Pesticides Research, 83 – The Effect of Esfenvalerate and Prochloraz on Amphibians with special reference to Xenopus laevis and Bombina bombina

The Effect of Esfenvalerate and Prochloraz on Amphibians with special reference to Xenopus laevis and Bombina bombina

4 Discussions

4.1 The use of amphibians as an ecotoxicological test organism
   4.1.1 The effects of aquatic contaminants on amphibians compared with
   the effects of other aquatic test organisms
   4.1.2 The effects of pesticides on different kinds of amphibians
4.2 A standardised toxicity test using embryos of amphibians
   4.2.1 Toxicity tests using embryos of amphibians
   4.2.2 Influence of the test protocols and the developmental stages of
   the amphibians on the ecotoxicological results
   4.2.3 Why introduce a new test with Bombina bombina
   4.2.4 Facilities for maintaining adults laboratory populations of Xenopus
   and Bombina
   4.2.5 Principles and methods for breeding amphibians with special focus
   on Bombina bombina
   4.2.6 Toxicity test using embryos of Xenopus laevis and Bombina
   bombina
   4.2.7 The effects of esfenvalerate on embryos of amphibians with
   special reference to Xenopus and Bombina
   4.2.8 In vivo observations
   4.2.9 Recordings of malformation, growth inhibition, mortality and
   teratogenic screening
4.3 The effects of prochloraz on embryos of amphibians with special reference to Xenopus and Bombina
   4.3.1 In vivo observations
   4.3.2 Recordings of malformation, growth inhibition and mortality at the
   end of the test
4.4 Comparison of the effects of esfenvalerate and prochloraz on amphibians with the results from the OECD standard tests
4.5 Conclusion

4.1 The use of amphibians as an ecotoxicological test organism

Due to their thin and permeable skins, and prolonged exposure first to the aquatic environment and then to the terrestrial, amphibians may be particularly sensitive to environmental contaminants. Populations of many amphibians have declined and some species have disappeared from certain regions around the world, a phenomenon which appears to have accelerated during the last years. Agricultural pesticides may contribute to the decline in amphibian populations (Phillips, 1990; Berrill et al. 1994). Much of the amphibian life cycle occurs in pons, streams, and temporary pools that are often associated with agricultural areas receiving pesticide applications. In addition, breeding and larval development of amphibians occur in spring and summer at the same time that heavy application of pesticides on agricultural lands occurs. Thus, the purpose of this part of the project is to provide a broader knowledge in handling the problems concerning the effects of pesticides on the ecology in small ponds in intensively farmed agricultural land. It is the idea to collect knowledge of amphibians' tolerance of pesticides from the literature and develop a new laboratory test system with Bombina bombina (test guideline development) for future testing of pesticides on amphibians living in Europe. The new test system is evaluated using esfenvalerate and prochloraz, two pesticides with different action and physico-chemical properties.

4.1.1 The effects of aquatic contaminants on amphibians compared with the effects of other aquatic test organisms

The toxicity of 11 organic compounds considered hazardous to water resources was evaluated using embryo-larva stages of up to eight fish and amphibian species. The animal test species exhibited varying degrees of sensitivity to the selected toxicant. For some test chemicals, the most sensitive species was amphibians, in other cases the fish was the most sensitive species. In most instances, higher LC₅₀ values were obtained in tests with B. fowleri (fowler’s toad), R. palustris (pickerel frog), Xenopus laevis (African clawed frog), and the fathead minnow. The species which exhibited the largest susceptibility to organic compounds were the rainbow trout, R. pipiens (leopard frog), and the European common frog R.temporaria (Black and Wesley, 1982). Howe et al. (1989) investigated the effect of atrazine and alachlor to the Northern leopard frog and American toads and compared their sensitivity with rainbow trout and channel catfish. Overall, rainbow trout and channel catfish appeared to be less sensitive than amphibian larvae. However, in some specific tests with atrazine or tests with early-stage larvae, the toxicity was roughly similar. The 96 h LC₅₀ values for later-stage amphibian larvae were approximately two to four times lower than those observed in fish.

This stresses the importance of using different test organisms representing various trophic levels, taxa, and functions in the environment in harzard and risk assessment.

