殺虫剤メトフルトリンの環境中代謝物における水生生物急性毒性

Introduction

Pyrethroid is one of the most important chemical classes of insecticide for both agricultural and public hygiene uses; it is not only exhibits an excellent biological activity but also readily degrades biotically and abiotically in the environment.^1–4) It can be well understood from its mode of action that pyrethroid is highly toxic to fish and arthropods (crustaceans and insects) in general, but extremely less toxic to other invertebrates such as mollusk and aquatic plants including algae.⁵⁾ Rapid degradation of pyrethroid in the environment makes its aquatic risks under practical uses acceptable and manageable, while ecological risk assessment of corresponding metabolites is very limited at the present. In general, metabolic transformation of a pesticide either destroys a toxicophore structure or introduces a new functional group, generally leading to increased molecular hydrophilicity, and the changes in the mode of action or the uptake potential result in less toxicity of metabolites than the parent.⁶⁾ Most metabolites formed via ester cleavage show far less toxicity (LC₅₀, EC₅₀) to sensitive taxa by 2–6 orders of magnitude than do corresponding pyrethroids,^7–9) while limited information is available on the aquatic toxicity of metabolites having an intact ester linkage, as reported for 4′-OH-bifenthrin.¹⁰⁾

Metofluthrin (I) (SumiOne®, Eminence®) [2,3,5,6-tetrafluoro-4-(methoxymethyl)benzyl(1R,3R)-2,2-dimethyl-3-((1EZ)-prop-1-enyl)cyclopropanecarboxylate] is a new pyrethroid insecticide with an extremely high knockdown activity, especially against mosquitoes.^11,12) Similarly to other pyrethroids, I exhibits high acute toxicity to common carp (Cyprinus carpio) and rainbow trout (Oncorhynchus mykiss) with 96-hr LC₅₀ of 0.0012 and 0.00306 mg/L, respectively, and to Daphnia magna with 48-hr EC₅₀ of 0.0047 mg/L, while it is much less toxic to green algae (Pseudokirchneriella subcapitata) with 72-hr E_bC₅₀ of 0.16 mg/L.^12–14) Considering the typical use pattern of I, which is distributed into the air by vaporization as a mosquito adulticide, its direct emission into the aquatic environment is most unlikely. Even if emitted into the aquatic environment, I degrades via either hydrolysis to the corresponding acid and alcohol moieties or sunlight photolysis with successive oxidation at the prop-1-enyl side chain with no remarkable change in an E/Z isomer ratio because of similar degradation rates and no isomerization.¹⁵⁾ Furthermore, when I is distributed in the terrestrial environment, aerobic microbes in soil rapidly metabolize it via ester cleavage followed by successive oxidation¹⁶⁾ or it is photodegraded by sunlight on the soil surface.¹⁷⁾ Eight major metabolites (II–IX) detected through environmental fate studies, as shown in Fig. 1, have unique chemical structures different from other pyrethroids.

Fig. 1. Structures of metofluthrin metabolites in the environment with their route of formation.

The objective of this study has been to determine basic aquatic ecotoxicological profiles of the major metabolites identified in environmental fate studies of metofluthrin, using three standard aquatic species (fish, daphnid, algal), in relation to the structural modification from metofluthrin.

Materials and Methods

1. Chemicals

The metabolites of I (II–VI and VIII) were prepared as test substances in our laboratory according to the reported methods.^11,15–17) The chemical purity of each metabolite was determined by HPLC to be greater than 98%. VII (99.4%) and IX (>97%) were purchased from Showa Denko K.K. (Tokyo, Japan) and Tokyo Kasei Kogyo Co., Ltd. (Tokyo, Japan), respectively, and used without further purification. All other chemicals were of a reagent grade and purchased from commercial suppliers unless otherwise noted.

