Differential Toxicity to Cd, Pb, and Cu in Dragonfly Larvae (Insecta: Odonata)

Abstract

Odonate larvae are important organisms in aquatic ecosystems but have been rarely studied in laboratory toxicity tests. Only a few previous studies have been conducted on odonates and their responses to heavy metals. We exposed two species of libellulid larvae (Anisoptera: Libellulidae) to equimolar concentrations of cadmium, lead, or copper in 7-day survival tests. Larvae were tolerant of high concentrations of cadmium and lead, as no significant decrease in survival was observed at exposures as high as 0.893 and 2.232 mM, respectively. In contrast, larvae were more sensitive to copper exposure, demonstrating significantly decreased survival to exposures as low as 2.360 microM. In whole animal samples, larvae accumulated very high concentrations (>1000 microg/g dry weight) of all three metals in an exposure-related manner. Much of this accumulation could probably be attributed to adsorption or accumulation of metal within the exoskeleton, because odonate larvae are known to sequester metals into this material. Our results were generally consistent with previous observations indicating that odonates are tolerant to metal exposures, even in comparison with other aquatic invertebrates. However, there are few studies that have used odonates in toxicity tests and compared these organisms to other aquatic life. Based on their abundance and their simple requirements in the laboratory, we believe that odonate larvae can be useful toxicological model organisms.

Full text

Differential Toxicity to Cd, Pb, and Cu in Dragonfly Larvae
(Insecta: Odonata)
V. D. Tollett Æ E. L. Benvenutti Æ L. A. Deer Æ
T. M. Rice
Received: 16 November 2007 / Accepted: 11 March 2008 / Published online: 18 April 2008
Ó Springer Science+Business Media, LLC 2008
Abstract Odonate larvae are important organisms in
aquatic ecosystems but have been rarely studied in labora-
tory toxicity tests. Only a few previous studies have been
conducted on odonates and their responses to heavy metals.
We exposed two species of libellulid larvae (Anisoptera:
Libellulidae) to equimolar concentrations of cadmium, lead,
or copper in 7-day survival tests. Larvae were tolerant of
high concentrations of cadmium and lead, as no significant
decrease in survival was observed at exposures as high as
0.893 and 2.232 mM, respectively. In contrast, larvae were
more sensitive to copper exposure, demonstrating signifi-
cantly decreased survival to exposures as low as 2.360 lM.
In whole animal samples, larvae accumulated very high
concentrations ([1000 lg/g dry weight) of all three metals
in an exposure-related manner. Much of this accumulation
could probably be attributed to adsorption or accumulation
of metal within the exoskeleton, because odonate larvae are
known to sequester metals into this material. Our results
were generally consistent with previous observations indi-
cating that odonates are tolerant to metal exposures, even in
comparison with other aquatic invertebrates. However,
there are few studies that have used odonates in toxicity
tests and compared these organisms to other aquatic life.
Based on their abundance and their simple requirements in
the laboratory, we believe that odonate larvae can be useful
toxicological model organisms.
Odonates (Insecta: Odonata; dragonflies and damselflies)
are abundant and important members of a variety of
freshwater ecosystems (Corbet 1999). The aquatic larvae
are predators of invertebrates as well as vertebrates such as
fish and amphibian larvae. Odonate larvae, in turn, serve as
an important prey base for fish and other aquatic predators.
Upon metamorphosis and emergence, adult odonates
become important predators on insects and continue to act
as a food source for terrestrial predators such as amphibi-
ans and birds.
Because they have such an important role in freshwater
systems, odonate larvae are included in many environ-
mental assessments (Rutherford and Mellow 1994;
Karouna-Renier and Sparling 2001; Scher and Thie
`
ry
2005). However, few studies have documented the
responses of odonates to environmental contaminants. As
might be expected, the effects of insecticides have been
frequently studied (Anadu et al. 1996; Beketov 2004;
Bhardwaj and Tyagi 1993; Giddings et al. 1996; Hardersen
and Wratten 2000; Rohr and Crumrine 2005; Schroer et al.
2004). However, less information has been collected on the
responses of odonates to metal contaminants. None of these
studies are very recent, and some are not very thorough.
Sloof (1983) exposed a variety of invertebrate larvae,
including the odonate Ischnura elegans, to toxicants,
including mercury and cadmium. I. elegans larvae had
considerably higher tolerance of metals, based on 48-h
median lethal concentrations (LC
50
s), compared to most
other species tested. Jones (1985) made casual observations
of odonate larvae development in settling tanks within a
former tin mine. Some malformations were observed, but
no empirical data were provided. Correa (1985) exposed
Somatochlora cingulata larvae to aluminum and low pH
levels. Adverse effects such as decreased oxygen con-
sumption were the result of exposure to low pH rather than
aluminum. Meyer et al. (1986) exposed Libellula depressa,
Libellula quadrimaculata, and Aeshna cyanea larvae to
V. D. Tollett  E. L. Benvenutti  L. A. Deer  T. M. Rice (&)
Department of Biological Sciences, University of South
Alabama, Mobile, AL 36688, USA
e-mail: trice@jaguar1.usouthal.edu
123
Arch Environ Contam Toxicol (2009) 56:77–84
DOI 10.1007/s00244-008-9170-1

lead and measured organ bioaccumulation and oxidative
enzyme activity. Most of the accumulated lead was found
in the cuticle, and enzyme activity was suppressed. Mackie
(1989) determine 96-h LC
50
s for Enallagma sp. larvae
exposed to cadmium, lead, and aluminum. Cadmium was
found to be the most toxic of the three metals. Rockwood
and colleagues (1990, 1991) observed decreases in weight
and oxygen uptake and changes in hemolymph chemistry
in Libellula julia larvae exposed to aluminum. The most
recent study was conducted by Tennessen (1993), who
documented hatching success and development of Libell-
ula lydia and Pachydiplax longipennis larvae exposed to
iron.
