Environ. Sci. Technol. 2005, 39, 7154-7157
Life Stage Specific Impact of Dimethoate on the Predatory Mite Hypoaspis aculeifer Canestrini (Gamasida: Laelapidae) L A R S - H E N R I K H E C K M A N N , †,‡ KRISTINE MARALDO,† AND P A U L H E N N I N G K R O G H * ,† National Environmental Research Institute, Department of Terrestrial Ecology, Vejlsøvej 25, P.O. Box 314, DK-8600 Silkeborg, Denmark
Toxicants may affect juveniles more than adults because of physiological and behavioral aspects. When developing toxicity tests, this issue is often not addressed, and the most sensitive end point may be ignored. The topic was raised by a European working group aiming to develop a standard toxicity test with the predatory mite, Hypoaspis (Geolaelaps) aculeifer Canestrini, as this species was included in the EU Guidance document for testing of pesticides. To assess whether the juvenile life stages are the most susceptible, we examined the acute toxicity of dimethoate on larvae, protonymphs, deutonymphs, males, and females of H. aculeifer. The mites were exposed to 0, 2, 4, and 6 mg dimethoate kg-1 for 7 days in an OECD artificial soil (5% organic matter). Total juvenile biomass, reproduction, mortality, and population growth rate (pgr) λ were assessed at the end of the test. A comparison of mortality ranked the sensitivity of the life stages: Larvae (LC50 ) 3.8 mg kg-1) > protonymphs (LC50 ) 5.3 mg kg-1) > males (LC50 ) 5.6 mg kg-1) > deutonymphs (LC50 ) 7.1 mg kg-1) > females (LC50 ) 7.6 mg kg-1). Effects on reproduction and pgr were significant at 2 mg dimethoate kg-1, with population decline starting at this concentration. Thus, a test system with H. aculeifer including reproduction as end point is a rational approach, as reproduction will encompass juvenile mortality, at least with respect to dimethoate. Moreover, we suggest that pgr should be included in chronic standard tests because of high ecological relevance and the feasibility of applying it.
Introduction European guidelines on risk assessment of pesticides have recently included Hypoaspis (Geolaelaps) aculeifer Canestrini as a test species, but at the same time they have indicated the need for further development of existing protocols (1, 2). This study is part of the research and development activities of a soil mite test working group including European institutions and businesses engaged in developing a standard toxicity test with the predatory mite H. aculeifer (end points survival and reproduction). The aim of the soil mite test working group is to satisfy the just-mentioned European need * Corresponding author phone: (45) 89201588; fax: (45) 89201413; e-mail:
[email protected]. † National Environmental Research Institute. ‡ Present address: The University of Reading, School of Animal and Microbial Sciences, Division of Zoology, P.O. Box 228, Reading, RG6 6AJ, United Kingdom. 7154
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 18, 2005
for soil toxicity tests. Currently, different important test parameters (e.g., life stage specific sensitivity and soil chemistry) are being tested by the members of the group. Among the soil mesofauna, predatory mites (Gamasida) play a key role in the soil ecosystem (3), especially in agriculture, where they may have a potential regulatory effect on the populations of plant parasitic nematodes (4). Traditionally, ecotoxicological tests predominantly expose adult individuals to determine the impact of a given toxicant on lethal and sublethal end points (5). This approach is reasonable since growth and reproduction are more sensitive end points than lethality (5). However, toxicants may affect juveniles more than adult life stages because of physiological (e.g., surface-volume ratio) and behavioral aspects (6-8). Moreover, size differences in species with sexual dimorphism may cause the smaller sex to be more vulnerable. Furthermore, timing of exposure may be very relevant to an organism, since all species have one or several points in their life cycle where they are more susceptible (e.g., during molting) to the effects of contamination (8, 9). In our opinion, determining the most sensitive life stage and end point is critical when developing a toxicity test. Ecotoxicological experiments studying effects at the population level often include the mortality of one or more juvenile life stages, for example, ref 10. However, this is rarely determined in a standard toxicity test, and the published literature on life stage sensitivity to given toxicants is in general very sparse regarding invertebrates, for example, refs 6-8. The objective of this study was to estimate the mortality of the life stages, larvae, protonymphs, deutonymphs, males, and females of H. aculeifer when exposed to the organophosphorous insecticide dimethoate. We hypothesized that larvae would be most sensitive to dimethoate because of their softer cuticle, greater surface-volume ratio, and metabolism, which overall could result in a higher uptake of insecticide. Furthermore, our aim was to determine the most sensitive end point by comparing total juvenile biomass, reproduction, mortality, and population growth rate.