2. Test organisms

Fathead minnow (P. promelas), D. magna, and the unicellular green algal species, P. subcapitata (ATCC22662), were chosen as test species because of recommendations in international test guidelines, such as those of the OECD (Organisation for Economic Co-operation and Development).¹⁸⁾ Parental organisms were originally obtained from the National Institute for Environmental Studies (Ibaraki, Japan) for fathead minnows and from Sumika Technoservice Corporation (Hyogo, Japan) for D. magna and P. subcapitata. The cultures of fathead minnows were held in tap water dechlorinated with activated charcoal at ca. 25°C with a 16-hr daylight photoperiod. They regularly fed on a standard commercial fish food (TetraMin®, Tetra Werke, Germany), and juveniles (total length, 1.5–2.9 cm) were used for bioassays. D. magna cultures, held at ca. 20°C with a 16-hr daylight photoperiod, regularly fed on commercially available chlorella (Chlorella V12, Chlorella Industry Co., Ltd., Tokyo, Japan), and <24-hr-old neonates were used. Elendt M4 medium referenced in OECD guideline 202¹⁸⁾ or ASTM Hard Reconstituted Fresh Water¹⁹⁾ was used as culture water. Pure water with an electrical resistivity of more than 17 MΩ cm, provided by a Barnstead E-pure D4643 (4Module E-pure, Barnstead Thermolyne Co., Iowa, USA), was used to prepare the culture media. Fathead minnows and D. magna were not fed during the bioassays. Precultures of P. subcapitata were prepared prior to each bioassay from stocks in a refrigerator and incubated at 23–26°C in the medium referenced in OECD guideline 201¹⁸⁾ under continuous shaking (Multishaker MMS-310, Tokyo Rikakikai Co., Ltd.) and illumination (fluorescent bulbs).

3. Bioassays

The acute toxicity of each metabolite on three species was examined basically in accordance with the corresponding international guidelines of the OECD.¹⁸⁾ The same types of media for culturing were used for the bioassays. The aqueous solution of a metabolite (V–IX) was prepared at a desired concentration by its direct dissolution, followed by serial dilution with the media. Otherwise, a metabolite (II–IV) was first dissolved in N,N-dimethylformamide (DMF), followed by dilution with the media at 0.1 mL DMF/L. When a test substance was partly dissolved in the media, the supernatant by decantation was used for the exposure and chemical analysis. The chemical analysis of each metabolite was regularly conducted in each test, together with measurements of temperature (Multi-thermometer, JAPAN PET DRUGS Co., Ltd.) and pH (Model B-212, Horiba Ltd., Japan). Except in the algal tests, the dissolved oxygen (DO) concentration in the test solution was also measured (SevenGo pro, Mettler Toledo, Columbus, Ohio, USA).

3.1. Fish acute toxicity test

Groups of seven fathead minnows were exposed without replication in 1000 mL of each test solution for 96 hr under static conditions using 1-L glass beakers. In the case of III, 20-L size stainless steel vessels (27×27×32 cm) filled with a 10-L test solution were used instead under static-renewal (every 48 hr) conditions to maintain exposure concentrations. The test vessels were partly immersed in a temperature-controlled water bath maintained at 25±1°C under a photoperiod of 16 hr/day using fluorescent bulbs (ca. 500–1000 lx). Except in the cases of III and VI, the test solutions were mildly aerated to maintain the DO levels. Biological observations including mortality and sublethal toxic symptoms (e.g., loss of equilibrium) were made daily, and mortality was defined as the lack of movement after gentle stimulation.

3.2. Daphnid acute immobilization test

Groups of five D. magna were exposed in 50 mL (for VI) or 100 mL (in the other cases) of each test solution for 48 hr without aeration under static conditions using 100-mL glass beakers, similarly to the fish study but at 20±2°C. Four replicates were established at each test concentration and for the control. Biological observations including immobilization and sublethal toxic symptoms (e.g., erratic swimming) were made daily. Immobilization was defined as the lack of free swimming within 15 sec after gentle agitation.

3.3. Algal growth inhibition test

P. subcapitata at the estimated initial population of 10⁴ cells/mL was exposed in 5 mL of each test solution for 96 hr under static shaking at 200 rpm using 10-mL glass vials loosely covered with a transparent film. Three replicates were established at each test concentration and twice for the control (water and solvent) group. The test vessels were placed on a horizontal shaking incubator (Multishaker MMS-310, Tokyo Rikakikai Co., Ltd.) maintained at 24±2°C under continuous illumination with fluorescent bulbs (1700–2500 lx measured with the Digital Illuminometer T-1H, Minolta Camera). The test vessels were repositioned daily in order to reduce/normalize bias between replicates and test groups. Algal biomass (cell density) was estimated daily by measuring the fluorescence of chlorophyll-a in algal cells using a fluorescence spectrophotometer (LS-55 luminescence spectrometer, PerkinElmer, Inc.). Each sample (0.05–0.5 mL) removed from the test vessels was diluted with acetone and the fluorescence in a quartz cuvette (1 cm pathlength) was measured at excitation and emission wavelengths of 430 and 667 nm, respectively.²⁰⁾

Since a remarkable pH decrease below 5 was observed for the test solutions of VI and IX at 100 mg/L due to their acidity and the limited buffering capacity of the OECD medium, the pH effect on algal toxicity was conveniently examined for VI at 100 mg/L by readjusting the pH of the exposure OECD medium to ca. 8 with 0.1 N NaOH.