Because there are so few laboratory studies regarding
effects of metals on odonate larvae, we conducted a series
of basic exposures with cadmium, copper, and lead to
determine the effects on survivability. Equimolar expo-
sures were used so that we could directly compare the
relative toxicity of these three metals. The study species
were larvae of P. longipennis and Erythemis simplicicollis
(Anisoptera: Libellulidae). To our knowledge, this work is
the first study of copper toxicity in odonate larvae. In
addition to toxicity tests, tissue levels of metals were
measured in P. longipennis so that we could compare
toxicity and bioaccumulation. This topic has also not been
adequately addressed in odonates. Our ultimate goal was to
provide recent information regarding the toxicity of metals
to odonate larvae.
Materials and Methods
Collection of Larvae
Odonate larvae were collected from 2004 to 2006 from a
small pond on the campus of the University of South
Alabama, Mobile, Alabama, USA. The most common
species included P. longipennis, E. simplicicollis, and
Ladona deplanata (Anisoptera: Libellulidae). Specimens
were identified through dichotomous keys (Richardson
2003). The most consistently abundant species in all years
was P. longipennis; therefore, these larvae were used in
the majority of the experiments described in this article.
Larvae were collected with a D-frame net within leaf
litter and aquatic vegetation. Specimens as small as 5 mm
and as large as 40 mm (late instar P. longipennis) were
collected; most individuals were between 10 and 25 mm.
Larvae were brought back to the laboratory room and
allowed to acclimate to room temperature (23°C) in the
collection water. Water from the pond contained only
trace amounts of metals (Pb and Cu \5 lg/L; Cd below
detection limit).
Maintenance of Odonate Larvae
The maintenance and housing design has been described
previously (Rice 2008). Briefly, after adjusting to labora-
tory temperature, larvae were placed into housing
chambers made from 480-mL (16-oz) plastic drinking
cups. Four 3 9 5-cm windows were cut into the sides of
each cup; each window was covered with nylon window
screen (mesh size: 0.84 mm). Typically, one to three lar-
vae, depending on size, were placed in each chamber. A
maximum of 10 housing chambers were placed in a
38 9 38 9 16.5-cm translucent polyethylene box. Each
plastic box was filled to a depth of 11 cm with reconsti-
tuted hard water (FETAX solution; ASTM 2000). The
water did not to need to be changed because it was filtered
and recirculated. The entire system was held in a laboratory
room under a 12-h light:12-h dark photoperiod regime at
23°
C. Larvae were collected for all trials within 2 weeks or
less prior to use in experiments. The size of larvae used for
experiments ranged between 10 and 15 mm. They were fed
two to three Daphnia magna, two to three times per week,
prior to use in experiments. Daphnia magna were pur-
chased from Carolina Biological Supply (Burlington, NC,
USA). They were maintained in 1-L polypropylene tri-
corner beakers containing reconstituted hard water and
were fed a yeast/fish flake/cereal grass mix (YTC)
according to standard methods (Landis et al. 2005).
General Experimental Design
All experiments were conducted over 7 days in the labo-
ratory room at 23°C and a 12-h light:12-h dark
photoperiod. Exposure chambers consisted of 400-mL
polyethylene beakers filled to a test volume of 360 mL of
FETAX solution at the designated treatment concentration,
with one larva per beaker. Water quality parameters gen-
erally measured in the exposure chambers were as follows:
pH = 6.24, hardness = 120 mg CaCO
3
/L, and tempera-
ture = 23°C. Stock solutions of 10 g/L Pb, Cu, and Cd
were made from metal salts (lead nitrate: Pb(NO
3
)
4
; copper
sulfate pentahydrate: CuSO
4
 5H
2
O; cadmium chloride
hemipentahydrate: CdCl
2
 2½ H
2
O), which were dissolved
in ultrapure water. Larvae were not fed during the expo-
sure. Four to five replicate beakers (i.e., four to five larvae)
per treatment were used, depending on availability of lar-
vae. Beakers were checked once per day for dead larvae, as
determined by lack of response to prodding. A complete
water change was conducted on day 3 of each trial.
Exposure to Cd, Pb, and Cu
The main series of experiments used P. longipennis larvae
that were exposed to nominal concentrations of 0, 0.045,
78 Arch Environ Contam Toxicol (2009) 56:77–84
123

0.357, 0.893, and 2.232 mM of Cd, Pb, or Cu. One trial
was conducted with all three metals during the same 7-day
exposure period, with four replicate beakers per treatment
for each metal. Additionally, a second Cd trial (five repli-
cates per beaker) and two trials each of Cu or Pb (four
replicates per beaker) were conducted during single 7-day
periods. Statistical analysis (below) was conducted on
composite data for each metal.