Materials and Methods Test Species. Hypoaspis (Geolaelaps) aculeifer Canestrini is a good candidate representing the Gamasida as they do not overkill their prey nor exhibit cannibalism, unless during extreme starvation (11), and the species is easily cultured (12). The original mite culture came from Koppert Biological Systems, Rotterdam, The Netherlands, and was reared at the National Environmental Research Institute, Silkeborg, Denmark. Test Chemical. The organophosphorus insecticide and acaricide dimethoate (O,O-dimethyl S-(N-methylcarbomoyl methyl) phosphorodithioate) was selected for this study, as its effect is well documented in other tests with H. aculeifer (13-17), and it is used as a toxic reference compound. A commercial formulation of dimethoate was used for all experiments. A commercially formulated sample was obtained directly form the manufacturer (Cheminova Agro A/S, Lemvig, Denmark) containing 400 g L-1 of the active ingredient (ai) and the organic solvents xylene and cyclohexanone. Experimental Design. A total of five experiments were performed comprising the three juvenile stages and sexually mature males and females. The juvenile mites were synchronized using the method of Krogh (16), except adult mites that were collected directly from the stock culture. At the beginning of the experiments, the ages of the five life stages were larvae 1 ( 1 d, protonymphs 6 ( 1 d, deutonymphs 11 10.1021/es050130d CCC: $30.25
2005 American Chemical Society Published on Web 08/18/2005
TABLE 1. Lethal and Sublethal Effect Concentrations of Dimethoate on Hypoaspis aculeifer after 1 Week of Exposurea
larvae protonymphs deutonymphs males females a
mortality LC50 (mg kg-1)
biomass EC50 (mg kg-1)
3.8b (2.9-4.7) 5.3c (4.7-5.9) 7.3c (5.0-9.6) 5.6c (4.8-6.4) 7.5c (4.6-10.4)
3.0 (2.6-3.4) 3.7 (3.3-4.2) 9.2 (4.2-14.3)
reproduction EC50 (mg kg-1)
% mortality at 6 mg kg-1
1.6 (1.2-2.0)
96b (90-102) 73c (60-86) 41e (25-56) 64c,d (49-79) 44d,e (30-58)
Values in brackets are the 95% confidence limits. Different superscript letters signify a significant difference (P < 0.05).
( 1 d, males >20 d, and females >20 d, respectively. The added larvae, protonymphs, and deutonymphs consisted of an unidentified mixture of males and females. Mite taxonomist curator Peter Gjelstrup at The Natural History Museum, Aarhus, Denmark, examined and verified the species and initial life stages. All experiments consisted of a control and three concentrations of dimethoate, and each treatment level was replicated three to five times. The applied concentrations of dimethoate were prepared in an increasing arithmetic series (i.e., 2, 4, and 6 mg kg-1 dry weight, DW). The soil substrate was a 5% organic matter (OM) OECD artificial soil (18) consisting of 5% Sphagnum-peat, 21% caoline clay, and 74% quartz sand. The initial water content was 20% corresponding to approximately 50% of the water-holding capacity (WHC). Precalculated amounts of dimethoate were mixed with demineralized water, while controls received demineralized water. An amount of 30 g soil wet weight (WW) was placed in microcosms (diameter 6.0 cm, height 5.5 cm), and 10 mites were added within 90-180 min after application of dimethoate. Two hundred small springtails Folsomia candida Willem (4 ( 4 d) were added as prey prior to the addition of the mites. Microcosms were kept in a climate room (constant temperature 20 ( 1 °C) with a 12:12 h light:dark regime. After an incubation period of 7 days, the mites were extracted by a modified MacFadyen-type high-gradient extractor similar to the equipment described by Petersen (19). The total duration of extraction was 48 h. Initially, the temperature was set at 25 °C and for every 12 h the temperature was automatically raised by 5 °C until reaching a maximum temperature of 40 °C. The mites were collected in plastic cups containing a saturated solution of benzoic acid (3%). The extractor was further equipped with controlled cooling (2-3 °C) of the collecting cups ensuring the preservation of the collected mites. After extraction, the mites were kept cool (5 °C) until counted manually under a stereomicroscope. Subsequently, mites were preserved in 70% ethanol. The mean individual body mass (µg DW) of each juvenile stage was estimated by pooling all the individuals from a given treatment (e.g., larvae exposed to 2 mg dimethoate kg-1). Ten mites were then randomly selected from the pooled samples (4 concentrations by 3 stages ) 12 pooled samples) and were dried at 60 °C