4.1.2 The effects of pesticides on different kinds of amphibians

Berrill et al. (1993) found differences in sensitivity among five species of amphibian embryos and tadpoles (the frogs Rena sylvatica, Rena pipiens, Rena clamitans, the toad Bufo americanus, the salamander Ambystoma madeculatum) exposed to permethrin and efenvalerate at concentrations between 10 and 200 µg/l. Even though no significant mortality of embryos, anuran tadpoles, or salamander larvae occurred either during or after exposure to these two pyrethroids, growth was delayed after exposure. Furthermore, newly hatched R. clamitans tadpoles recovered slowlier than R. pipiens following exposure to low levels of both pyrethroids. Older tadpoles of B. americanus were also more sensitive than R. pipiens tadpoles at approximately the same stage. Larvae of the salamander A. maculatum, the most sensitive of the organisms tested, either recovered far more slowly or failed to recover at all. This indicates that different amphibian species have different tolerance to pesticides.

Black and Wesley (1982) tried to determine whether species tolerance in different amphibian species varies in a predictable manner with ecological adaptations or with other criteria which could be applied in extrapolating test results to natural aquatic ecosystems. The evaluated amphibian species were Ambystoma gracile (north-western salamander), Rena pipiens (leopard frog), Rena temporaria (European common frog), Rena palustris (pickerel frog), Bufo fowleri (fowler's toad), and Xenopus laevis (African clawed frog). These species were selected to represent different patterns of reproduction and variations in ecological habitat and geographical distribution to determine whether such factors correlate with susceptibility to organic toxicants. The toxicity of 11 organic compounds and two metals was evaluated using embryo-larva stages of the different species. The animals exhibited varying degrees of sensitivity to the selected toxicants. The most sensitive amphibian species were generally those which normally are restricted to aquatic or moist terrestrial habitats, whereas the most tolerant amphibians included those semi-aquatic and terrestrial species which appear to be more broadly adapted ecologically. However, Xenopus laevis (the African clawed frog) was an exception, since the results revealed that this aquatic organism was generally less sensitive to organic chemicals than other amphibian species. This was surprising and may be of some concern since Xenopus laevis is the test organism used in the only internationally recognised ecotoxicological test with amphibians.

4.2 A standardised toxicity test using embryos of amphibians

4.2.1 Toxicity tests using embryos of amphibians

FETAX ^[2] (Frog Embryo Teratogenesis Assay-Xenopus) is a standard methodology for toxicity tests of chemicals using embryos of amphibians and is designed to use embryos of the South African clawed frog Xenopus laevis. Although the FETAX toxicity test is designed explicitly for the use of Xenopus laevis, the procedure has been found useful for conducting developmental toxicity tests with other species of frogs, however, with some necessary modifications.

The procedure is applicable to all chemicals either individually or in formulations, commercial products or mixtures. With appropriate modification these procedures can also be used to conduct tests on temperature, dissolved oxygen, pH, physical agents and on materials such as aqueous extracts of water-insoluble materials and sediment.

Methodologies for the evaluation of amphibian toxicity have normally been based on mortality, malformation and growth inhibition.

Teratogenic screening

The amphibian embryo has been and remains a classical model for experimental embryological studies. It is an intact developing system that undergoes events as cleavage, gastrulation, morphogenesis and organogenesis, comparable with those of other vertebrates. We therefore incorporated this aspect into the present study.

A developmental toxicant is a test material that affects any developmental process. Therefore, a developmental toxicant affects embryo mortality and malformation. It is important to measure developmental toxicity because embryo mortality, malformation, and growth inhibition can often occur at concentrations far less than those required to affect adult organisms. A teratogen is a test material that causes abnormal morphogenesis (malformation). The Teratogenic Index or TI is a measure of developmental hazard (Dumont et al. 1983).

FETAX is a potential rapid test for identifying developmental toxicant. Data may be extrapolated to other species including mammals. FETAX might be used to prioritise samples for further tests which use mammals. Validation studies using compounds with known mammalian or human developmental toxicity, or both, suggest that the predictive accuracy will approach or exceed 85% (Dawson and Bantle, J.A. 1987; Courchesne, and Bantle, 1985; Dumont et al. 1983; Sabourin et al. 1985, Sabourin and Faulk 1987). The accuracy rate compares favourably with other currently available “in vitro Teratogenesis screening” (Schuler et al. 1982; Greenberg, 1982; Kitchin et al. 1981).

