AVOIDANCE AND REPRODUCTION TESTS WITH THE PREDATORY MITE HYPOASPIS
ACULEIFER: EFFECTS OF DIFFERENT CHEMICAL SUBSTANCES
OLUGBENGA J. OWOJORI,*yz KAROLINA WASZAK,yand JOERG ROEMBKEy
yECT Oekotoxikologie GmbH, Floersheim, Germany
zDepartment of Zoology, Obafemi Awolowo University, Ile-Ife, Nigeria
(Submitted 15 May 2013; Returned for Revision 15 August 2013; Accepted 4 October 2013)
Abstract: Few toxicity data exist in the literature on the toxicity of chemicals to the predatory mite Hypoaspis aculeifer, but no
information is available on its avoidance response. To assess the relevance of the avoidance behavior of H. aculeifer and the relative
sensitivity of the mite in comparison with other invertebrates, avoidance and reproduction tests were conducted with 7 chemicals using
standardized guidelines. The chemicals (deltamethrin, chloropyrifos, dimethoate, Cu, NaCl, phenanthrene, and boric acid) were selected
so as to cover varying chemical classes. For all 3 pesticides tested, avoidance response showed lower sensitivity than reproduction and
survival (avoidance median effective concentration [EC50] >reproduction EC50/median lethal concentration [LC50] values). However,
for Cu, NaCl, and phenanthrene, the avoidance response showed similar sensitivity as reproduction (avoidance EC50 reproduction
EC50 values), whereas for boric acid, similar sensitivity as survival (avoidance EC50 LC50 values). Although the mite H. aculeifer
appears less sensitive to some of the chemicals tested than most other soil invertebrates, its status as the only predator among organisms for
which standardized tests are available affirms its inclusion in routine ecotoxicity assessment. The results of the avoidance test with
H. aculeifer suggest its potential usefulness as a rapid screening test for risk assessment purposes. Environ Toxicol Chem 2014;33:230–
237. #2013 SETAC
Keywords: Metals Organic chemicals Avoidance Pesticides Laboratory
INTRODUCTION
Soil pollution is a worldwide problem that needs to be
properly addressed with robust risk assessment strategies, for
example, following a tiered approach. For a proper risk
assessment, a battery of tests is desirable; therefore, the
development of new test methods covering wide arrays of soil
organism groups and useful at different tiers is often advocated
[1]. To obtain a balanced battery of tests, several factors,
including feasibility of the test and its ecological relevance, are
issues of consideration [2]. Apart from this, organisms chosen
should include those that are representatives of the ecosystem to
protect; therefore, organisms with different life histories,
representing different functional groups, different taxonomic
groups, and routes of exposure are necessary [2,3]. Although
arthropods are among the most diverse soil organism groups
living in soil, the current set of available tests shows an
underrepresentation when compared with other groups such as
oligochaetes (mainly enchytraeids and earthworms [4,5]).
Among micro-arthropods, Collembola are the most commonly
used group, with standardized reproduction test protocols
available [6,7], but a reproduction test for the predatory mite
Hypoaspis aculeifer has already been standardized [8,9].
Standardized avoidance tests with earthworms [10] and
Collembola [11] were recently introduced for the risk assess-
ment of contaminants in soil. Avoidance response of soil
organisms is increasingly advocated as a tool for rapid
identification and assessment of contaminated soils. Although
test duration of 2 d is often used (e.g., Owojori and
Reinecke [12]) and also recommended for earthworms and
collembolans [10,11], sufficient evidence is available to suggest
a shorter duration [13–15]. Soil organisms avoid contaminants
because they possess chemoreceptors highly sensitive to
chemicals in their environment. Avoidance behavior often
occurs at lower concentrations than those affecting survival and
similar levels to those affecting reproduction and thus can be
used to predict an impending effect at individual and population
levels [16,17]. Avoidance behavior is also an ecologically
relevant measurement endpoint because it influences the energy
budget of the individual organism as well as ecosystem structure
because habitat functioning of the soil is directly impacted when
organisms leave the area [18].