for 24 h. Subsequently, the mites were weighed on a Sartorius Micro SC 2 balance (Sartorius AG, Goettingen, Germany). Data Analysis and Statistical Methods. The total biomass of the juvenile stages was calculated by multiplying the mean individual body mass by the number of surviving mites. Population growth rate (pgr) was estimated by calculating the dominant eigenvalue of the population projection matrix (20). The dominant eigenvalue or λ is the finite rate of increase, where a value of 1 signifies a constant pgr, while values above and below 1 signify an increase and decrease in pgr, respectively. Population growth rates were calculated by the Microsoft Excel software add-in PopTools version 2.6.4 (www.cse.csiro.au/poptools). Prior to statistical analysis, data were normalized to the mean control value of λ. Statistical analyses for estimation of concentrations causing 50% mortality and an effect of 50% on the biomass
FIGURE 1. Mortality rates of different life stages of Hypoaspis aculeifer after 1 week of exposure to dimethoate. The mean mortality of the different life stages is shown as plots. and reproduction, that is, LC50 and EC50, and point estimates were performed with the SAS/STAT version 8.02 procedure NLMIXED (21). Ninety-five percent confidence limits were computed by this procedure by approximate standard errors for the estimates using the delta method, and these standard errors were used to compute corresponding confidence limits (21). Nonlinear modeling was used to estimate doseresponse relationships by fitting the binomially distributed mortality data to the mortality rate (m) formula:
m ) c + (1 - c) Φ(a + bd)
(1)
where c is control mortality rate, Φ (phi) is the cumulative normal probability function, a slides the curve along the x axis, b determines the slope, and d is the mg kg-1 concentration of dimethoate in soil. Inferences were made by calculating contrasts between the estimates. A one-way ANOVA test was performed in Enterprise Guide Version 1 (SAS Institute Inc, Cary, NC) to test for significant differences between the population growth rates of the different treatments. For all tests, a significant level of 5% was applied.
Results Mortality. Mortality was the main end point of the study, as it was possible to compare the specific sensitivity of the different life stages all together. Larvae had a significantly lower LC50-value than the other life stages (P < 0.05, Table 1). At 6 mg dimethoate kg-1, the mortality of protonymphs and males was significantly higher (P < 0.05) compared with deutonymphs and females. However, male mortality was only weakly significantly different (P ) 0.061) from female mortality (Table 1). Contrasting mortality at 6 mg dimethoate kg-1 clustered the five life stages significantly (P < 0.05) into threegroups: larvae,protonymphsandmales,anddeutonymphs and females (Table 1, Figure 1). Sublethal Effects. During the 1-week exposure, reproduction resulted in approximately one larva per female in the controls. Reproduction was significantly (P < 0.05) lower VOL. 39, NO. 18, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7155
FIGURE 2. Population growth rate (λ) of Hypoaspis aculeifer after 1 week of exposure to dimethoate (mean + 95% C. L.). Asterisks * signify a significant difference (P < 0.05) between treatment and control. in all the exposed treatments compared with the control, ranging between 0.32 and 0.48 larvae per female (data not shown). Reproduction was the most sensitive end point, with an EC50 of 1.6 mg dimethoate kg-1 (Table 1). Furthermore, the total biomass of larvae, protonymphs, and deutonymphs was estimated. This end point was the product of the mean individual body mass of a juvenile stage at a given concentration multiplied by the corresponding number of surviving mites (Table 1). The mean individual body mass of larvae, protonymphs, and deutonymphs ranged between 1.0 and 3.0, 2.1 and 3.5, and 8.2 and 9.7 µg DW, respectively. The mean body mass of the larvae and protonymphs were negatively correlated with the concentration of dimethoate, while the mean body mass of deutonymphs showed a slightly positive correlation (data not shown). Population Growth Rate. The population growth rate λ was significantly (P < 0.05) lower in all the exposed treatments compared with the controls. The mean pgr of the control individuals was stable, that is, λ ) 1, whereas population decline (λ < 1) was evident starting at the lowest concentration of 2 mg dimethoate kg-1 (Figure 2).