4.2.2 Influence of the test protocols and the developmental stages of the amphibians on the ecotoxicological results

Sensitivity of different stages

For many organic compounds, amphibian embryos have proved to be considerably more sensitive than larvae of the same species (Black and Wesley, 1982). Considering the limited data available from chronic life-cycle studies and the low cost of short-term embryo-larval bioassays, such tests may provide a useful means of quantifying the toxicity of aquatic contaminants. In some cases, older and more developed embryos were, however, found to be more sensitive to toxicants. Thus, older embryos of R. clamitans were found to be more sensitive to pyrethroid exposure than younger, less developed embryos (Berrill et al., 1993). Ecotoxicity data on the Northern leopard frogs and the American toads exposed to atrazine and alachlor also indicated that late-stage amphibian larvae may be more sensitive to some herbicides than earlier stage larvae (Howe et al. 1998). In a comparison of the effect of a given pesticide on different amphibians, it is therefore important to take the stage of development of the amphibians into consideration. Thus, acute toxicity data presented only for early-stage larvae should be interpreted with caution because they may, in some cases, underestimate the hazards to later life stages.

Media

Two principally different kinds of media have been employed in different studies with amphibian embryos: A “synthetic” salt solution (FETAX medium) with known composition (Appendix 2) and a medium, normally composed of naturally filtrated water (water from pond, streams and temporary pools). Several ecotoxicity studies indicate the influence of organic content and medium formulation on the toxic effects of different chemicals on both aquatic and terrestrial organisms (Di Toro et al. 1991; Berhin et al. 1981; Larsen and Nilsson 1983). This has especially been shown for complexed agents and reactive chemicals. A reduction of bioavailability of fenvalerate, by sorption to particulate and dissolve organic matter was demonstrated by Strawbridge et al. (1992) and Coats et al. (1989). Thus, fenvalerate was 6 times less toxic to mosquito larvae at 50 mg/l dissolved humic acid compared with the toxicity without humic acid.

Temperatures

Furthermore, the effect of many pesticides e.g. pyrethroids may be temperature dependent. Thus, according to Coats et al. (1989) pyrethroids are usually more effective at colder temperatures. On the contrary, Materna et al. (1995) found that the effect of esfenvalerate on larval leopard frogs increased with increasing temperature (the mortality was higher at 22°C compared with that at 18°C). Even though there is a significant uncertainty whether the effect of a given pesticide increases or decreases with increasing temperature, it seems clear that the temperature may influence the effect concentration of a pesticide. This factor should be taken into consideration when evaluating the toxicity of a pesticide.

4.2.3 Why introduce a new test with Bombina bombina

Development of a new test guideline for ecotoxicological tests with Bombina will enable completion of the already recognised frog embryo teratogenesis assay with the South African clawed frog Xenopus laevis (FETAX test). This test has a number of deficiencies which make its use for ecotoxicological risk assessment in Denmark less suited.

Xenopus laevis represents a very special limited ecosystem in Africa. It is of course a disadvantage in the evaluation of pesticides effect on amphibians in Europe. The sensitivity of embryo and tadpole stages of a particular species may vary with temperature. Populations in Northern latitudes as well as spring-breeding species may therefore be affected in a different manner.
The standard FETAX test also involves removal of the jelly coat of the embryos with a cystein treatment which may influence the sensitivity of a toxicant on the embryos compared with normal embryos with a jelly coat.
It is generally assumed that the most sensitive amphibian species are generally those which are normally restricted to aquatic or moist terrestrial habitats. Even though Xenopus is restricted to the aquatic environment, some studies revealed, however, that Xenopus laevis is less sensitive to several organic chemicals compared with other amphibian species (Black and Wesley, 1982).

There is therefore a requirement to introduce a new test with amphibians from Europe for ecotoxicological risk assessment of pesticides. The fire-bellied toad is particularly in focus because it is essential to develop a new amphibian test with an easily breeding frog/toad which, like Bombina sp., is living in nature in Europe. Such a test method will make it possible to compare laboratory tests with field tests. This is important in order to carry out an ecotoxicological risk assessment in accordance with the EU Commission's new suggestion for approval of plant protection measures in "Uniform Principles". Especially when using pesticides near sensitive small amphibian biotopes, toleration criteria limits must be made to ensure that the amphibian populations will survive.