4. Chemical analysis

At least at the initiation and termination of exposure, the concentration of each metabolite was determined by direct HPLC analysis of each test solution after an appropriate dilution. A Shimadzu HPLC system (LC-10AD pump, SCL-10A system controller and SPD-10A UV detector at 230 nm) equipped with an L-column ODS (5 µm, 4.6mmϕ×150 mm; Chemicals Evaluation and Research Institute, Japan, Tokyo) was operated at a flow rate of 1.0 mL/min under the isocratic condition. The mixing ratio (v/v) of acetonitrile/0.05% trifluoroacetic acid water as a mobile phase and the typical retention time of each metabolite in parenthesis are as follows: II, 35/65 (13.3, 16.2, 17.3, and 18.5 min); III, 3/2 (6.8 min); IV, 2/3 (17.5 min); V, 2/3 (8.1 min); VI, 1/9 (8.1 min); VII, 1/3 (11.5 min); VIII, 1/3 (8.5 min); and IX, 5/95 (5.2 min).

5. Statistical analysis

Acute toxicity values, fish 96-hr LC₅₀, daphnid 48-hr EC₅₀ and algal 96-hr EC₅₀, were determined on the basis of the arithmetic mean measured test concentrations. When the measured concentration at the end of exposure decreased by >20% as compared with the initial concentration, a time-weighted mean concentration was used instead. Among similar response variables on algal test, such as “biomass” (area under the growth curve, E_bC₅₀) and “yield” (final cell density minus the initial one, E_yC₅₀), the latter one was selected as a representative sensitive one for the algal evaluation.¹⁸⁾ The probit or linear interpolation method was applied. The former method was used for sufficient datasets for fish and daphnids (III for fish and II for daphnids). Computer programs of PROBIT (ver. 1.5, U.S. EPA) and ICP (ver. 2.0, U.S. EPA) were used for these regression analyses. If the dataset was sufficient, their 95% confidence intervals were also estimated.

Results

1. Fish acute toxicity test

The water temperature, DO, and pH were mostly within the normal ranges: 24–26°C, 5.0–8.9 ppm, and 5.8–8.4, respectively. Some low DO values of 2.8–4.8 ppm, sporadically observed in the exposures of II, IV, and VIII, had no impact on the fish, as no abnormal behavior, such as a loss of equilibrium, was observed in any test group. Exposure to V–IX, formed via ester cleavage, resulted in no mortality even at the highest nominal concentration of 100 mg/L, as listed in Table 1. Since dead fish were observed only at 24 hr, the cumulative mortality was found to be independent of an exposure period of up to 96 hr. Among the metabolites having an intact ester linkage, II and IV showed no and slight toxicity at the nominal concentration of 100 mg/L, respectively. In contrast, a clear dose-response was observed for III at 0.16–1.1 mg/L, and the lowest LC₅₀ value among the metabolites was estimated to be 0.44 mg/L.

Table 1. Fish acute toxicity of metofluthrin metabolites

Metabolite	Test concentration (mg/L)		Cumulative mortality (%)	96-hr LC50 (mg/L) [95% C.I.]	ECOSAR estimated LC50 (mg/L)f
	Nominal	Mean measured	96 hrd
II	0 (Solvent)a	NAc (<1.0)	0	>48	13 (ester)36 (neutral)
	100a,b	48	0
III	0 (Solvent) a	NA (<0.040)	0	0.44e [0.30–0.65]	4.4 (aldehyde)8.6 (ester)21 (neutral)
	0.25	0.16	0
	0.50	0.29	29
	1.0	0.56	57*
	2.0	1.1	100
	4.0	2.2	100
IV	0 (Solvent) a	NA (<10)	0	>77	64 (ester-acid)13 (neutral)
	100b	77	14
V	0	NA (<2.0)	0	>92	180 (neutral organic acid)
	100b	92	0
VI	0	NA (<2.0)	0	>94	40000 (neutral organic acid)
	100a	94	0
VII	0	NA (<1.0)	0	>95	140 (benzyl alcohol)370 (neutral)
	100	95	0
VIII	0	NA (<20)	0	>120	7400 (neutral organic acid)
	100	120	0
IX	0	NA (<1.0)	0	>99	58000 (neutral organic acid)
	100a	99	0

*: Survived fish showed toxic symptoms (loss of equilibrium or hyperactive swimming). a: The results at the lower test concentrations or of negative control were omitted because of their insignificance. b: Since undissolved test substance was observed in the test solution, the exposure and chemical analysis were conducted using a supernatant solution by decantation. c: Not applicable (The limit of detection was indicated in the parenthesis). d: Since cumulative mortalities at 24, 48 and 72 hr were identical with those at 96 hr, those values were omitted. e: LC₅₀ value was estimated by probit analysis. f: Chemical class used for estimation in the parentheses.