Additional experiments were also performed. Because
of low survivability in the initial Cu treatments described
earlier, a single 7-day trial with P. longipennis was con-
ducted using low levels of Cu, at exposures of 0, 0.295,
0.590, 1.180, and 2.360 lM, with four replicates per
treatment. In 2004, E. simplicicollis were abundant in the
collection pond. Therefore, we exposed larvae (similar in
size to P. longipennis) to Cd at 0, 0.022, 0.045, 0.134,
0.357, and 0.893 mM. Two separate trials, with five rep-
licates per exposure, were conducted.
Metal Levels in Whole Animal Samples
Using P. longipennis, four larvae were selected each from
0-, 0.045-, or 2.232-mM exposures of Cd, Pb, or Cu from
various trials to determine metal levels in whole animal
samples. Tissue processing methods were modified from
US EPA Method 3051 (US EPA 1994). Specimens were
frozen after removal from the particular exposure experi-
ment. Individual larvae were thawed, rinsed with ultrapure
water, and placed in a 45-mL Teflon digestion vial with
2 mL ultrapure nitric acid. These vials were placed into
Parr
Ò
microwave digestion bombs (Parr Instrumental
Company, Moline, IL, USA), which were then placed into
a microwave and heated at 750 W for 3 min. Bombs were
cooled, vented, and then microwaved a second time at 600
W for 2 min. The resulting digestate was quantitatively
transferred to an acid-washed 50-mL centrifuge tube for
analysis and diluted to 20 mL with ultrapure water. To
report the data on a dry weight basis, a subset of larvae
were weighed wet, dried for 48 h at 65°C, and weighed
again. The average dry:wet ratio was 10%; this value was
used to convert the wet weights of digested larvae to dry
weight.
Standard reference materials were digested with batches
of larvae samples to monitor extraction efficiency. These
reference materials consisted of 0.25 g of NIST 1566b
(National Institute of Standards and Technology: oyster
tissue) during Pb and Cd extraction and NIST 2976 (mussel
tissue) during Cu extraction. Recovery efficiency from
these reference materials was 95–100%. Metal levels were
analyzed on a Varian
Ò
SpectrAA220 graphite furnace
atomic absorption spectrophotometer (Varian, Inc., Palo
Alto, CA, USA). The instrumental detection limit was
approximately 0.01 lg metal/g dry weight.
Statistical Analysis
Statistical analysis consisted of analysis of variance
(ANOVA) to compare survival time among exposure
concentrations within each metal. Separate analyses were
conducted on the main series with Cd, Pb, and Cu (com-
posite of all separate metal trials), on the low Cu trial, and
on a composite of the two Cd trials with E. simplicicollis.
Tukey’s multiple comparisons were used to separate sig-
nificant differences among all metal treatments within a
particular metal. Student’s t-tests were used to compare
survival time between P. longipennis and E. simplicicollis
within 0.045-, 0.357-, or 0.893-mM Cd exposures.
ANOVA was also used to compare metal levels in larvae
among 0-, 0.045-, and 2.232-mM treatments within a
metal. For this analysis, data were log
10
-transformed due to
extreme heterogeneity of variances among the treatments.
Results
Analysis of composite trials with Pb revealed no significant
difference in survival time for P. longipennis exposed to
any level of Pb (F
4, 55
= 0.64, p = 0.639; Fig. 1). For com-
posite Cd trials, survival time was significantly lower in the
2.232-mM treatment compared to 0.357- and 0.045-mM
treatments (F
4, 40
= 3.18, p = 0.023; Fig. 1). In contrast to
results from Cd or Pb exposure, P. longipennis larvae
exposed from 0.045 to 2.232 mM Cu showed significant
decreases in survival time compared to unexposed lar-
vae (F
4, 55
= 24.29, p = 0.0001; Fig 1). No other significant
0
1
2
3
4
5
6
7
control 0.045 0.357 0.893 2.232
Concentration (mM of metal)
Mean Days of Survival
cadmium
lead
copper
y
y
y
y
b
xab a a ab
Fig. 1 Mean survival (±1 SE) for P. longipennis larvae over 7-day
exposures to equimolar concentrations of cadmium, lead, or copper.
There were two trials with cadmium (N = 9) and three trials each
with lead and copper (N = 12). Each trial used four to five larvae per
treatment. Within Cd or Cu treatments, different letters indicate
significant differences in survival time. There were no significant
differences among any Pb treatments
Arch Environ Contam Toxicol (2009) 56:77–84 79
123

differences were observed among any other concentrations.
Cu levels as low as 2.360 lM significantly decreased sur-
vival time compared to unexposed larvae (F
4, 15
= 3.55,
p = 0.031; Fig. 2).
A significant treatment effect was detected in survival
time for E. simplicicollis larvae exposed to Cd (F
4, 54
=
2.92, p = 0.021; Fig. 3). However, the only significant
pairwise comparison was between the 0.893- and 0.045-
mM treatments. E. simplicicollis larvae exposed to 0.045 or
0.357 mM Cd had a similar survival time compared to that
of P. longipennis in the same treatments (df = 17,
t \ 1.71, p [ 0.05; Fig. 4) but a significantly lower sur-
vival time in the 0.893-mM treatment (df = 17, t = 3.36,
p = 0.001; Fig. 4).