Bombina bombina, Bombina variegata and Bombina orientalis are animals belonging to the Discoglossidae family (Amphibia, Anura) which is spread out in all Europe and Asia. The two European fire-bellied toads B. bombina and B. varigata hybridize easily in nature and form fully viable and fertile hybride mutually (a natural phenomenon in Eastern Europe). So far studies have shown that wherever their ranges meet a zone of hybridisation is formed, separating areas inhabited by the pure forms. Such hybrid zones have already been discovered in Poland, Austria, Hungary and Czechoslovakia (Rafinska, 1991).

Therefore a new standardised toxicity test with Bombina bombina was developed. This type of frog is chosen due to the considerable geographical distribution. Another reason is that the Bombina bombina is relevant in Denmark, as their population are dynamically and ecologically well-tested in among others Funen's County. Furthermore, it has an aquatic lifestyle outside the hibernation season.

4.2.4 Facilities for maintaining adults laboratory populations of Xenopus and Bombina

The facilities for maintaining adults laboratory populations of the South African clawed frog Xenopus laevis are well-described in the FETAX test guideline and information regarding the basic biology and development of this species has been reported by Deuchar (1972, 1975). For long there has, however, been a need for a reliable method whereby other laboratory populations of anurans could be both maintained and perpetuated at modest costs and with minimal time expenditure. The availability of reliable methods for the breeding and maintenance of anurans in a time-efficient operation and at modest costs is important to biologists in many disciplines. The results from the present study showed that it was rather easy to keep adults laboratory populations of Bombina bombina. However in contrast to Xenopus which can be kept in large aquaria, Bombina requires a vivarium with both water and a land area with hiding and feeding place. It is possible to keep the animals active the whole year and to prevent an hibernation period by keeping a constant temperature, a photo period of 12 h light, and constant feeding. Furthermore, Bombina bombina can be bred year-round as demonstrated in the present study, a requirement which is absolute necessary for a test organism used for ecotoxicological tests. The only problem in keeping adults populations of Bombina is that these organisms only eat food (crickets, meal worms, small earthworms, flies, and other suitable invertebrates) that is moving. This means that the animals must be fed by hand frequently which is a very time consuming procedure and therefore a disadvantage in connection with ecotoxicological testing.

4.2.5 Principles and methods for breeding amphibians with special focus on Bombina bombina

Even though it is generally assumed to be very difficult to breed Bombina bombina under captive conditions (Wilkinson, 1994), the results from the present study revealed that it is possible even to breed Bombina bombina throughout the year and get a sufficient number of healthy embryos. The success of the present study in breeding and getting a large number of healthy embryos of Bombina bombina may be caused to application of the right amount of hormone. To ensure a sufficient amount of HCG hormone we used the same procedure as described for Xenopus and adjusted the amount of hormones according to the weight of Bombina. The amount of HCG injected depended on the time of year and the condition of the adults. However, the most important reason for the success of the present study in breeding and getting a large number of healthy embryos at any time we wanted may be ascribed to the injection of PMG to the females 2 days before the animals were injected with HCG in order to mature the oocyttes.

A large number of available embryos facilitate statistical analysis and allow the construction of concentrations-response curves with narrow confidence limits which is a fundamental issue in ecotoxicological testing.

With minor modifications the system should be suitable for many other anuran species.

In view of the apparent decline of the amphibian species in the nature, it would seem that the herpetologist could play a significant role in the conservation of this species by the proper maintenance and breeding of this species which may then be used for provision of toadlets for reintroduction programmes. The success of such activities would of course depend upon the retention and maintenance of suitable habitats in which reintroduction could take place.

4.2.6 Toxicity test using embryos of Xenopus laevis and Bombina bombina

The toxicity test with embryos of Xenopus was performed according to the procedure described in the ASTM standard. All tests were conducted at 24±0.5°C. Temperatures higher than 26°C may cause malformation whereas lower temperatures may prolong the test.