Of the oligochaetes and arthropods for which standardized
guidelines are available for assessing effect on reproduction,
only the predatory mite H. aculeifer has never been assessed for
relevance of the avoidance test. This species has a global
distribution, is easy to handle, and because it is predaceous,
represents a higher trophic level than species in other
invertebrate groups (e.g., earthworm, Collembola, and enchy-
traeid) for which standardized tests are available. The standard-
ized protocol for the assessment of pollutants on its reproduction
was done with a ring test involving several laboratories based on
information with 2 chemicals—boric acid and dimethoate [8]—
suggesting the need for tests with more chemicals to ascertain its
sensitivity relative to other species. Mites are considered to have
intermediate sensitivity to pollutants in comparison with other
soil invertebrates [8], but this also depends on the pollutant in
question [19]. Therefore, a study evaluating the toxicity of
various contaminants and the avoidance response of the
predatory mite H. aculeifer is warranted.
The present study was therefore designed to investigate the
avoidance behavior and life-cycle response of the predatory mite
H. aculeifer to several chemical groups with varying or similar
modes of action. This was done with a view to assess the
* Address correspondence to gowojori@oauife.edu.ng.
Published online 9 October 2013 in Wiley Online Library
(wileyonlinelibrary.com).
DOI: 10.1002/etc.2421
Environmental Toxicology and Chemistry, Vol. 33, No. 1, pp. 230–237, 2014
#2013 SETAC
Printed in the USA
230
usefulness of its avoidance response as a tool for risk assessment
at a lower tier and its relative sensitivity in comparison with
other soil invertebrates. The following chemicals were select-
ed: 3 insecticides (deltamethrin, dimethoate, and chlorpyrifos),
1 metal (Cu), 1 polycyclic aromatic hydrocarbon (PAH;
phenanthrene), 1 salt compound (NaCl), and a reference
chemical (boric acid).
Deltamethrin is a type II pyrethroid, and it interferes with the
normal production and conduction of nerve signals in the
nervous system by delaying the closing of the activation gate for
the sodium ion channel [20]. It is a contact and systemic poison.
Dimethoate and chlorpyrifos are organophosphates, which are
known to inhibit cholinesterase, an enzyme important for the
proper functioning of the nervous system [21]. Copper was
chosen because it is a widespread contaminant used as a
fungicide, algicide, and herbicide. It is toxic because of its ability
to denature cellular proteins and deactivates enzyme systems. It
causes cellular stress to organisms because of reduced lysosomal
stability [22], and there is evidence of neurotoxicity to some
invertebrates [23]. Phenanthrene is one of the most common
PAHs and is considered a human health risk in soil. It causes
narcosis in invertebrates and alters the function of the cell
membrane, which can result to mortality in some cases [24].
Sodium chloride acts by causing imbalance in ionic and osmotic
regulation, thereby affecting water uptake in the organism [25].
Boric acid is the reference substance adopted for several
standardized tests [26]. It acts as a stomach poison and may
also have some toxic effects on the nervous system of
insects [27].
The specific goals of the present study were to 1) assess the
avoidance behavior of the predatory mite to the 7 contaminants
selected, 2) compare the results of the mite avoidance test with
mite survival and reproduction tests for these contaminants, 3)
assess the suitability of boric acid as a reference substance for the
predatory mite avoidance test, and 4) assess the relative
sensitivity of H. aculeifer in comparison with other soil
organisms commonly used in toxicity tests.
MATERIALS AND METHODS
Test organism
The test organism used in the present study is the predatory
mite species H. (Geolaelaps) aculeifer Canestrini (Acari:
Laelapidae). This species of predatory mite is a relevant
representative of soil fauna. It is distributed worldwide and can
easily be collected and reared in the laboratory. Information on
its biology and ecology and use in ecotoxicological testing is
available [28,29]. A standardized toxicity test for its assessment
of chemical is now available [9]. Specimens used for the present
study were taken from a culture kept in the ECT Oekotox-
ikologie GmbH laboratory (Floersheim, Germany). The original
culture was collected from the field in approximately 2005 and
was provided by Mitox. Specimens were cultured in the
laboratory, kept at 20 8C with a photoperiod of 16:8 h, light:dark
cycle in plastic containers lined with an 8:1 ratio of plaster of
Paris and activated charcoal substrate. The substrate was
moistened once a week with deionized water, and cheese mites
(Tyrophagus putrescentiae) were added ad libitum as a food
source.