2. Daphnid acute immobilization test

Similarly to the fish tests, the water temperature, DO, and pH were almost within the normal ranges: 19–23°C, 7.8–10.6 ppm, and 7.0–8.9, respectively. No or negligible immobilization (≤10%), even at 100 mg/L, was observed for VI–IX (Table 2), while the exposure to V at 100 mg/L caused 90% cumulative immobility after 48 hr, resulting in the EC₅₀ value of 71 mg/L. IV showed no immobilization at up to a 100 mg/L difference from a slight fish toxicity, while the dose-response toxicity was clearly observed for II at 13–55 mg/L (EC₅₀=52 mg/L). III was most toxic to daphnids among the tested metabolites with the EC₅₀ (48-hr) value of 6.3 mg/L.

Table 2. Daphnid acute immobilization by metofluthrin metabolites

Metabolite	Test concentration (mg/L)		Cumulative immobility (%)		48-hr EC50 (mg/L) [95% C.I.]	ECOSAR estimated EC50 (mg/L)f
	Nominal	Mean measured	24 hr	48 hr
II	0 (Solvent) a	NAc (<1.0)	0	0	52d [44–67]	23 (ester)23 (neutral)
	13a	13	0	0
	25	24	0	5*
	50	49	0	40*
	100	55	0*	60*
III	0 (Solvent) a	NA (<0.040)	0	0	6.3e [5.5–6.6]	3.8 (aldehyde)16 (ester)13 (neutral)
	0.25	0.16	0	0
	0.50	0.32	0*	0*
	1.0	0.60	10*	5*
	2.0	1.2	0*	0*
	4.0	2.4	0*	10*
	8.0	4.7	5*	25*
	16	9.4	45*	100
	32	19	95*	100
IV	0 (Solvent) a	NA (<10)	0	0	>76	110 (ester-acid)8.6 (neutral)
	100b	76	0	0
V	0	NA (<1.0)	0	0	71e [66–77]	110 (neutral organic acid)
	13a	12	0	0
	25	24	0	10*
	50	47	0	0*
	100	92	15*	90*
VI	0	NA (<2.0)	0	0	>92	19000 (neutral organic acid)
	100a	92	0	0
VII	0	NA (<1.0)	0	0	>93	100 (benzyl alcohol)210 (neutral)
	100	93	0	0
VIII	0	NA (<10)	0	0	>120	4000 (neutral organic acid)
	100	120	0	0
IX	0	NA (<1.0)	0	0	>95	29000 (neutral organic acid)
	25a	24	0	0
	50	50	0	0*
	100	95	0	10

*: Survived daphnids showed toxic symptoms (hypoactive swimming). a, b, c, f: Same as Table 1. d: EC₅₀ value was estimated by probit analysis. e: EC₅₀ value was estimated by linear interpolation.

3. Algal growth inhibition test

The exponential increase of a mean cell density, >16 times per 72 hr, showed excellent algal growth in the control groups at 23–26°C without any unusual variation, indicating the validity of all exposures. Any remarkable growth inhibition, ≥25% for 96 hr, was not observed for II, IV, and VIII, even at 100 mg/L, while the other metabolites exhibited clear dose-responses. Similarly to the fish and daphnid tests, III was most toxic among the tested metabolites with the E_yC₅₀ value of 2.6 mg/L. While V and VII partly inhibited (ca. 60%) at the highest concentrations, VI and IX showed very steep dose-responses from 0 to 100% inhibition at the highest two concentrations, with significantly acidic pH values (4.0–5.9) being recorded. The additional bioassay of VI at the readjusted pH of ca. 8 showed insignificant growth inhibition even at 100 mg/L (E_yC₅₀>100 mg/L). In the other cases, the measured pH values of the test solutions were in the range of 6.2–8.9, reflecting OECD medium conditions and algal photosynthesis. The E_yC₅₀ value of each metabolite was thus estimated as listed in Table 3.