For Bombina bombina a slightly modification of the standard test was necessary due to the lower number of healthy embryos available from this species. Thus for each concentration, only 5 embryos was used instead of the 25 embryos normally used in the ASTM standard. Other test conditions and test procedure like temperature, medium, and the renewal procedure etc. followed the ASTM test guideline. We found it very important to make as few changes in the test design as possible, especially when the effect of different toxicants is compared in different species of amphibians. However, one significant change of the standard procedure was made in the test design with Bombina. According to the ASTM standard the jelly coat is removed from the embryos at the beginning of the test with a cystein treatment. In the present test with Bombina, the jelly coat is not removed. This deviation from the standard procedure was chosen because we were afraid that the cyctein treatment may influence the sensitivity of the embryos to a given toxicant. An important part of the present project was to link our laboratory test with the field investigations in Funen's Country (these investigations will be published in a separate report) and it was therefore important to maintain the jelly coat so that these experiments are more comparable with the natural conditions.

The standard exposure time for the FETAX test is 96 h and the attainment of stage 46 in the controls of Xenopus at 24°C. However, deviations from this standard exposure time was necessary in the tests with Bombina because the development of the embryos was slower under comparable experimental conditions. In Bombina the attainment of stage 46 of the controls occurred after 120 h at 24°C. This means that embryos of Bombina are exposed to the test substance for a longer period of time than in the standard FETAX test. This should be taken into consideration when the toxicity of a given test substance is evaluated and the effects concentrations are compared. However, in an evaluation and comparison of the effects of a given test substances on different amphibian species we found it more important to ensure that the embryos went through the same developmental stages, rather than to maintain the same time of exposure. To ensure that different embryos had passed through the same developmental stages is of course especially important in an evaluation of the developmental toxicity of a test material.

6-aminonicotinamide was used as a reference toxicant to evaluate the Bombina test design. For 6-aminonicotinamide, a mortality and malformation database for reference purpose exists. From this published data base for 6-aminonicotinamide, the 96 h LC₅₀ is 2500 mg/l (95% CI = 2350 to 2650) and the 96 h EC₅₀ for malformation is 5.5 mg/l (95% CI = 3.9 to 6.9), or a TI of 455 (Dawson et al. 1989). The results from the present study revealed that the effect of the positive control 6-amino-nicotinamide on Bombina bombina was comparable with the one found with Xenopus. Furthermore, the effect of 6-aminonicotinamide on both malformation and mortality was comparable with those described in the database for Xenopus. The most conspicuous malformations caused by 6-aminonicotinamide in Bombina were the same as those found in Xenopus. Thus, no significant difference was found in the effect of 6-aminonicotinamide on embryos of Bombina bombina and Xenopus laevis.

4.2.7 The effects of esfenvalerate on embryos of amphibians with special reference to Xenopus and Bombina

Esfenvalerate is an insecticide used against insect pest in agriculture, gardening, fruit farming and forestry. The compound is effective against several insect species. It is a pyrethroid insecticide which contains the ss-isomer of fenvalerate, the most toxic of the four stereoisomers. It is sold under the trade name Sumi Alfa which contains 5% a.i.

4.2.8 In vivo observations

Esfenvalerate affected the embryos gradually during the exposure period and a dose-response relationship was found in both Xenopus and Bombina. The initial response of embryos was a decrease in activity and the characteristic spasmodic twisting and at the highest concentrations some of the embryos were immobilised.

After 96 h, corresponding to the end of the test with Xenopus periodic spasmodic twisting was seen in most of the embryos of both Xenopus and Bombina at a concentration as low as 1 µg/l. The intensity increased with increasing concentration of esfenvalerate and at 10 µg/l (for Xenopus) and 50 µg/l (for Bombina) the embryos were immobilised caused by constant spasmodic twisting. Furthermore the heartbeat was slow. The results from the present study are in agreement with the effect seen in leopard frog (Rena pipiens) exposed to esfenvalerate where a decrease in activity was found at concentrations of 1.3 µg/l and a twisting response was seen at concentrations of 3.6 µg/l (Materna et al., 1995). In this study, spasmodic twisting associated with an uncoordinated spiral swimming was also seen at concentrations of above 5 µg/l.

Thus, the results from the in vivo observations revealed that esfenvalerate affected the embryos of Xenopus and Bombina in the same way even though the embryos of Xenopus seem to be more sensitive compared with the effects found in Bombina.