Pachydiplax longipennis larvae exposed to 0.045 or
2.232 mM Cd, Pb, or Cu accumulated high levels of these
metals (Table 1). There was considerable variability within
each treatment, but, in general, the level of metals was
consistent with exposure level. For both Cd and Cu, there
were significant differences in metal concentrations among
all three treatments (F
2, 9
[ 62.74,p \ 0.001, based on log
10
–
transformed data; Table 1). For Pb, unexposed larvae had
significantly lower concentrations than both 0.045 and
2.232 mM, but there were no differences between the two
Pb exposure treatments (F
2, 9
= 9.19, p \ 0.001, based on
log
10
-transformed data; Table 1).
Discussion
Both P. longipennis and E. simplicicollis larvae exhibited
high tolerance to Pb and Cd, at least in terms of surviv-
ability. No appreciable mortality was observed in either
species at concentrations below 0.893 mM (100 mg Cd/L,
185 mg Pb/L). Pachydiplax longipennis larvae were able to
tolerate up to 2.232 mM Cd and Pb (250 mg Cd/L, 462 mg
Pb/L). Only exposures to Cu demonstrated any effect on
mortality at concentrations above 2.360 lM(150lgCu/L).
All of the concentrations of Pb, Cu, or Cd that caused
mortality were well above any concentration to which
odonate larvae would be exposed in the field, except under
extreme contamination scenarios. These concentrations
also greatly exceed the US EPA-recommended Criterion
Continuous Concentration to protect aquatic life [CCC at
water hardness of 100 mg/L CaCO
3
:Pb\ 2.50 lg/L
(0.012 lM); Cu 9.00 lg/L (0.142 lM); Cd \ 0.25 lg/L
(0.002 lM); US EPA 2005].
The mortality from exposure to Cu but not to Cd or Pb
might be due to the ability of aquatic insects such as
odonates to more readily bioaccumulate Cu. Metal-binding
metallothionein proteins have been found in some species
0
1
2
3
4
5
6
7
control 0.295 0.590 1.180 2.360
Concentration (
µµ
M copper)
Mean Days of Survival
b
a
ab
ab
ab
Fig. 2 Mean survival (±1 SE) for P. longipennis larvae over a 7-day
exposure to copper. Data consists of a single 7-day trial with four
larvae per treatment. Different letters indicate significant differences
in survival time
0
1
2
3
4
5
6
7
control 0.022 0.045 0.134 0.357 0.893
Concentration (mM cadmium)
Mean Days of Survival
b
aba
ab
ab
ab
Fig. 3 Mean survival (±1 SE) for E. simplicicollis larvae over 7-day
exposures to cadmium. Data consists of a composite of two trials.
Each trial used five larvae per treatment; total sample sizes = 10 for
each treatment. Different letters indicate significant differences in
survival time
0
1
2
3
4
5
6
7
398.0753.0540.0
Concentration (mM Cd)
Mean Days of Survival
P. longipennis
E. simplicicollis
*
Fig. 4 Mean survival (±1 SE) for P. longipennis and E. simplici-
collis larvae over 7-day exposures to cadmium. Data for each species
consisted of a composite of two trials. Each trial used four to five
larvae per treatment. Asterisks indicate significant differences in
survival time between species within a treatment. See Figure 1
(P. longipennis) and Figure 3.(E. simplicicollis) for other details
80 Arch Environ Contam Toxicol (2009) 56:77–84
123

of insect larvae. Cu is preferentially bound more readily by
these proteins than Cd, whereas Pb is minimally bound
(Maroni and Watson 1985, Suzuki et al. 1988, 1989).If
odonate larvae have metal-binding proteins, then they
might bioaccumulate Cu more readily that Cd or Pb and
then show toxic effects. However, the presence of these
proteins in odonates remains unexplored.
The concentrations in the present study were also con-
siderably higher than levels used in the few previous
experiments on odonates exposed to Cd or Pb; no inves-
tigators have examined the toxicity of Cu in odonates.
Meyer et al. (1986) exposed Libellula depressa, L. quad-
rimaculata, and A. cyanea larvae to 20 lg/L (0.097 lM) of
Pb for 6 weeks. No mortality was observed, but activity of
oxidative enzymes was decreased. Meyer et al. (1986) also
observed that food-catching behaviors were markedly
decreased after 2 weeks of exposure. In contrast, we
observed no changes in appetite for E. simplicicollis
exposed to 0.357 mM (73.97 mg/L) during a single 14-day
trial (data not shown). Mackie (1989) conducted 96-h
exposures of Cd, Pb, and other metals with Enallagma sp.
larvae. Median lethal concentrations (LC
50
) ranged from
7.05 to 10.66 mg Cd/L (0.063 to 0.095 mM), well below
the highest concentration used in our experiments where no
mortality was observed. The mortality observed by Mackie
(1989) might be explained by differences in species or in
general experimental design. In contrast to exposures to
Cd, Mackie (1989) observed no mortality in Enallagma sp.
larvae exposed to concentrations of Pb above 60 mg/L
(0.290 mM). These results were consistent with our
observations of little appreciable mortality in Pb-exposed
libellulids. Sloof (1983) exposed I. elegans larvae to Cd
and determined the LC
50
to be [56 mg/L (0.500 mM).