Discussion This study demonstrated that the acute toxicity of dimethoate showed a positive correlation between stage (proportional to age) and survival. The larval stage was the most sensitive life stage, whereas males and protonymphs were more sensitive than females or deutonymphs at 6 mg dimethoate kg-1. Reproduction and population growth rate (pgr) were the most sensitive end points. Furthermore, the estimated pgr, which includes the impact on survival of juvenile stages and reproduction, revealed population decline starting at 2 mg dimethoate kg-1. The different sensitivity between life stages may be explained jointly by stage-specific surface-volume ratio (4, 5), behavior (16), and the structural complexity of the cuticle (22), which thickens with age in H. aculeifer (23). An interesting finding was that deutonymphs were more tolerant than males. Deutonymphs and males have approximately the same size and may have similar metabolic activity. Males may have a high level of activity because they are sexually mature, but female deutonymphs have both a higher growth rate and consumption rate than male mites from an age of about 10 days (16). The exposed deutonymphs were a mixture of future males and females. However, this would hardly be the sole explanation for the dissimilarity in sensitivity of males and deutonymphs. Females were more tolerant than males, 7156
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 18, 2005
which may be a reflection of sexual dimorphism as males are smaller than females (10-15 vs 50-60 µg DW, respectively) (15). In retrospective, it would have been interesting to investigate egg tolerance, as other researchers report this life stage to have a variable tolerance (6, 8, 24). The LC50 found herein is similar to the results of Krogh and Axelsen (14). They report an adult LC50 above 4 mg dimethoate kg-1 in a LUFA 2.2 soil, which is comparable to the 5% OM OECD artificial soil regarding humus, pH, and WHC. In a natural sandy loam, with less OM, the LC50 of adult H. aculeifer was approximately 0.9 mg dimethoate kg-1 (16). In a previous study, we found the LC50 of protonymphs to be 3.3 mg dimethoate kg-1 in a 2.5% OM OECD soil (unpublished results), indicating that the toxicity of dimethoate increases with decreasing OM (25). The EC50 for reproduction was likewise within previous reported values for dimethoate on H. aculeifer being 0.87 mg kg-1 in sandy loam (16) and 2.7 mg kg-1 in LUFA 2.2 (14). No studies were directly comparable to the impact of dimethoate on juvenile total biomass. However, Folker-Hansen et al. (17) report EC10 on growth rate of females to be 0.59 mg dimethoate kg-1. Surprisingly, the EC50 of deutonymphs was higher than LC50, which could be because the smallest of the initial deutonymphs died first, thus artificially increasing the mean body mass. Population growth rate is an important summary parameter, however, it is specific to the temporal and spatial context in which it is measured (26). Therefore, it might seem ambitious that we estimated the impact of an acute insecticide exposure on population dynamics. However, we found it interesting to assess how a short-term exposure did in fact influence the population level. Our results revealed that population decline started at 2 mg dimethoate kg-1, caused by the high impact on both reproduction and survival of especially larvae. However, as this was a short-term exposure, a field population would probably recover within months unless recolonization would be impaired by factors such as low migration, interspecific competition, and climate. Comparing the survival of the different life stages with the other end points revealed that reproduction and pgr were the most sensitive end