4.2.9 Recordings of malformation, growth inhibition, mortality and teratogenic screening

The results revealed that at concentrations up to 2.5 µg/l no malformations were seen. At a concentration above 2.5 µg/l, the malformations increased with increasing concentrations of esfenvalerate. Most embryos possessed multiple malformations and the same malformations were seen in both Xenopus and Bombina. That the effect of esfenvalerate seems to be less pronounced in experiments with Bombina compared with that found with Xenopus (a factor of 10 less sensitive) can not be ascribed to a less developed embryos of Bombina, since the test with Bombina had been prolonged to ensure that the embryos of both species went through the same stages of development. This is especially important when the effects of pyrethroids are investigated since it has been shown that older embryos of some amphibians are more sensitive to pyrethroid exposure than younger embryos (Berrill et al., 1993). As pyrethroids apparently act primarily on sodium and calcium channels in nervous tissue, the higher sensitivity of the late-stage embryos probably reflects the more differentiated state of the nervous system. The higher effect concentration found in Bombina may be ascribed to the jelly coat which may have a protective effect, even though it does not prevent the penetration of esfenvalerate. However, it should be noted that Xenopus and Bombina responded to esfenvalerate alike and that the difference in sensitivity was less than a factor of 10. Such a difference can easily be ascribed to normal differences in species sensitivity. Thus, the effect of esfenvalerate on different fish is found to differ in the same range (cf. Appendix 3).

In Xenopus, addition of 10 g esfenvalerate caused a slight reduction of the mean length of the embryos compared with the controls. However, no further reduction of the length of the embryos was seen with increasing concentrations of esfenvalerate, and esfenvalerate did not affect the mean length of the embryos of Bombina at any concentration tested. Thus, the growth inhibition does not seem to be a sensitive effect parameter in the evaluation of esfenvalerate in either Xenopus or Bombina.

Based on the present study, it can be concluded that esfenvalerate has only a small effect on the mortality on embryos of both Bombina and Xenopus and that the LC₅₀ values are higher than 150 µg/l in both species. These results are in agreement with the results found by Berrill et al. (1993) who tested the effect of fenvalerate on embryos and newly hatched tadpoles of Rena clamitans, Rena pipiens, and larvae of the salamander A. maculatum. In that study, embryos and newly hatched tadpoles did not die at concentrations up to 100 µg/l. However, a higher mortality was seen in leopard frog exposed to esfenvalerate at 22°C, and the LC₅₀ value was 7.3 µg/l (Materana et al., 1995).

When the effect of Sumi Alfa was tested no significant change in effect was seen compared with the samples where only the active ingredient was added with DMSO as a vehicle. Thus, it is not expected that the formulated product has a significantly higher toxicity on amphibians in the environment compared with the active ingredient dissolved in DMSO.

It is interesting that the in vivo observation revealed an effect at 1 µg/l in both species while the EC₅₀ value based on malformation was 3 and 30 µg/l for Xenopus and Bombina, respectively. The LC₅₀ values were higher than 150 µg/l in both species. This indicates that wrong conclusions could be drawn if the effect of esfenvalerate on amphibian was only based on LC₅₀ values. Such sublethal effects that we witnessed are likely to have serious implications on the long-term success of the exposed individuals, thus these organisms are more susceptible to predation. Furthermore, the twisting response is so obvious and so abnormal that it may prove to be a useful indicator-behaviour in the field work. The results of the present study also revealed that esfenvalerate has a teratogen index which indicates that the pesticide might be considered as a teratogen. The present study also revealed that the effect of esfenvalerate on amphibians based on the results from the standard toxicity test with Xenopus (FETAX) gives approximately the same results as those found with embryos of Bombina bombina.

4.3 The effects of prochloraz on embryos of amphibians with special reference to Xenopus and Bombina

Prochloraz is a contact fungicide of the imidazole group acting by inhibiting ergosterol synthesis in the target organisms and primarily used in cereals, grasses and rape against a number of important diseases.

4.3.1 In vivo observations

The number of malformation increased with increasing concentrations of prochloraz and the time of exposure. The most characteristic malformation caused by prochloraz in both species was edema especially in the cardiac region. The heartbeat decreased with increasing concentrations of prochloraz.

The lowest effect concentration found for prochloraz was 1.5 and 2 mg/l for Xenopus and Bombina, respectively, where some embryos exhibited edema especially in the cardiac region.