Chessman and McEvoy (1998) did not conduct exposures,
but, instead, they calculated indexes of sensitivity to vari-
ous environmental insults in Australian watersheds. The
authors determined that lestid and libellulid larvae
appeared to be relatively insensitive to metal pollution
compared to other insults such as sewage or dams. The
results of the above studies all suggest that odonate larvae,
at least libellulids, are tolerant of water-borne heavy met-
als. Even when effects were observed, the exposure
concentrations were typically in the milligram per liter
(millimolar) range, well above any levels that would be
expected in the field.
The above-cited studies represent the only investigations
of odonate responses to Cd or Pb. Furthermore, because no
experiments have been conducted with Cu, we cannot be
confident that the high mortality we observed compared to
Cd or Pb exposure would be expected or instead is a unique
phenomenon among aquatic insects. Considering that od-
onates include two distinct suborders (Zygoptera and
Anisoptera) and a variety of natural histories within these
groups, more controlled studies need to be conducted with
a variety of species to more fully examine metal bioaccu-
mulation and toxicity in these organisms.
It would be useful to compare the sensitivity of metals
between odonates and other aquatic invertebrates, but only
two studies have made such direct comparisons. Sloof
(1983) determined that I. elegans were more tolerant to Cd
compared to other noninsect aquatic invertebrates but
similar in tolerance to other aquatic insect larvae. Mackie
(1989) observed that Enallagma sp. were more tolerant to
Cd or Pb compared to molluscs or ephemeropteran larvae.
Because of the lack of any other direct comparisons to
odonates, an indirect alternative would be to examine how
other aquatic invertebrates respond differentially to Cd, Pb,
and Cu. Warnick and Bell (1969) exposed a variety of
aquatic insect larvae to Cd, Cu, and Pb and measured
LC
50
s. All species were most sensitive to Cu and least
sensitive to Pb. Nehring (1976) observed that plecopteran
and ephemeropteran larvae were both more sensitive to Cu
compared to Pb (Cd not tested). Anderson et al. (1980)
determined that the chironomid Tanytarsus dissimulis was
most sensitive to Cd compared to Cu and least sensitive to
Pb. Rayms-Keller et al. (1998) observed that mosquito
larvae (Aedes aegypti) were less sensitive to Cu compared
to Cd (Pb not examined). Milani et al. (2003) exposed four
species of aquatic invertebrates (not odonates) to Cd and
Cu (Pb not examined). Based on 96-h LC
50
s during water-
only exposures, Hyalella sp. and Chironomus sp. were
more sensitive to Cd, whereas Hexagenia sp. and Tubifex
sp. were more sensitive to Cu. These studies generally
indicate that aquatic insect larvae, much like the odonate
larvae in the present study, are tolerant of Pb compared to
Cd or Cu. In contrast, sensitivity to Cu versus Cd varies
Table 1 Mean concentrations (±1 SE) of cadmium, lead, or copper in whole-body samples of P. longipennis larvae (N = 4) exposed to
equimolar concentrations over 7 days
Metal 0 mM 0.045 mM 2.232 mM
Cadmium 10.91 ± 8.64 a 1,085.52 ± 200.83 b 21,423.80 ± 3,330.18 c
Lead 336.39 ± 139.76 a 90,066.66 ± 16,730.46 b 189,320.60 ± 47,302.76 b
Copper 33.95 ± 10.14 a 3,190.68 ± 625.07 b 20,783.31 ± 8,612.65 c
Note: Metal levels are in micrograms per gram dry weight; the mean dry:wet weight ratio of P. longipennis was 10%. Different letters indicate
significant differences among treatments within a metal, based on log
10
-transformed data.
Arch Environ Contam Toxicol (2009) 56:77–84 81
123

among insect species. It should be noted that the above-
cited studies observed toxicity at concentrations far below
those used in the present study. Therefore, in general, od-
onates do appear to be more tolerant to metals compared to
other aquatic invertebrates.
We measured considerable amounts of Cd, Pb, and Cu
in P. longipennis larvae collected during various trials.
These whole-body concentrations appear to be extraordi-
narily high, especially considering that little toxicity was
observed for Cd or Pb. We do not believe that these high
levels are the result of sample contamination. Certainly,
there was a great deal of insoluble metal precipitate that
formed in the exposure chambers and the larvae would be
covered with this material. However, we thoroughly rinsed
the specimens prior to both freezing and processing. Fur-
thermore, recovery from standard reference materials
processed with the larvae samples was never higher than
100%, indicating no contamination of reagents. Therefore,
we are confident that the levels presented here represent the
actual levels in the specimens. However, much of this
metal might not be bioavailable but adhered onto or
sequestered into the exoskeleton. In this event, rinsing the
specimens prior to digestion would not remove these bound
metal forms.
There is considerable documentation that odonates and
other aquatic insect larvae sequester high levels of metals,
particularly Pb, in the cuticle (for a review, see Hare 1992).
For example, Giesy et al. (1981) observed that the exuviae
of Pantala hymenaea exposed in artificial microcosms
contained 68% of the Cd in whole specimens. Meyer et al.
(1986) exposed three species of anisopteran larvae to Pb
and measured levels in multiple organs and exoskeleton.
Composite samples of these species demonstrated that
most accumulated Pb was sequestered in the exoskeleton
compared to the brain, fat bodies, midgut, or rectum. Gupta
(1995) collected Crocothemis servilia from lakes in India
and measured Cd, Pb, and Cu levels. The greatest pro-
portion of whole-body metal levels was sequestered into
the exoskeleton (100%, 75%, and 68% for Cd, Pb, and Cu,
respectively). Based on these previous observations, it is
possible that P. longipennis larvae in the present study had
high levels of metals adsorbed onto or accumulated into
their exoskeletons. These metal species were unlikely to be
very bioavailable, given the lack of any obvious toxicity
even at high exposure concentrations.
