Evironmental project, 986 – Background report on pre-validation of an OECD springtail test guideline

Background Report on Pre-validation of an OECD Springtail Test Guideline

2 Materials and methods

2.1 Background - species identification and comparison
     2.1.1 Literature search - Ecology and biology of species
2.2 Sensitivity comparison
     2.2.1 General sensitivity comparison
     2.2.2 Laboratory experiments - Comparison of age/size and toxicity
     2.2.3 Sex-ratio effects on offspring
2.3 Identifying test concentration ranges for ring-testing
     2.3.1 Test conditions

According to the aims the studies were divided in two parts. The first part deals with studies on the importance of ecological, biological and physiological for choice of species. These issues will be dealt with though specific aspects (see 2.1). Second, to identify test-concentration ranges for future positive control compounds three compounds will be tested based on previous ring-test programmes and test guidelines (see 2.2).

2.1 Background - species identification and comparison

2.1.1 Literature search - Ecology and biology of species

Based on a literature search and experience a short description of the ecology and biology was performed. Emphasis was put on background for choice of the species included in the guideline (see 3.1).

2.2 Sensitivity comparison

The two selected species differ in various aspects and may hence in theory differ in sensitivity to chemical toxicity. This issue was studied via two approaches.

2.2.1 General sensitivity comparison

To study the general chemical sensitivity of the species a literature survey was performed. In this review two compounds were selected for further analysis. The first compound was an choline-esterase inhibiting pesticide, dimethoate. For this compound there was already considerable knowledge available. The second compound was a heavy metal, copper, with a more broad mode-of-action. For this group relevant information is also available in the published literature. Hence, in the present report a comparison was made between the two species, F. fimetaria and F. candida, with regard to the toxicity of Copper and Dimethoate. The data has been obtained via searches in international databases such as WOS, BIOSIS, SCIC and ECOTOX. Only original literature has been cited (see 3.2.1).

2.2.2 Laboratory experiments - Comparison of age/size and toxicity

One of the notable differences between the two species is the size difference. In theory the two species may differ in sensitivity to chemicals due to such a size difference, i.e. the surface area compared to volume is larger for smaller animals (F. fimetaria), which will influence uptake chemicals taken up via outer cuticle.

In accordance, an experiment was set up to study whether the toxicity differed between the two species and to see if this could be attributed to size differences. The experiment was based on exposing a range of age-classes (hence also size classes) to one concentration of a chemical and observing differences in mortality between the age-classes (see Table 1). Due to the limited time-span of the project it was decided to use a compound, dimethoate, already tested for both species and to use the LC₅₀ for Folsomia fimetaria reported by Krogh (1995) (see 3.2.2).

2.2.2.1 Test compound

Dimethoate (purity: 400 g L^-1 Cheminova Agro A/S) (CAS no. 60515)

Table 1. Age of animals in size classes employed for both Folsomia fimetaria and Folsomia candida.

Size class	1	2	3	4	5	6	7	8	9
Age/days	0-2	3-5	6-8	9-11	12-14	15-17	18-20	21-23	24-26

Hence, the mortality after exposure for one week to dimethoate 2 mg kg^-1 and control was tested for 9 size/age-classes of Folsomia fimetaria and Folsomia candida in 5% OECD soil. They were fed 15 mg Baker's yeast in the beginning of the experiment. Ten individuals were added to each replicate container and extracted after 1 week. There were 2 replicates per species and size-class and treatment.

2.2.2.2 Test soil, counting and statistics

See section 2.2.1.3-2.2.1-5

2.2.3 Sex-ratio effects on offspring

Experiments were conducted to see if it was possible to include offspring sex-ratio as an effect parameter in the draft guideline. To study this adult Folsomia fimetaria were exposed for 42 days to soil contaminated with two potential sex-ratio disrupting compounds, ethynyl-estradiol and methyl-testosterone. The experiment was continued for a total of 42 days to allow the first clutch of the juveniles to become adult (Table 2). Allowing the first clutch of the juveniles to become adult made it potentially possible to identify the sexes on the offspring. The maximum concentrations used were 1000 mg a.s. kg^-1 soil corresponding to the limit-testing for plant protection products (see 3.2.3).