points. An important issue when designing an ecotoxicological test system is to ensure that the most sensitive and ecological relevant end point or life stage is covered by the test. Reproduction covers a range of physiological and behavioral aspects such as, for example, fertility and juvenile mortality. Thus, a soil toxicity test based on H. aculeifer reproduction and survival of adults will encompass juvenile mortality, at least with respect to dimethoate. However, Forbes and Calow (10) conclude that pgr is a better measure of responses to toxicants than are individual-level effects, because pgr integrates potentially complex interactions among life-history traits and provides a more relevant measure of ecological impact. Thus, including pgr in a chronic standard test would not only, as in this case, add a sensitive end point, it would also improve the overall ecological relevance of the test. In this study, a demographic approach was used, that is, estimating reproduction and survival rates of several life stages. However, in a standard soil test, the approach can only be partially demographic, since the life stage specific survival rates cannot be estimated individually. However, the natural rate of increase, r, is a direct measure of pgr and, like a demographic approach, it integrates survivorship and fecundity. The natural rate of increase is calculated by the following equation:
r ) loge(Nf/No)/∆T
(2)
where Nf is the final number of individuals, No is the initial number of female individuals (as only females count), and ∆T is the change in time (number of days the experiment
was run). Positive values of r indicate a growing population, r ) 0 indicates a stable population, and negative r values indicate a population in decline and headed toward extinction. Although this is not really demography, the approach gives a measure of population growth and has been used by several authors (reviewed in ref 27). Thus, implementing r in future and present chronic toxicity tests would be straightforward as Nf and No are already being estimated.
Acknowledgments We would like to acknowledge the soil mite test working group for positive feedback and valuable discussions and Dr. Peter Gjelstrup for verifying the different life stages of Hypoaspis aculeifer. Moreover, we would like to thank research technicians Karsten T. Andersen, Zdenek Gavor, Elin Jørgensen, and Mette Thomsen for experimental assistance and Richard M. Sibly and three anonymous reviewers for valuable comments on the manuscript. This research was supported by the Department of Terrestrial Ecology, National Environmental Research Institute, Denmark.
Literature Cited (1) European Commission 2002 Guidance Document on Terrestrial Ecotoxicology Under Council Directive 91/414/EEC. http:// europa.eu.int/comm/food/fs/ph_ps/pro/wrkdoc/wrkdoc09_ en.pdf (accessed August 10, 2005). (2) EPPO 2003. Environmental risk assessment scheme for plant protection products. Chapter 8: Soil organisms and functions. EPPO Bull. 33: 195-209. (3) Larink, O. Springtails and mites: Important knots in the food web of soils. In Fauna in Soil Ecosystems; Benckiser, G., Ed.; Marcel Dekker: New York, 1997; pp 225-264. (4) Sharma, R. D. Studies on plant parasitic nematode Tylenchorhynchus dubius. Dissertation, Mededelingen Landbouwhogeschool Wageningen, 1971, 71, 154 pp. (5) Walker, C. H.; Hopkin, S. P.; Sibly, R. M.; Peakall, D. B. Principles of Ecotoxicology; Taylor & Francis: London, 1996. (6) Green, D. W. J.; Williams, K. A.; Pascoe, D. The acute and chronic toxicity of cadmium to different life-history stages of the freshwater crustacean Asellus aquaticus (L). Arch. Environ. Contam. Toxicol. 