We measured relatively high levels of Pb even in our
unexposed control larvae, and we have no good explana-
tion for these high levels. Analysis of water from the
collection area indicated only trace amounts of all three
metals (Pb and Cu \5 lg/L; Cd below detection limit).
The levels in the unexposed treatments were higher than
any Pb amounts reported in whole-larvae samples from
field sites. In most cases, the Pb concentrations in odonates
collected from these areas were \20 lg/g dry weight
(Anderson 1977; Barak and Mason 1989; Karouna-Renier
and Sparling, 2001; Mathis et al., 1979; Nummelin et al.,
2007; Scheuhammer et al., 1997). Gupta (1995) did mea-
sure up to 50 lg/g dry weight in C. servilia from lakes in
India, but there was no indication that these lakes were
contaminated.
It is possible that odonate larvae are capable of bioac-
cumulating high amounts of Pb even under low-level
exposures, perhaps from the collection area or the exposure
water. As described previously, Meyer et al. (1986)
exposed three species of Anisoptera to Pb and found that
the exoskeleton had the highest levels of bioaccumulation.
Furthermore, the levels in the exoskeleton of Pb-exposed
and unexposed larvae were nearly equivalent. This obser-
vation, much like that in the present study, indicated that
unexposed larvae were carrying relatively high levels of Pb
(Meyer et al.
1986). However, the concentrations of Pb
reported by Meyer et al. (\10 lg/g dry weight using our
dry:wet weight ratios) were considerably lower than
observed in the present study. These investigators used a
lower exposure concentrations (20 lg/L = 0.097 lM) for
6 weeks. Meyer et al. (1986) hypothesized that most of the
Pb in the odonate exoskeleton was held in the mesocuticle,
which becomes reincorporated into the new cuticle during
subsequent molting periods. Therefore, previously accu-
mulated Pb remains and so there would be potential, even
under low Pb-exposure scenarios, for odonate larvae to
accumulate a relatively high Pb burden with repeated
molts. We did observe molting by several larvae during
various experiments; therefore, the extremely high levels of
metals we measured in unexposed P. longipennis could be
consistent with these observations from Meyer et al.
(1986). Unfortunately, Meyer et al. (1986) are the only
investigators who observed Pb uptake within multiple
organs between exposed and unexposed odonate larvae.
Therefore, more studies are required that examine the sites
of uptake of Pb and other metals in odonates and other
aquatic insects.
Unlike the relatively high Pb levels in unexposed P.
longipennis in the present study, the levels of Cd and Cu in
unexposed larvae were comparatively low. These levels
probably represent low background levels; they were
within values reported in previous studies for odonate
larvae collected from various field sites. These field levels
were generally \25 lg Cu/g and \2 dry lg Cd/g dry
weight (Anderson 1977; Barak and Mason 1989; Currie
et al. 1997; Karouna-Renier and Sparling 2001; Nummelin
et al. 2007; Scheuhammer et al. 1997). The values mea-
sured in the present study bracket these previously reported
concentrations. Some investigators have reported higher
levels, in some cases from contaminated areas. Brown
(1977) measured Cu ranging from 48 to 768 lg/g dry
82 Arch Environ Contam Toxicol (2009) 56:77–84
123

weight from Libellula sp. and Agrion sp. larvae collected in
a mine drainage area in Cornwall, United Kingdom. Mathis
et al. (1979) measured Cd levels of *35.4 lg Cd/g wet
weight (354 lg/g dry weight based on our dry:wet ratio)
from power-plant ponds in Illinois, United States. Gupta
(1995) measured up to 45 lg Cu/g dry weight, but only
6 lg Cd/g dry weight, in C. servilia from lakes in India.
Based on these studies, we believe that the Cd and Cu
levels in unexposed P. longipennis larvae in the present
study could be considered within normal background
values.
In conclusion, P. longipennis and E. simplicicollis lar-
vae appear to be tolerant of heavy metals and capable of
accumulating high body burdens with little effect on
mortality. We suggest that future studies examine other
sublethal end points, such as predator-avoidance, devel-
opment, or changes in appetite. In field situations, odonate
larvae might not be at risk of toxic effects from metal
exposure even when high levels have been bioaccumulated.
However, predators of odonate larvae might be at risk from
ingestion of metal-laden larvae. Unfortunately, no studies
have examined the transfer of metals from odonate larvae
to predators.
Similar properties of metal tolerance have been
observed for frog larvae, which also do not demonstrate
toxic effects until high levels of metals have been bioac-
cumulated (Ferreira et al. 2004; Hopkins et al. 2000; Rice
et al. 1999, 2002). Odonate larvae share some similar
properties to frog larvae. The larval stage of both types are
abundant members of a variety of freshwater ecosystems
and serve as food base for high-level predators such as fish.
Furthermore, both odonate larvae and frog larvae undergo
extensive metamorphosis into adults that are important
insect predators while still serving as an important food
base. Much support in recent years has been presented for
frog larvae as important organisms for environmental
toxicology in both field and lab studies (e.g., Hopkins et al.