4.3.2 Recordings of malformation, growth inhibition and mortality at the end of the test

The result revealed that the effect concentrations were a little bit higher in Bombina compared with those found in Xenopus even though Bombina embryos had been exposed to the test substance one day more than embryos of Xenopus. However, the same malformation was seen in both Bombina and Xenopus. Many embryos had multiple malformations. The most characteristic malformations caused by prochloraz were severe edema especially optic and cardiac malformations where the heart often composed of a single straight tube but also brain, edema, and gut malformation were common.

Growth inhibition did not seem to be a very sensitive effect parameter when we look at the effect of prochloraz.

For Xenopus and Bombina, no significant mortality was seen at concentrations below 3 mg/l. At the end of the experiments, the LC₅₀ values were found to be 4.5 mg/l and more than 10 mg/l for Xenopus and Bombina, respectively. This indicates that malformation is the most sensitive end point. Furthermore, the results of this test also revealed that prochloraz might be considered as a teratogen.

The present study has also demonstrated that the jelly coat has no significant influence on the effect of prochloraz on the embryos of Bombina. Thus, the difference in response in the two groups of embryos can not be ascribed to the present of the jelly coat in the experiments with Bombina. Tt should be kept in mind that the only distinct difference in the effect concentration between Xenopus and Bombina was found for mortality and even in this case the difference at effect concentration was only about a factor of 2 which can easily be ascribed to interspecies differences in sensitivity. Thus, it can be concluded that prochloraz has the same effect on Bombina and Xenopus and at the same concentrations. Furthermore, it can be concluded that the most sensitive end point, which can easily be quantified, is the malformation where an EC₅₀ value of 1.4 and 2.1 mg/l was calculated for Xenopus and Bombina, respectively. Limited data on the toxicity of herbicides to amphibian larvae are available for comparison with our results. Thus, we have been unable to find any data on prochloraz for comparison with our results on Xenopus and Bombina.

4.4 Comparison of the effects of esfenvalerate and prochloraz on amphibians with the results from the OECD standard tests

It is well-known that the toxicity of esfenvalerate is very high to fish with 96 h LC₅₀ values for fathead minnows, bluegill, and killifish in the range 0.0007-0.002 mg/l. Likewise, the toxicity to Daphnia is very high with an EC₅₀ (48 h) value = 0.00024 mg/l (cf. Appendix 3). The LC₅₀ values for Xenopus and Bombina found in the present study are higher than 0.15 mg/l the highest concentration tested (the water solubility of esfenvalerate is 0.002 mg/l). Thus, amphibians seem to be less sensitive to esfenvalerate compared with both fish and Daphnia. However, when the effect of esfenvalerate is based on malformation of the amphibian embryos EC₅₀ values of 0.003 and 0.029 mg/l were found in Xenopus and Bombina, respectively, which is comparable with the value found in the standard test. Furthermore, in vivo observation has shown an effect of esfenvalerate on embryos of both Xenopus and Bombina at concentrations as low as 0.001 mg/l.

Prochloraz is toxic to fish with 96 h LC₅₀ values for rainbow trout, harlequin fish and bluegill sunfish in the range 1.5-2.9 mg/l. Likewise, the toxicity to Daphnia is high (EC₅₀ (48 h) = 2.6 mg/l) while it is very high to algae with an IC₅₀ (96 h) of 0.073 mg/l. (cf. Appendix 3). The LC₅₀ values for Xenopus and Bombina found in the present study range from 4.5 to more than 10 mg/l which indicate that prochloraz has a lower toxicity to amphibians than the normally used aquatic test organisms in the OECD standard test. However, when we look at the effect of prochloraz to induce malformation EC₅₀ values of 1.4 and 2.1 mg/l were found for Xenopus and Bombina, respectively. Based on these values prochloraz is toxic to amphibians and the effect of prochloraz on amphibians is in the same range as that found in fish and Daphnia.

4.5 Conclusion

When the toxicity of different organic compounds was evaluated using different test organisms the animal test species exhibited varying degrees of sensitivity to the selected toxicant. For some test chemicals, the most sensitive species was amphibians in other cases the fish was the most sensitive species. This stresses the importance of using different test organisms representing various trophic levels, taxa, and functions in the environment in hazard and risk assessment.

Much of the amphibian life cycle occurs in ponds, streams, and temporary pools that are often associated with agricultural areas receiving pesticide applications. Agricultural pesticides may contribute to the decline in amphibian populations and it is therefore important to investigate the effect on pesticides on amphibians.