Test soil
Experiments were done in artificial soil as per Organisation
for Economic Co-operation and Development (OECD) guide-
line 226 [25]. It consisted of 74.8% sand (Quarzwerke GmbH),
20% kaolin clay (Ziegler & Co GmbH), and 5% sphagnum peat
(Torfwerk Albert Thomann GmbH) by dry weight. The pH was
adjusted to 6.0 0.5 by adding 0.2% CaCO
3
. The maximum
water holding capacity of the OECD artificial soil was 45.8%.
The pH–CaCl
2
and water holding capacity of the test soils were
assessed according to OECD [5] and International Organization
for Standardization [30], respectively. Each soil treatment used
in the experiment was moistened so as to make 50% of the
maximum water holding capacity.
Test substances
Test chemicals used included 3 pesticides: deltamethrin
(added as Decis flussig, 2.8% active substance [a.s], Bayer Crop
Science GmbH), dimethoate (added as Bi 58, containing 400 g a.
s./L, Compo GmbH), and chlorpyrifos (added as Pestanal,
purity: 99.9%, CAS No: 2921-88-2, Sigma-Aldrich), as well as 1
metal (Cu; added as CuCl
2
.2H
2
O, CAS: 10125-13-0, Alfa Aesar
GmbH), 1 PAH (phenanthrene; C
14
H
10
, purity: 98%, Sigma-
Aldrich), 1 salt compound (NaCl; purity: 99.5%, Art No: 3957.1,
Carl Roth GmbH), and a reference toxicant (boric acid; H
3
BO
3
,
purity: 99.8%, Art No: 6943.2, Carl Roth GmbH). Selected
test concentrations for each chemical were based on nominal
values calculated from range finding tests or data available from
literature. For metals, pesticides (except chlorpyrifos), and boric
acid, the test chemicals were added as aqueous solution in
deionized water so as to make up 50% of the maximum water
holding capacity of the soil. Prepared soil treatments were
introduced into the test vessels and used immediately in all cases
except for Cu and boric acid, where the soils were allowed to
equilibrate for 3 d before they were used in the experiment. For
phenanthrene and chlorpyrifos, the compounds were introduced
to a portion (10% mass) of the soil, in acetone (1:2, v/w soil) as
carrier, and left in a fume hood for at least 2 h to allow the
acetone solvent to evaporate, after which the other portion (90%
of total soil) was added. The soil was then thoroughly mixed
with a hand mixer for approximately 10 min and subsequently
moistened to the required moisture level, and animals were
introduced immediately. In these cases when acetone was used
as solvent, apart from the water control, a solvent control
(control soil spiked with same amount of acetone used in the
treatment soils) was also added for the reproduction tests. For all
avoidance test treatments with chemicals delivered in the
acetone solvent carrier, an equivalent amount of acetone was
added to the soil in the nontest chemical-treated half of the
vessel. This ensured that no avoidance behavior could be caused
solely by the solvent.
Experimental procedures
Avoidance behavior of mites. For the assessment of mite
avoidance to pollutants, the 7 chemicals were tested individual-
ly, combined with a control soil in a 2-chambered avoidance test
using the modified earthworm/Collembola avoidance protocol
[10,11]. The cylindrical 150-mL glass beakers used (5.5 cm
diameter, 7.5 cm height) were divided into 2 sections by drawing
a line on the outside and labeling it with the name of the
corresponding treated and control soil. Using a piece of plastic
fitted transversally as a divider in the vessel, half of the vessel
was filled with treated soil, the other filled with control soil.
Dual-control tests were performed to ensure that no avoidance
behavior is detected when the control soil is used in both sides of
the test chamber. The volumes of soil used on each half of the
container were 10 g dry weight. Five replicates were used for
each treatment. After the soils were introduced, the plastic
dividers were removed, and 10 mites were placed on the surface
Avoidance and reproduction tests with Hypoaspis aculeifer Environ Toxicol Chem 33, 2014 231
at the dividing line between the 2 halves of each test soil. The
vessels were then closed tightly with paraffin. The tests were
carried out in a climate chamber at 20 18C with a light:dark
cycle of 16 h:8 h for all treatments. The animals were not fed
during the test. At the end of the test period of 48 h, the control
and the contaminated soil sections were carefully separated by
reinserting the plastic divider. The number of mites in each
section was counted after a 48-h extraction with a modified
Berlese-Tullgren apparatus.