2000; Rice et al. 1999, 2002; Sparling et al. 2006). We
propose that the same recognition be given to odonate
larvae.
Acknowledgments We thank Claire Kelley, Lauren Gertz, and
Denise Cook for maintenance of larvae and Jyoti Rai for metal
analysis. Dr. Gene Cioffi provided consultation and access to ana-
lytical instrumentation. This work was supported by funds provided to
T. M. Rice from the Department of Biology, University of South
Alabama.
References
Anadu DI, Anaso HU, Onyeka OND (1996) Acute toxicity of the
insect larvicide Abate
Ò
(tempephos) on the fish Tilapia mela-
nopleura and the dragonfly larvae Neurocordelia virginiensis.J
Environ Sci Health Part B 31:1363–1375
Anderson RV (1977) Concentration of cadmium, copper, lead, and
zinc in thirty-five genera of freshwater macroinvertebrates from
the Fox River, Illinois and Wisconsin. Bull Environ Contamin
Toxicol 18:345–349
Anderson RL, Walbridge CT, Fiandt JT (1980) Survival and growth
of Tanytarsus dissimilis (Chironomidae) exposed to copper,
cadmium, zinc, and lead. Arch Environ Contam Toxicol 9:329–
335
ASTM (American Society for Testing, Materials) (2000) Standard
guide for conducting the frog embryo teratogensis assay-
Xenopus (FETAX). E1439–98, Annual Book of ASTM Stan-
dards, Vol 11.04. ASTM, West Conshocken, PA
Barak NAE, Mason CF (1989) Heavy metals in water, sediment and
invertebrates from rivers in eastern England. Chemosphere
19:1709–1714
Beketov MA (2004) Comparative sensitivity to the insecticides
deltamethrin and esfenvalerate of some aquatic insect larvae
(Ephemeroptera and Odonata) and Daphnia magna. Russ J Ecol
35:200–204
Bhardwaj AC, Tyagi N (1993) Impact of organochlorine pesticides on
the morphobehavior of Pantala flavescens Fabr. (Libellulidae:
Odonata). J Environ Biol 14:89–94
Brown BE (1977) Effects of mine drainage on the River Hayle,
Cornwall. A: factors affecting concentrations of copper, zinc,
and iron in water, sediments and dominant invertebrate fauna.
Hydrobiology 52:221–233
Chessman BC, McEvoy PK (1998) Towards diagnostic indices for
river macroinvertebrates. Hydrobiology 364:169–182
Corbet PS (1999) Dragonflies: behavior and ecology of Odonata.
Comstock Publishing Association, Ithaca, NY
Correa M, Coler RA, Yin C-M (1985) Changes in oxygen consump-
tion and nitrogen metabolism in the dragonfly Somatochlora
cingulata exposed to aluminum in acid waters. Hydrobiology
121:151–156
Currie RS, Fairchild WL, Muir DCG (1997) Remobilization and
export of cadmium from lake sediments by emerging insects.
Environ Toxicol Chem 16:2333–2338
Ferreira CM, Lombardi JV, Machado-Neto JG, Bueno-Guimara
˜
es
HM, Soares SRC, Saldiva PHN (2004) Effects of copper
oxychloride in Rana catesbeiana tadpoles: toxicological and
bioaccumulative aspects. Bull Environ Contam Toxicol 73:465–
470
Giddings JM, Biever RC, Annunziato MF, Hosmer AJ (1996) Effects
of diazinon on large outdoor pond microcosms. Environ Toxicol
Chem 15:618–629
Giesy JP, Bowling JW, Kania HJ, Knight RL, Mashburn S (1981)
Fates of cadmium introduced into channels microcosm. Environ
Int 5:159–175
Gupta A (1995) Metal accumulation and loss by Crocothemis servilia
(Drury) in a small lake in Shillong, northeastern India (Anisop-
tera: Libellulidae). Odonatologica 24:283–289
Hardersen S, Wratten SD (2000) Sensitivity of aquatic life stages of
Xanthocnemis zealandica (Odonate: Zygoptera) to azinphos-
methyl and carbaryl. J Mar Freshw Res 34:117–123
Hare L (1992) Aquatic insects and trace metals: bioavailability,
bioaccumulation, and toxicity. Crit Rev Toxicol 22:327–369
Hopkins WA, Congdon J, Ray JK (2000) Incidence and impact of
axial malformations in larval bullfrogs (Rana catesbeiana)
developing in sites polluted by a coal-burning power plant.
Environ Toxicol Chem 19:862–868
Jones SP (1985) A note on the survival of dragonflies in adverse
conditions in Cornwall. J Br Dragonfly Soc 1:83–84
Karouna-Renier NK, Sparling DW (2001) Relationships between
ambient geochemistry, watershed land-use and trace metal
concentrations in aquatic invertebrates living in stormwater
treatment ponds. Environ Pollut 112:183–192
Arch Environ Contam Toxicol (2009) 56:77–84 83
123

Landis WG, Coggan CE, Gorsuch JW, Morcock RE, Palmieri MA
(2005) 8711-Daphnia. In: Eaton AD, Clesceri LS, Rice EW,
Greenberg AE (eds) Standard methods in water and wastewater.