Table 3. Algal growth inhibition by metofluthrin metabolites

Metabolite	Test concentration (mg/L)		Mean cell density,g ×104 cells/mL		96 hr EyC50 (mg/L) [95% C.I.]	ECOSAR estimated EC50 (mg/L)f
	Nominal	Mean measured	24/48/72 hr	96 hr (yield inhibition %)
II	0 (Solvent)a	NAc (<2.0)	4.7/22/86	300 (−)	>61	8.4 (ester)25 (neutral)
	25a	22	4.7/23/88	310 (−2)
	50	37	5.0/22/82	280 (7)
	100	61	5.0/20/70	230 (23)
III	0 (Solvent)a	NA (<0.20)	4.6/26/120	360 (−)	2.6d [2.5–2.7]	7.6 (aldehyde)5.4 (ester)16 (neutral)
	1.3	0.84	4.6/25/100	340 (6)
	2.5	1.7	5.0/25/93	250 (30)
	5	3.5	4.3/21/56	120 (68)
	10	7.8	3.7/8.0/10	24 (94)
	20	15	3.7/7.9/6.9	6.1 (99)
IV	0 (Solvent)a	NA (<10)	4.6/26/120	360 (−)	>74	38 (ester-acid)11 (neutral)
	100b	74	4.3/25/110	340 (6)
V	0	NA (<0.10)	5.1/30/140	440 (−)	84d [82–85]	120 (neutral organic acid)
	50a	55	5.1/28/120	450 (−2)
	100	95	3.6/17/57	160 (64)
VI	0	NA (<2.0)	5.3/24/96	310 (−)	75d [75–76]	7600 (neutral organic acid)
	50a	51	6.4/25/93	330 (−4)
	100	100	1.0/1.2/1.0	0.58 (100)
	VI (pH 8)	0	NMe	5.4/38/125	509 (−)		>100
		100	NM	5.0/24/88	315 (38)
VII	0	NA (<1.0)	4.0/25/120	390 (−)	97d [92–100]	40 (benzyl alcohol)130 (neutral)
	25a	26	3.8/25/110	400 (−4)
	50	53	3.8/25/98	360 (5)
	100	110	3.7/19/45	150 (62)
VIII	0	NA (<10)	4.0/25/120	390 (−)	>110	2300 (neutral organic acid)
	100	110	3.7/24/95	390 (−1)
IX	0	NA (<1.0)	4.5/23/95	290 (−)	75d [74–75]	11000 (neutral organic acid)
	50a	51	5.1/23/95	300 (−2)
	100	99	0.63/0.69/1.7	2.2 (99)

a, b, c, f: Same as Table 1. d: E_yC₅₀ value was determined by linear interpolation. e: Not measured. g: Estimated by measuring fluorescence of chlorophyll-a in algal cells.

Discussion

Although the chemical structure prerequisite for aquatic toxicity is not clearly defined for pyrethroids, modification at a molecular end may cause a slight change in the toxicity of a metabolite. In fact, 4′-OH-bifenthrin, only hydroxylated at the 4′-position of the biphenyl moiety, still showed high toxicities to fish (LC₅₀, 0.0039 mg/L) and daphnids (EC₅₀, 0.0012 mg/L).¹⁰⁾ Furthermore, ester cleavage at the side chain in the acid moiety of acrinathrin and/or hydration of its α-benzyl cyano group greatly reduced its toxicity by a factor of 55 to >5×10⁵ in fish (LC₅₀, 0.33–0.39 mg/L) and daphnids (EC₅₀, 0.548 to >10 mg/L).²¹⁾ These observations may indicate the relevance of molecular hydrophobicity of pyrethroids with their aquatic toxicity. Among the metabolites with modification at the prop-1-enyl side chain of I, significantly low toxicity was observed for II and IV in three tested species (LC₅₀ and EC₅₀, >48 mg/L). In contrast, III showed a higher toxicity than II and IV to all species, but much less than I by two to three orders of magnitude to pyrethroid-sensitive taxa (fish and daphnid) and one order of magnitude to pyrethroid-insensitive ones (alga). Gobas et al. applied the diffusion rate of a hydrophobic chemical through membrane-diffusion layer barriers to express the rates of chemical uptake and elimination in fish,²²⁾ and they suggested the constant uptake and proportional decrease in elimination with an increasing log K_ow (n-octanol/water partition coefficient) above 3–4. The log K_ow values of II–IV are estimated to be 3.1–3.5 by the EPI-Suite²³⁾ and much lower than I (5.0),¹⁴⁾ suggesting higher elimination of these metabolites as one of the reasons for their lower toxicity.

The daphnid EC₅₀/NOEC ratio of III was calculated to be 39 (6.3 mg/L vs. 0.16 mg/L), while the corresponding values of I¹³⁾ and twenty pyrethroids¹³⁾ were much smaller, 1.6 and 3.8±2.8, respectively. Such a moderate concentration-response pattern in the sublethal effect of III, different from pyrethroids, indicated that the observed toxicity did not originate from pyrethroid-specific mode of action but from other mechanisms. Since aldehydes are well known as toxic chemicals,²⁴⁾ the formyl group in III is most likely to be a toxicophore. This is supported by the toxicity of aldehyde metabolites of fluridone and glyphosate.⁶⁾ Therefore, the prop-1-enyl side chain at a molecular end is considered to be one of the key structures relating to aquatic toxicity of I.