Survival and reproduction of mites. Effects on survival and
reproduction were assessed according to procedures described in
OECD guideline 226 [9] for all 7 compounds except dimethoate
and boric acid because data were available in the literature for
these compounds [8]. Ten adult female mites between ages 28 d
and 35 d were introduced in each test vessel containing 20 g dry
weight soil. The mites were fed a small amount (tip of a spatula)
every 3 d with cheese mites. Moisture changes were assessed
weekly by weighing the containers, and the corresponding water
loss was replenished. Four replicates were used in all treatments,
except in tests in which a solvent control was not needed where 6
replicates were used for the control treatments. For each
treatment, 1 extra replicate for moisture and pH measurements
was kept aside until the end of the test. Testing was conducted
using the 16:8 h light/dark photoperiod cycle and temperature of
20 8C for 14 d. At the end of the test, the number of mites in each
container was counted after 48 h extraction with a modified
Berlese-Tullgren apparatus. Measurement endpoints included
number of surviving adults and juveniles produced after 14-d
exposure.
Statistics
All data were checked for normality and homogeneity of
variance with the Shapiro-Wilks W test and Levene’s test,
respectively. All statistical analyses were done using the
software R.
Avoidance tests. Results of avoidance tests are presented in
graphs in terms of average of net response expressed as a
percentage and calculated as follows
NR ¼CTðÞ=N100
where NR is the net response, Cis the number of mites observed
in the control soil, Tis the number of mites observed in test soil,
and Nis the total number of mites per replicate [10,11].
A positive net response (þ) indicates avoidance, and a
negative net response (–) indicates an attraction to the test soil. In
the avoidance test, confirmation of unbiasness that approxi-
mately 50% 10% of mites had chosen either side of the dual-
control chamber in the dual-control test was done with pairwise
comparison with 1-tailed Student’sttest. The avoidance median
effective concentration (EC50) values were estimated using the
trimmed Spearman Karber method. One-way analysis of
variance was used to test the effects of exposure on mite
avoidance response. When differences were observed, Tukey
post-hoc comparison was used to ascertain where the differences
lie. When the data failed the normality test (chlorpyrifos data),
the Kruskal-Wallis one-way analysis of variance on ranks was
performed.
Reproduction tests. Differences in survival and reproduction
of mites between water control and acetone control treatments
were checked with a Student’sttest. One-way analysis of
variance was used to test the effects of increasing exposure
concentrations on survival and reproduction of the mites. When
differences were observed, Tukey post hoc comparison was used
to ascertain where the differences lie. The median lethal
concentration (LC50) and reproduction EC50 values were
calculated using trimmed Spearman Karber and nonlinear
regression methods (logistic-3 parameter), respectively [31].
Significant differences between the avoidance EC50 values
for each compound and the reproduction EC50 values and the
LC50 values were ascertained when confidence limits do not
overlap.
RESULTS
Avoidance behavior of mites in exposure to 7 contaminants
The proportions of mites in the dual control (2 untreated
sides) for all of the compounds were in average within 60:40% as
specified in the International Organization for Standarization
guidelines [10,11]. Further pairwise comparison showed no
significant difference (p>0.05) in the number of mites in either
side in all of the dual control tests. Therefore, the validity
criterion requiring no significant difference in avoidance
response in dual control was achieved in all tests. Also, the
validity criterion of the collembolan avoidance test [11],
requiring at least 80% recovery, was adopted in the present
study. Overall, in all tests, the mean (standard error) number of
mites recovered was high at 89.7 3.5%, with highest mean
number recovered found for chlorpyrifos (93.7 1.5%) and
lowest (84 7.2%) found for dimethoate. In all of these tests,
the mean number of mites recovered was not significantly
affected by concentrations of the chemical except in the highest
concentration of dimethoate (10 mg/kg), where only 46% of
mites could be recovered. Apart from this, in only 1 test (5400
mg/kg NaCl) was the mean number of mites recovered (78%)
less than 80%. This was caused by an outlier (detected with a box
plot analysis), and when the outlier was removed, 85% of the
mites were recovered. Based on the validity criterion, only the
former case (10 mg/kg dimethoate) was excluded from the
analysis.
