Zinc selenite

Long-term toxicity to fish

Description of key information

Reliable chronic data were only identified for freshwater fish and vary between 10 and 1000 µg Se/L. The most critical waterborne NOEC was a 258-d NOEC for pre-spawning mortality of bluegills (Lepomis macrochirus) obtained from a study with Na2SeO3. All reliable chronic NOEC values were used for the derivation of a PNECwater based on the statistical extrapolation approach.

Key value for chemical safety assessment

EC10, LC10 or NOEC for freshwater fish:

10 µg/L

Additional information

For chronic toxicity to fish, 90 studies were identified as potentially relevant. After review, 83 studies were considered relevant, including 15 reviews. Finally, 18 studies were identified reporting reliable effect concentrations. These studies include data from toxicity studies with SeO₂, H₂SeO₃, Na₂SeO₃, seleno-L- and seleno-DL-methionine. Reliable data were also available for studies using combinations of these test substances (including Na₂SeO₄). All reliable data were for freshwater fish (Pimephales promelas, Pogonichthys macrolepidotus and Brachydanio rerio (Cyprinidae), Lepomis macrochirus (Centrarchidae), Oncorhynchus mykiss and Oncorhynchus tshawytscha (Salmonidae), and Acipenser transmontanus (Acipenseridae)). Results are available from studies in which fish were exposed to waterborne Se only, dietary Se only, or a combination of waterborne and dietary Se. The available effect concentrations include NOEC values expressed as waterborne Se, dietary Se and/or internal Se.

Only results based on dissolved Se concentration in the exposure media and obtained from studies in which Se was added to the test water only, are taken into account for the derivation of a PNECwater for direct effects. Reliable NOEC values based on waterborne Se are available for 4 freshwater fish species (B. rerio, P. promelas, L. macrochirus and O. mykiss) and range from 10 to 1000 µg Se/L. The lowest NOEC was obtained in a study in which adult bluegills (Lepomis macrochirus) of 3 to 4 years old were exposed to a series of waterborne Se concentrations (added as Na₂SeO₃) in outdoor experimental streams in which fish were allowed to feed on naturally present food (Hermanutz et al., 1992). The adult fish were exposed for 258 days (pre-spawning period) + 98 days post-spawning. The NOEC of 10 µg Se/L represents a 258-d NOEC for pre-spawning mortality. The highest chronic NOEC for fish originates from a study in which fertilized eggs were exposed to a series of waterborne Se concentrations (added as SeO₂) and larval mortality was observed until 10 days post-hatch (Niimi and LaHam, 1975). All results are based on added dissolved Se concentrations. The data for Se-(D)L-methionine were not taken into account for the assessment of direct effects of selenite to aquatic organisms because there is some concern on different biochemical behaviour of this selenium containing amino acid compared to inorganic Se compounds. Because no data are available for selenate, no comparison can be made between the long-term toxicity of selenite and selenate compounds to fish.

Results for dietary toxicity of Se to fish can be used for the secondary poisoning assessment of Se in the aquatic food chain. Reliable NOECoral values (expressed as mg Se/kg diet) were obtained from studies in which Se was added to the diet only. Results are available for 6 freshwater species and values range from 3.7 to 89.8 mg Se/kg diet (dw). The lowest NOEC originates from a study in which juvenile rainbow trout (Oncorhynchus mykiss) were exposed for 20 weeks to a series of dietary Se concentrations (added as Na₂SeO₃ to a casein-torula yeast diet) (Hilton et al., 1980). After 20 weeks of dietary exposure, a NOEC of 3.7 mg Se/kg diet was determined for both survival and growth (body weight). The highest NOEC represents an 8-week NOEC for the endpoints condition factor and hepatosomatic index for white sturgeon (Acipenser transmontanus) exposed to a series of dietary Se concentrations (added as seleno-L-methionine) (Tashjian et al., 2006).

Although not used in the standard risk assessment, the available effects concentrations based on internal Se concentrations are discussed below. Reliable NOEC values based on internal Se were obtained from a series of studies in which Se was added either to the water or the diet only or added to both water and diet. The NOEC values (not considering unbounded values) varied between 3 and 12.6 mg Se/kg dw (whole body). The NOEC of 3 mg Se/kg dw results from a study in which juvenile bluegills (L. macrochirus) were exposed for 90 days to a series of dietary Se concentrations (added as seleno-L-methionine) (Cleveland et al., 1993). The NOEC was obtained for the endpoint survival. The NOEC of 12.6 mg Se/kg dw was reported by Hamilton et al. (1990) and resulted from a study in which fall Chinook salmon fingerlings (Oncorhynchus tshawytscha) were exposed for 120 days to a series of dietary Se concentrations (added as seleno-DL-methionine). The NOEC was obtained for the endpoints length and weight.

Some of the studies also reported reliable NOEC values based on Se concentrations in eggs or ovaries. It is generally accepted that tissue-based effect concentrations are more predictive of Se toxicity than diet borne and waterborne effect concentrations. From the NOEC values identified as reliable it is clear that tissue-based NOEC values show less variation than diet borne NOEC values and diet borne NOEC values in their turn show less variation than waterborne NOEC values.