Xenopus laevis is the test organisms used in the only internationally recognised test with amphibians. However, this test has a number of deficiencies which make its use for ecotoxicological risk assessment in Denmark less suited. Thus, Xenopus laevis represents a very special limited ecosystem in Africa. Obviously, it is a disadvantage when evaulating pesticides' effects on amphibian in Europe. In addition, some results indicate that this organism is less sensitive to organic chemicals than other amphibians. Furthermore, the standard test involves removal of the jelly coat of the embryos which may influence the sensitivity of the test system. Therefore a new toxicity test with Bombina bombina was developed since this organism has widespread wild living and closely related species represented in Europe. This makes the important comparison between laboratory tests and field test possible. The results from the present study revealed that we are able to breed Bombina bombina throughout the year and get a sufficient number of healthy embryos for ecotoxicity testing.

Esfenvalerate has only a small effect on the mortality on embryos of both Bombina and Xenopus during the present experimental conditions where the LC₅₀ values were higher than 150 µg/l in both species. However, when the number of malformations were used as an end point it was demonstrated that the EC₅₀ values for malformations were only 3 µg/l and 29 µg/l, for Xenopus and Bombina, respectively. It should be noted that Xenopus and Bombina responded to esfenvalerate in the same way and that the change in sensitivity was less than a factor of 10. Such a difference can easily be ascribed to normal differences in species sensitivity. Furthermore, it is interesting that the in vivo observation revealed effects at 1 µg/l in both species. The effects are twisting and apparent partial paralysis of the embryos. Such sublethal effects that we witnessed are likely to have serious implications on the long-term success of the exposed individuals, thus, these organisms are more susceptible to predation. It indicates that wrong conclusions could be drawn if the effect of esfenvalerate on amphibian was only based on LC₅₀ values.

For Xenopus and Bombina, no significant mortality was seen at prochloraz concentrations below 3 mg/l and the LC₅₀ values were found to be 4.5 mg/l and more than 10 mg/l for Xenopus and Bombina, respectively. However as for esfenvalerate, the most sensitive end point which can easily be quantified is the malformation where EC₅₀ values of 1.4 and 1.5 mg/l were calculated for Xenopus and Bombina, respectively. Many embryos possessed multiple malformations, and the same malformations were seen in both Bombina and Xenopus. The present study has also demonstrated that the jelly coat has no influence on the effect of prochloraz on the embryos of Xenopus. Even though the LC₅₀ value is higher in Bombina compared with the one found in Xenopus it should be noted that even in this case the difference in effect concentration was only about a factor of 2 which can easily be ascribed to interspecies differences. It can be concluded that prochloraz has the same effect on Bombina and Xenopus and at the same concentrations.

Embryos of Bombina and Xenopus are clearly unlikely to be killed by short-term exposure to low concentrations of esfenvalerate and prochloraz. Despite concentrations raised to environmentally unrealistic levels of 0.1 and more than 4 mg/l for esfenvalerate and prochloraz, respectively for 96 h, the embryos did not die. If mortality is our only measure of response to pesticide exposure, we would conclude that these amphibians were relatively tolerant of these pesticides. However, the sublethal effects that we witnessed are likely to have serious implications on the long-term success of the exposed individuals. Thus, when the effect of these pesticides is based on e.g. malformation of the amphibian embryos the EC₅₀ values found in Xenopus and Bombina are comparable with those found in the standard aquatic tests with fish and Daphnia.

Finally, the present study has demonstrated that the effect of pesticides on amphibians is highly dependent on several factors. Thus, several studies have found significant differences in sensitivity among different species of amphibian species and it is a well-known phenomenon that temperature may influence the effect concentration of pesticides. Furthermore, several studies indicate that the sensitivity of an amphibian to a given chemical is highly dependent on the different larvae stages of the amphibian. Therefore additional research with more chemicals, the sensitivity on different larval stage in several species, and the effects of temperature must be conducted to make definitive conclusions concerning hazards. Only with such information will it be possible to determine if pesticide use poses a serious threat to amphibian populations in Denmark.

Fodnoter

[2] This guide is under the jurisdiction of ASTM Committee E-47 on Biological Effects and Environmental Fate and is the direct responsibility of Subcommettee E47.01 on Aquatic Toxicology.

Current edition approved Sept. 15, 1991, Published November 1991.