Among the 3 pesticides used, no significant mite avoidance
(p>0.05) of chemicals could be seen (Figure 1A–C). In the case
of deltamethrin, significant attraction (p<0.05) was seen at
32 mg/kg (Figure 1A). Converse to the case of the pesticides, a
clear dose–response curve was found for the mite avoidance of
Cu, NaCl, and phenanthrene, but it was less obvious for boric
acid (Figure 2A–D). In the case of Cu and phenanthrene, almost
complete avoidance was observed as from 2560 mg/kg and
100 mg/kg, respectively.
Survival and reproduction of mites in exposure to 5 toxicants
No significant changes occurred in soil pH because of
chemical spiking or test duration. The mean adult survival in all
control soils was 80% or greater, and mean juvenile production
was 100 or more, with a coefficient of variation of 30% or less.
Therefore, all validity criteria [9] were met in these tests. In the 2
cases in which extra acetone controls were used, the statistical
differences between the mite performance acetone and water
controls were evaluated. For chlorpyrifos, no significant
difference (p>0.05) was seen in the mean percentage of
survival and reproduction of mites in water and acetone control
groups; therefore, the data were pooled together. Similarly, for
phenanthrene, no significant difference was seen in survival
between those in water control and those in acetone control, and
the survival data were also pooled together. However, a
significantly higher number of juveniles (p<.05) were
produced in the water control than in the acetone control;
therefore, only the reproduction data for the acetone control was
232 Environ Toxicol Chem 33, 2014 O.J. Owojori et al.
compared with all other treatments in the phenanthrene
treatment.
The proportions of adults that survived and juveniles
produced when the mites were exposed to deltamethrin and
chlorpyrifos are shown in Figure 3A and B. For deltamethrin,
reproduction was a more sensitive parameter than survival,
because significantly (p<0.05) lower juvenile production was
found at 10 mg/kg, whereas that for survival was found at the
highest concentration (32 mg/kg). For chlorpyrifos, a steep
dose–response curve was found for survival and reproduction,
with effects (p<0.05) only found at the highest concentration
(10 mg/kg). For Cu, NaCl, and phenanthrene, a clear dose–
response curve was seen for reproduction but less so for survival
(Figure 4A–C). In all 3 cases, reproduction ceased at the highest
concentration used, and only few animals survived at these
concentrations. For Cu, significant effects (p<0.05) on survival
and reproduction was found at 5120 and 2560 mg/kg,
respectively, whereas for NaCl, significant effects (p<0.05)
on survival and reproduction was found at 31 492 mg/kg and
7920 mg/kg, respectively. For phenanthrene, a significant effect
(p<0.05) on survival and reproduction was found from 320 mg/
kg and 100 mg/kg, respectively
Comparison of avoidance and reproduction tests. The avoid-
ance EC50 values obtained for all of the compounds were
compared with the reproduction EC50 values and the LC50
values (Table 1). For the 3 pesticides (deltamethrin, chlorpyrifos,
and dimethoate), the avoidance EC50 values were higher than
the LC50 values. For boric acid, the avoidance EC50 was higher
than the reproduction EC50 values but similar to the LC50
values, whereas for Cu, NaCl, and phenanthrene, the avoidance
EC50 values estimated were lower or similar to the reproduction
EC50 values.
DISCUSSION
The avoidance behavior of the predatory mite H. aculeifer to
5 groups of chemicals (metal, PAH, salt, pesticide, and an acid),
some with varying, others with similar modes of action, were
tested in the present study. The mites showed similar sensitivity
in the avoidance and reproduction tests for 3 groups (metal-Cu,
PAH-phenanthrene, and salt-NaCl) of the chemicals tested and
similar sensitivity in the avoidance and survival tests for the
reference substance boric acid. However, lower sensitivity of the
avoidance test than reproduction and survival was found for all 3
pesticides tested in the present study. The modes of action of the
3 pesticides cause negative effects on the nervous system
[20,21]. This could have been responsible for the inability of the
mites to avoid them, because a mite might be paralyzed at
sublethal concentrations and might be unable to avoid the
chemicals at such concentrations. However, this cannot be
generalized because phenanthrene, which is a narcotic com-
pound, also has effects on the nervous system [24], but the mites
showed strong avoidance of phenanthrene. This may be an
indication that the lack of sensitivity of the predatory mite
avoidance test for the pesticides in the present study might not be
explained by the mode of action only.