American Public Health Association, Washington, DC, p 8-101
Mackie GL (1989) Tolerances of five benthic invertebrates to
hydrogen ions and metals (Cd, Pb, Al). Arch Environ Contam
Toxicol 18:215–223
Maroni G, Watson D (1985) Uptake and binding of cadmium, copper,
and zinc by Drosophila melanogaster larvae. Insect Biochem
15:55–64
Mathis J, Cummings TF, Gower M, Taylor M, King C (1979)
Dynamics of manganese, cadmium, and lead in experimental
power plant ponds. Hydrobiology 67:197–206
Meyer W, Harisch G, Sagredos AN (1986) Biochemical and
histochemical aspects of lead exposure in dragonfly larvae
(Odonata: Anisoptera). Ecotoxicol Environ Safe 11:308–319
Milani D, Reynoldson TB, Borgmann U, Kolasa J (2003) The relative
sensitivity of four benthic invertebrates to metals in spiked-
sediment exposures and application to contaminated field
sediment. Environ Toxicol Chem 22:845–854
Nehring RB (1976) Aquatic insects as biological monitors of heavy
metal pollution. Bull Environ Contamin Toxicol 15:147–154
Nummelin M, Lodenius M, Tulisalo E, Hirvonen H, Alanko T (2007)
Predatory insects as bioindicators of heavy metal pollution.
Environ Pollut 145:339–347
Rayms-Keller A, Olson KE, McGaw M, Oray C, Carlson JO, Beaty
BJ (1998) Effect of heavy metals on Aedes aegypti (Diptera:
Culicidae) larvae. Ecotoxicol Environ Safety 39:41–47
Rice TM (2008) A review of methods for maintaining odonate larvae
in the laboratory, with a description of a new technique.
Odonatologica 37:41–54
Rice TM, Blackstone BJ, Nixdorf WL, Taylor DH (1999) Exposure to
lead induces hypoxia-like responses in bullfrog larvae (Rana
catesbeiana). Environ Toxicol Chem 18:2283–2288
Rice TM, Oris JT, Taylor DH (2002) Effects on growth and changes
in organ distribution of bullfrog larvae exposed to lead
throughout metamorphosis. Bull Environ Contam Toxicol
68:8–17
Richardson JS (2003) Identification manual for the dragonfly larvae
(Anisoptera) of Florida. Florida Department of Environmental
Protection, Tallahassee, FL
Rockwood JP, Coler RA (1991) The effect of aluminum in soft water
at low pH on water balance and hemolymph ionic and acid-base
regulation in the dragonfly Libellula julia Uhler. Hydrobiology
215:243–250
Rockwood JP, Jones DS, Coler RA (1990) The effect of aluminum in
soft water at low pH on oxygen consumption by the dragonfly
Libellula julia Uhler. Hydrobiology 190:55–59
Rohr JR, Crumrine PW (2005) Effects of an herbicide and an
insecticide on pond community structure and processes. Ecol
Appl 15:1135–1147
Rutherford JE, Mellow RJ (1994) The effects of an abandoned roast
yard on the fish and macroinvertebrate communities of sur-
rounding beaver ponds. Hydrobiology 294:219–228
Scher O, Thie
`
ry A (2005) Odonata, amphibia and environmental
characteristics in motorway stormwater retention ponds (South-
ern France) Hydrobiology 551:237–251
Scheuhammer AM, McNicol DK, Mallory ML, Kerekes JJ (1997)
Relationships between lake chemistry and calcium and trace
metal concentrations of aquatic invertebrates eaten by breeding
insectivorous waterfowl. Environ Pollut 96:235–241
Schroer AFW, Belgers JDM, Brock TCM, Matser AM, Maund SJ,
Van den Brink PJ (2004) Comparison of laboratory single
species and field population-level effects of the pyrethroid
insecticide k-cyhalothrin on freshwater invertebrates. Arch
Environ Contam Toxicol 46:324–335
Sloof W (1983) Benthic macroinvertebrates and water quality assess-
ment: some toxicological considerations. Aquat Toxicol 4:73–82
Sparling DW, West S, Ortiz-Santaliestra M (2006) Effects of lead-
contaminated sediment on Rana sphenocephala tadpoles. Arch
Environ Contam Toxicol 51:458–466
Suzuki KT, Sunaga H, Aoki Y et al (1988) Binding of cadmium and
copper in the mayfly Baetis thermicus larvae that inhabit a river
polluted with heavy metals. Comp Biochem Physiol C 91:487–
492
Suzuki KT, Sunaga H, Hatakeyama S, Sumi Y, Suzuki T (1989)
Differential binding of cadmium and copper to the same protein
in a heavy metal tolerant species of mayfly Baetis thermicus
larvae. Comp Biochem Physiol C 94:99–104
Tennessen K (1993) The common, remarkable Lydia. Argia 5:16–18
US EPA (United States Environmental Protection Agency) (1994)
Method 3051: microwave assisted acid digestion of sediments,
sludges, soils, and oils. SW-846, Test methods for evaluating
solid waste. US EPA, Washington, DC. Available from
www.epa.gov/epaoswer/hazwaste/test/sw846.htm
US EPA (United States Environmental Protection Agency) (2005)
Current national recommended water quality criteria. Availabe
from www.epa.gov/waterscience/criteria/wqcriteria.html
Warnick SL, Bell HL (1969) The acute toxicity of some heavy metals
to different species of aquatic insects. J WPCF 41:280–283
84 Arch Environ Contam Toxicol (2009) 56:77–84
123