The toxicity of the metabolites (VII, VIII, and IX), originating from the alcohol moiety of I, in fish and daphnids (LC₅₀ or EC₅₀, >93 mg/L) resembles that of 2,3,5,6-tetrafluoro-4-methylbenzoic acid²⁵⁾ and 2,3,5,6-tetrafluorobenzoic acid²⁷⁾ (LC₅₀ or EC₅₀, >100 mg/L), independent of a number of a carboxylic group. The very weak toxicity of V and VI (LC₅₀ or EC₅₀, 71 to >94 mg/L) was found to be common to the chrysanthemic acid derivatives of cypermethrin,^7,26) lambda-cyhalothrin,⁷⁾ tefluthrin,²⁵⁾ and acrinathrin²¹⁾ (LC₅₀ or EC₅₀; 3 to 180, >16 to 100, >15.8 to >182 and 69 to >110 mg/L, respectively). The last one, cyclopropyl dicarboxylic acid from acrinathrin, only shows an algal toxicity similar to VI.

The algal toxicity of dicarboxylic acid derivatives (VI and IX) with steep dose–response curves can be accounted for by lower pH values (4.0–5.9) of the corresponding solutions. The dissociation of carboxylic groups in these metabolites may control their uptake, and, as a result, it is supposed to alter their algal toxicity.²⁸⁾ The pKa values of VI and IX can be respectively estimated to be 6.73/4.26 and 1.96/0.95 by ACD/pKa DB (ver. 4.56, Advanced Chemistry Development Inc., Toronto, Canada). By comparing these pKas and the observed pH values (4.0–4.1) at the toxic level, 78% of VI should be present in an undissociated form, but IX is fully ionized. Since both metabolites exhibited the same E_yC₅₀ value of 75 mg/L irrespective of different ionized fractions, factors other than carboxylic acid dissociation are likely to control their toxicity. Such a low pH results in an extremely small fraction of HCO₃⁻ as a bio-available carbon source.²⁹⁾ Therefore, this may be a key factor in the observed algal growth inhibition of these highly acidic metabolites. The inhibitory effect on algal growth is known for highly acidic terephthalic acid³⁰⁾ and benzoic acid drivatives.³¹⁾ The medium pH effect on algal toxicity was further confirmed by the greatly reduced 96-hr E_yC₅₀ of >100 mg/L in the additional bioassay of VI at pH 8. Considering the unlikelihood of pH change by dilution in natural bodies of water related to its buffering capacity, the toxicity of VI and IX observed on a laboratory scale should be an unrealistic overestimate, and the toxic profile under standard pH conditions (>100 mg/L for VI) is appropriate for risk assessments as prescribed in the UK technical guidance.³²⁾

Metabolites generally show less aquatic toxicity than does pesticide itself, but unexpected toxicity is not ruled out. Instead of conducting any toxicity study, QSAR approaches^33,34) using the relationship of toxicity with physico-chemical parameters such as log K_ow have been applied to predict the ecotoxicity of a chemical. The ECOSAR program (ver. 1.11),³⁵⁾ one of the most popular models,³⁴⁾ estimates aquatic toxicity based on regression from training sets of experimentally obtained ecotoxicity data with a physico-chemical property (mainly estimated log K_ow) by taking account of 111 chemical classes, such as aldehydes and esters. ECOSAR predicted the LC₅₀ and EC₅₀ values of each metabolite mostly within one order of magnitude, but a few outliers were found (Tables 1–3). ECOSAR underestimated the fish toxicity of III when it was classified as ester and neutral organics, while the reasonably close value to that in our study was estimated when an aldehyde class was selected. This supports a formyl group as the dominant toxicophore in III, as mentioned above. Much greater underestimation of toxicity to three species was obtained by ECOSAR for acidic metabolites (VI, VIII, and IX), which may be accounted for by the limited maximum concentration taken in the toxicity tests (nominal 100 mg/L) as well as an acidity effect in the algal toxicity.