Since the compounds may bind to the soil particles and hence not be available to the organism a water- exposure test was conducted to ensure exposure. Hence, adult Collembola were exposed for 48 hours to 12 mg testosterone L^-1 or 2 mg estradiol L^-1 after which the adults were allowed to produce eggs over a period of 7 days. The hatched juveniles were allowed to growth for 23 days prior to size measuring.

Table 2. Test conditions of the sex-determination toxicity tests with Folsomia candida.

Test species	Folsomia fimetaria
Life stage	23-26 days (synchronised)
Exposure duration	42 days
Test volume	30 g wet weight
Temperature	20°C
Light cycle	16:8 light/dark
Feeding	every 14 days
Initial organisms per replicates	20

2.2.3.1 Test compounds

Methyl-testosterone (purity: 97%, Sigma Aldrich) (CAS no. 58-18-4)
17-βEthynyl-testosterone (purity: 85%, Sigma Aldrich) (CAS no. 57-63-6)

2.2.3.2 Test soil, counting and statistics

See section 2.2.1.3-2.2.1-5

2.3 Identifying test concentration ranges for ring-testing

Experiments were performed in order to identify the test concentrations used in a future ring-test programme and to identify test concentrations used for a positive control. The experiments included three compounds fipronil, 3,5-dichlorophenol and boric acid. The experiments were performed according to conditions stated in table 3.

2.3.1 Test conditions

2.3.1.1 Test organisms

The Collembolans Folsomia fimetaria and Folsomia candida (Collembola, Isotomidae) were used as test organisms. For fipronil and 3,5-dichlorophenol tests were performed with F. fimetaria and for boric acid tests were performed with both species.

Cultures of these species were kept on moist substrate of plaster of Paris and pulverised activated charcoal (as described in the guideline). Granulated yeast was added weekly as a food source. The cultures were maintained at the Department of Terrestrial Ecology, National Environmental Research Institute.

Table 3: Test conditions of the range determination toxicity tests with Folsomia candida and Folsomia fimetaria.

Test species	Folsomia candida	Folsomia fimetaria
Life stage	10-12 days (synchronised)	23-26 days (synchronised)
Exposure duration	14/21 days	14/21 days
Test volume	30 g wet weight	30 g wet weight
Temperature	20°C	20°C
Light cycle	16:8 light/dark	16:8 light/dark
Feeding	every 14 days	every 14 days
Initial organisms per replicates	10	20

2.3.1.2 Test compounds

All compound used were of analytic grade and obtained from Sigma-Aldrich (Germany) or Cheminova.

Boric Acid (purity: 100%, Sigma Aldrich) (CAS no. 10043-35-3)
Fipronil (purity: 97%, Sigma Aldrich) (CAS no. 1200068-37-3)
3,5-Dichlorophenol (DCP) (purity: 97%, Sigma Aldrich) (CAS no. 591-35-5)

2.3.1.3 Test soil

The test soil was an artificial mixed soil according to OECD guideline. Compared to the "standard" OECD soil, which has a 10% organic matter content, it was chosen in the present test to use a 5% organic matter soil. This latter choice was made to mimic field condition and reduce binding of test chemicals. Soil composition: Organic matter - 5%; clay - 21% and sand - 74% [the clay and sand content has been changed, compared to the original OECD soil, to keep the relative composition the same].

2.3.1.4 Counting

At the end of each experiment the animals were heat-extracted and counted with a computerised system employing a digital image processing (DIP) approach (Krogh et al. 1998). The guideline alternatively suggests to extract the animals by flotation and to manually count the organisms floating on the surface. It was chosen in this context to use the heat extraction and the automated counting since this is considered more precise than the flotation and manually counting.

2.3.1.5 Statistics

the data were checked for normality using a 2 test, and for homogeneity of variance by Barlett's test. Effect concentrations, EC₁₀ and EC₅₀, and confidence intervals were estimated by fitting a logistic model to the data (Lacey and Mallet 1991). The formulae were reparametrized by incorporation of the EC₁₀ and EC₅₀ into the equation (Krogh 1995). The estimates with a 95% confidence interval were performed with the SAS procedure PROC NLIN (Enterprise Guide 1.3)