A
Concentration of deltamethrin (mg/kg) soil
00.321 3.210 32
Net response (%)
-100
-50
0
50
100
C
Concentration of dimethoate (m
g
/k
g
) soil
02468
Net response (%)
-100
-50
0
50
100
B
Concentration of chlorpyrifos (mg/kg) soil
0 0.1 0.32 1 3.2 10
Net response (%)
-100
-50
0
50
100
*
Figure 1. Mean( standard error, n¼5) net response of Hypoaspis aculeifer after its exposure for 48 h in 2-chamber avoidance tests combining untreated soil with
deltamethrin- (A), chlorpyrifos- (B), and dimethoate- (C) treated soils. Positive values indicate avoidance, whereas negative values indicate attraction. Asterisks
indicate a significant difference at the 0.01 level when compared with control data. All concentrations are nominal values.
Avoidance and reproduction tests with Hypoaspis aculeifer Environ Toxicol Chem 33, 2014 233
Many soil invertebrates have been known to show strong
avoidance response to single and mixtures of toxicants.
Earthworms [17], collembola [32], oribatid mites [15], and
isopods [33] could avoid contaminants at concentrations similar
to that for which effects are found on reproduction or survival.
For enchytraeids, initial trials [34] with Enchytraeus albidus
indicate some promise, but later trials [35] showed lower
sensitivity than reproduction and with large data variation,
thereby discouraging their use in the context of risk assessment
and suggesting a need to consider other enchytraeid species. In
the present study, the predatory mite H. aculeifer has shown
more than moderate response to several chemicals. Although
more tests with more chemicals are needed, the data of the
present study show some promise.
Although the mite avoidance response was as sensitive as
reproduction for some of the chemicals (Cu, NaCl, and
phenanthrene), the avoidance test was less sensitive than
reproduction for other chemicals (deltamethrin, chlorpyrifos,
dimethoate, and boric acid). Nonavoidance or weak avoidance
observed for these chemicals and in fact other chemicals are not
peculiar to the predatory mite. The earthworm Eisenia fetida
showed no avoidance to organophosphate insecticides (diazinon
Concentration of deltamethrin (mg/kg) soil
00.321 3.210 32
# juvenile mites
0
100
200
300
0
20
40
60
80
100
120
Adult mite survival (%)
Concentration of chlorpyrifos (m
g
/k
g
) soil
WC AC 0.1 0.32 1 3.2 10
0
100
200
300
0
20
40
60
80
100
120
AB
Figure 3. Mean (standard error, n¼4) survival and reproduction of the predatory mites (Hypoaspis aculeifer) after exposure for 14 d in Organisation for
Economic Cooperation and Development soil with 5% organic matter content and spiked with deltamethrin (A) and chlorpyrifos (B). The bar chart represents
reproduction, left y-axis; the line graph represents survival, right y-axis. All concentrations are nominal values. WC ¼water control, AC ¼acetone control.
A
Concentration of copper (mg/kg) soil
0 320 640 1280 2560 5120
Net response (%)
-100
-50
0
50
100
C
Concentration of phenanthrene (mg/kg) soil
0 10 32 100 320 1000
Net response (%)
-100
-50
0
50
100
B
Concentration of NaCl (mg/kg) soil
0
3000
5400
9720
17496
31492
Net response (%)
-100
-50
0
50
100
D
Concentration of boric acid (mg/kg) soil
0 100 200 400 800 1600
Net response (%)
-100
-50
0
50
100
**
***
*
***
Figure 2. Mean( standard error, n¼5) net response of Hypoaspis aculeifer after its exposure for 48 h in 2-chamber avoidance tests combining untreated soil with
Cu- (A), NaCl- (B), phenanthrene- (C), and boric acid–(D) treated soils. Asterisks indicate significant difference at the 0.01 level when compared with control data.
All concentrations are nominal values. All concentrations are nominal values.
234 Environ Toxicol Chem 33, 2014 O.J. Owojori et al.