As a further ecotoxicological profiling of the metabolites in relation to a typical usage of the parent, an ecological risk assessment for each metabolite of I was conveniently conducted in terms of acute toxicity exposure ratio (TER), defined as the toxicity (LC₅₀, EC₅₀) value divided by its predicted environmental concentration (PEC), a so-called quotient method.³⁶⁾ In the case of SumiOne® Liquid vaporizer, a typical use pattern in the EU region, the PEC value of I was calculated to be 7.6×10⁻⁷ mg/L and 2.6×10⁻⁵ mg/kg in freshwater and soil,¹⁴⁾ respectively. The PEC value of each metabolite was conveniently estimated from that of I by assuming its complete degradation to the corresponding metabolite with runoff introduction from soil, as an unrealistic worst case. The surface water PEC of each metabolite in 5% run-off scenario was estimated with FOCUS STEP2 simulation³⁷⁾ by assuming a formation of 100%. The water solubility and soil K_oc value of each metabolite were conveniently estimated by EPI-Suite,²³⁾ and the worst-case DT₅₀ of 1000 days in soil was used. The TER values in the three species range from 1.8×10⁵ to 2.5×10⁶ for the most toxic metabolite (III); they are much higher for the other metabolites by one or two orders of magnitude (Table 4). These TER values are more than a hundredfold higher than those for I (1.6×10³–2.1×10⁵). Based on these large margins of safety, the risks to the aquatic ecosystem from all the metabolites as well as the parent under typical usage were considered to be negligible.

Table 4. A worst-case assumption of PEC values of each metabolite in water and resulting TER values

Metabolite	I	II	III	IV	V	VI	VII	VIII	IX
Molecular weight (g/mol)	360	394	348	364	154	158	224	238	238
Water solubility (mg/L)a	0.5	8.06	10.43	12.61	557.9	87320	4055	1904	13620
Koc (L/kg)a	6184	43.51	220.2	87.3	38.79	2.24	21.94	6.57	2.26
PEC (ng/L)b
from I in water	0.76	0.83	0.73	0.76	0.32	0.33	0.47	0.50	0.50
from I in soilc	0.41	3.4	2.5	3.0	1.4	1.5	2.0	2.2	2.2
Lowest TER, ×106
Fish	0.0016	14	0.18	26	66	63	48	55	45
Daphnid	0.0062	15	2.5	25	51	61	47	55	43
Algae	0.21	18	1	25	59	67	49	50	34

a: Estimated by EPI-Suite (Lower value as the worst-case) except I (measured value). b: Estimated by assuming 100% degradation of I as the worst-case. c: Estimated by EU FOCUS STEP 2 simulation (ver. 1.1), based on the soil PEC of I (2.6×10⁻⁵ mg/kg¹⁴⁾ converted using an application rate of 0.02 g/ha, under the scenario of no drift and crop interception) with the worst-case assumption (DT₅₀=1000 days, 100% occurrence and 5% runoff).