1986, 15, 465-471. (7) McCahon, C. P.; Pascoe, D. Use of Gammarus pulex (L) in safety evaluation tests - Culture and selection of a sensitive life stage. Environ. Toxicol. Chem. 1988, 15, 245-252. (8) Stark, J. D.; Banken, J. A. O. Importance of population structure at the time of toxicant exposure. Ecotoxicol. Environ. Saf. 1999, 42, 282-287. (9) Heckmann, L. H.; Friberg, N. Macroinvertebrate community response to pulse exposure with the insecticide lambdacyhalothrin using in-stream mesocosms. Environ. Toxicol. Chem. 2005, 24, 582-590. (10) Forbes, V. E.; Calow, P. Is the per capita rate of increase a good measure of population-level effects in ecotoxicology? Environ. Toxicol. Chem. 1999, 18, 1544-1556. (11) Usher, M. B.; Davis, P. R. The biology of Hypoaspis aculeifer (Canestrini) (Mesostigmata) - Is there a tendency towards social
behaviour? Acarologia 1983, 24, 243-253. (12) Schlosser, H. J.; Riepert, F. Entwicklung eines pru¨fverfahrens fu ¨ r chemikalien an bodenraubmilben (Gamasina) - Teil 1: Biologie der bodenraubmilbe Hypoaspis aculeifer Canestrini, (Gamasina) unter laborbedingungen. Zool. Beitr. 1992, 34, 395412. (13) Bakker, F.; van Stratum, P. Testing Hypoaspis aculeifer (Acari: Laelapidae) on standard soils: Notes on testing methodology and species sensitivity. IOBC/wprs Bull. 2003, 26, 99-105. (14) Krogh, P. H.; Axelsen, J. A. Test on the predatory mite Hypoaspis aculeifer preying on the collombolan Folsomia fimetaria. In Handbook of Soil Invertebrate Toxicity Tests; Løkke, H., van Gestel, C. A. M., Eds.; John Wiley & Sons: New York, 1998; pp 239-251. (15) Hamers, T., Krogh, P. H. Predator-prey relationships in a twospecies toxicity test system. Ecotoxicol. Environ. Saf. 1997, 37, 203-212. (16) Krogh, P. H. Effects of pesticides on the reproduction of Hypoaspis aculeifer (Gamasida: Laelapidae) in the laboratory. Acta Zool. Fennica 1995, 196, 333-337. (17) Folker-Hansen, P.; Krogh, P. H.; Holmstrup, M. Effect of dimethoate on the body growth of representatives of the soil living mesofauna. Ecotoxicol. Environ. Saf. 1996, 33, 207-216. (18) OECD Guideline for Testing of Chemicals 207. Earthworm, Acute Toxicity Tests: Paris, France, 1984; 9 pp. (19) Petersen, H. Some properties of two high gradient extractors for soil microarthropods. Natura Jutlandica 1978, 20, 95-121. (20) Caswell, H. Matrix population models; Sinauer Associates: Sunderland, MA, 1989. (21) SAS Institute Inc. SAS/STAT User’s Guide, Version 8; SAS Institute Inc.: Cary, NC, 1999. (22) Tang, J. X.; Siegfried, B. D. Bioconcentration and uptake of a pyrethroid and organophosphate insecticide by selected aquatic insects. Bull. Environ. Contam. Toxicol. 1996, 57, 993-998. (23) Karg, W. Acari (Acarina), Milben Parasitiformes (Anactinochaeta), Cohors Gamasina Leach, Raubmilben, Die Tierwelt Deutschlands, 59. Teil; Gustav Fischer Verlag: Jena, Germany: 1993; pp 1151. (24) Kjær, C.; Elmegaard, N.; Axelsen, J. A.; Andersen, P. N.; Seidelin, N. The impact of pheology, exposure and instar susceptibility on insecticide effects on a chrysomelid beetle population. Pestic. Sci. 1998, 52, 361-371. (25) Martikainen, E. A. T.; Krogh, P. H. Effects of soil organic matter content and temperature on toxicity of dimethoate to Folsomia fimetaria (Collembola: Isotomiidae). Environ. Toxicol. Chem. 1999, 18, 865-872. (26) Sibly, R. M.; Hone, J. Population growth rate and its determinants: an overview. Philos. Trans. R. Soc. London, Ser. B 2002, 357, 1153-1170. (27) Stark, J. D.; Banks, J. E. Population-level effects of pesticides and other toxicants on arthropods. Annu. Rev. Entomol. 2003, 48, 505-519.
Received for review January 19, 2005. Revised manuscript received May 31, 2005. Accepted July 18, 2005. ES050130D
VOL. 39, NO. 18, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7157
