999. Pollution Prevention and Abatement Handbook. Wash-
ington, DC.
Xie, L., Klerks, P.L., 2003. Responses to selection for cadmium resistance in the least
killifish, Heterandria formosa. Environ. Toxicol. Chem. 22, 313e320.