References

• 1)  J. Gan, F. Spurlock, P. Hendley and D. P. Weston (eds.): “Synthetic Pyrethroids Occurrence and Behavior in Aquatic Environments,” American Chemical Society, Washington D.C., 2008.
• 2)  T. Katagi: Top. Curr. Chem. 314, 167–202 (2012).
• 3)  T. R. Roberts, D. H. Hutson, P. W. Lee, P. H. Nicholls and J. R. Plimmer (eds.): “Metabolic Pathways of Agrochemicals Part II: Insecticides and Fungicides,” The Royal Society of Chemistry, Cambridge, 1999.
• 4)  J. P. Leahey (ed.): “The Pyrethroid Insecticides,” Taylor & Francis, London & Philadelphia, 1985.
• 5)  S. J. Maund, P. J. Campbell, J. M. Giddings, M. J. Hamer, K. Henry, E. D. Pilling, J. S. Warinton and J. R. Wheeler: Top. Curr. Chem. 314, 137–165 (2012).
• 6)  C. J. Sinclair and A. B. A. Boxall: Environ. Sci. Technol. 37, 4617–4625 (2003).
• 7)  I. R. Hill: Pestic. Sci. 27, 429–465 (1989).
• 8)  V. Zitko, W. G. Carson and C. D. Metcalfe: Bull. Environ. Contam. Toxicol. 18, 35–41 (1977).
• 9)  K. E. Day and R. J. Maguire: Environ. Toxicol. Chem. 9, 1297–1300 (1990).
• 10)  EFSA: EFSA J. 9, 2159–2260 (2011).
• 11)  K. Ujihara, T. Mori, T. Iwasaki, M. Sugano, Y. Shono and N. Matsuo: Biosci. Biotechnol. Biochem. 68, 170–174 (2004).
• 12)  N. Matsuo, K. Ujihara, Y. Shono, T. Iwasaki, M. Sugano, T. Yoshiyama and S. Uwagawa: SUMITOMO KAGAKU 2005-II, 1–15 (2005).
• 13)  US EPA OPP: Pesticide Ecotoxicity Database, http://www.ipmcenters.org/Ecotox/DataAccess.cf.m (Accessed 20 Oct., 2012)
• 14)  UK: Inclusion of Active Substances in Annex I or IA to Directive 98/8/EC, Assessment Report, Metofluthrin (2010), http://circa.europa.eu/Public/irc/env/bio_reports/library?l=/assessement_directive/metofluthrin_2010pdf/_EN_1.0_&a=d (Accessed 20 Oct., 2012)
• 15)  M. Nishiyama, Y. Suzuki and T. Katagi: J. Pestic. Sci. 35, 447–455 (2010).
• 16)  R. Kodaka, Y. Suzuki, T. Sugano and T. Katagi: J. Pestic. Sci. 32, 393–401 (2007).
• 17)  H. Nishimura, Y. Suzuki, M. Nishiyama, T. Fujisawa and T. Katagi: J. Pestic. Sci. 36, 376–384 (2011).
• 18)  OECD: OECD Guidelines for the Testing of Chemicals, Section 2: Effects on Biotic Systems, http://www.oecd-ilibrary.org/environment/oecd-guidelines-for-the-testing-of-chemicals-section-2-effects-on-biotic-systems_20745761 (Accessed 20 Oct., 2012)
• 19)  ASTM: “E729-96 Standard Guide for Conducting Acute Toxicity Tests on Test Materials with Fishes, Macroinvertebrates, and Amphibians,” 1995.
• 20)  P. Mayer, R. Cuhel and N. Nyholm: Water Res. 31, 2525–2531 (1997).
• 21)  EFSA: EFSA J. 8, 1872–1944 (2010).
• 22)  F. A. P. C. Gobas, A. Opperhuizen and O. Hutzinger: Environ. Toxicol. Chem. 5, 637–646 (1986).
• 23)  US EPA: “Exposure Assessment Tools and Models, Estimation Program Interface (EPI) Suite,” http://www.epa.gov/opptintr/exposure/pubs/episuite.htm (Accessed 20 Oct., 2012)
• 24)  J. L. M. Hermens: Environ. Health Perspect. 87, 219–225 (1990).
• 25)  EFSA: EFSA J. 8, 1709–1774 (2010).
• 26)  EUROPEAN COMMISSION: “Review Report for the Active Substance Alpha-cypermethrin,” SANCO/4335/2000 final (2000).
• 27)  The Netherlands: “Competent Authority Report, transfluthrin,” https://circabc.europa.eu/sd/d/57db8369-fc84-4e2f-b3ee-c876fd434ee5/Transfluthrin_Doc%20I_public%20version_first%20draft-JUL2010.pdf (Accessed 20 Oct., 2012)
• 28)  G. M. Fahl, L. Kreft, R. Altenburger, M. Faust, W. Boedeker and L. H. Grimme: Aquat. Toxicol. 31, 175–187 (1995).
• 29)  Y. Akiyama, Y. Ariga, M. Sakamoto and Y. Yokohama (eds.): “Sourui no Seitai,” Uchidaroukakohou, Tokyo, 1986 (in Japanese).
• 30)  M. V. Haley, N. A. Chester, C. W. Kurnas, W. T. Muse and C. T. Phillips: “Toxicity of Terephthalic Acid (TPA) Smoke Mix to Algae, Daphnia, Fathead Minnows, and Earthworms,” http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA299941 (Accessed 20 Oct., 2012)
• 31)  G. W. Stratton and C. T. Corke: Environ. Pollut. (Ser. A) 29, 71–80 (1982).
• 32)  UK Environment Agency: “The Direct Toxicity Assessment of Aqueous Environmental Samples Using the Pseudokirchneriella Subcapitata Freshwater Algal Growth Inhibition Test (2008),” http://www.environment-agency.gov.uk/static/documents/Research/bluebook219_2060295.pdf (Accessed 20 Oct., 2012)
• 33)  European Commission: “Comparative Review of (Q)SARs for Acute Toxicity,” http://ihcp.jrc.ec.europa.eu/our_labs/predictive_toxicology/doc/Report_Comparative_Review_of_QSARs_for_Acute_Tox.pdf (Accessed 20 Oct., 2012)
• 34)  P. Reuschenbach, M. Silvani, M. Dammann, D. Warnecke and T. Knacker: Chemosphere 71, 1986–1995 (2008).
• 35)  US EPA: “Ecological Structure Activity Relationships (ECOSAR),” http://www.epa.gov/oppt/newchems/tools/21ecosar.htm (Accessed 20 Oct., 2012)
• 36)  M. Miyamoto, H. Tanaka and T. Katagi: SUMITOMO KAGAKU 2008-I, 26–40 (2008).
• 37)  FOCUS: “FOCUS Surface Water Scenarios in the EU Evaluation Process under 91/414/EEC,” http://focus.jrc.ec.europa.eu/sw/index.html (Accessed 20 Oct., 2012)