Registration Dossier

Endpoint:

adult fish: sub(lethal) effects

Type of information:

experimental study

Adequacy of study:

key study

Study period:

1997-1999

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Study included atypical endpoints which have no toxicological relevance.

Principles of method if other than guideline:

Trout fed DecaBDE-treated diet for 16, 49, ir 120 d. 49-D group depurated for 71 d. DecaBDE was mixed with corn oil, mixed with Barents Sea cod that had been homogenized with 3% gelatin. 24 g portions were dried and fed to fish after soaking in water. Trout were fed either control cod chips or treated cod chips at 7.5 - 1- mg DecaBDE/kg bw/day. Analysis of liver and muscle. Some physiologic and biochemical parameters measured.

GLP compliance:

no

Analytical monitoring:

yes

Vehicle:

yes

Test organisms (species):

Oncorhynchus mykiss (previous name: Salmo gairdneri)

Test type:

flow-through

Water media type:

brackish water

Limit test:

yes

Total exposure duration:

120 d

Remarks on exposure duration:

also 49 d exposure with 71 d depuration

Post exposure observation period:

yes.

Hardness:

no info

Test temperature:

appr. 14 to 24 degrees C

pH:

no info

Dissolved oxygen:

no info

Salinity:

7.0-7.5 permillage

Nominal and measured concentrations:

0 or 7-10 mg/kg bw/d were the dose levels.

Reference substance (positive control):

no

Kierkegaard et al. (1999) investigated the uptake in trout of Dow FR-300-BA following administration in food. Rainbow trout were force-fed homogenized cod containing the suspended test article for a period of 16, 49 and 120 d. Doses ranged between 7.5 and 10 mg/kg/d.

Dow FR-300 -BA has not been manufactured since the 1980s. The published composition was described as 77.4% of the DecaBDE, 21.8% NonaBDEs and 0.8% OctaBDEs (Norris et al 1973, 1974, 1975). Kierkegaard et al. did not provide the composition of the mixture tested, but stated that the technical product contained "detectable amounts of non- (NoBDE), octa- (OcBDE), and heptabromodiphenyl ethers (HpBDE)" (pg 1612). Fig 2, pg 1614, inidcates peaks corresponding to Nona- and OctaBDEs in the "DeBDE technical product". Additionally, very small peaks corresponding to the heptaBDEs can be seen the "DeBDE prepared cod chips".

Results of physiologic and biochemical tests from control and exposed fish were compared using Student's t test by day of treatment except for those analyzed at depuration day 71.

Effects of treatment were not observed on condition or mortality. Although measured, body weights, lenghts and liver weights were not reported. An increase in the liver to body weight was reported in treated compared to control fish on d 120. However, the authors note that "control fish showed signficantly decreased liver weight and condition factor (especially male fish), probably as a result of water temperature" from days 16 -49 (pg 1613). Inspection of Table 1 indicates the DecaBDE-treated fish on d 16, 49 and 120 had better body condition factors than the control fish (although not statistically significant). The better condition of the DecaBDE-treated fish, coupled with the decrease in condition and liver weight of control fish from days 16 -49, explains the statistical difference in d 120 liver to body weight ratio in DecaBDE-treated fish. DecaBDE likely did not, as the publication infers, increase liver weight; rather the control fish were in poor condition and had lower liver weights as a result.

Differences compared to controls in blood lactate (treated higher) and lymphocyte (treated lower) levels or numbers at d 120 were reported. Information on historical values for these endpoints in fish were not provided, and thus it is difficult to assign biological significance to the statistical result. Blood lactate is not a typical endpoint used to assess toxicology. Lactate levels could have been higher in treated fish due to their better condition if condition was a factor in the speed/amount of swimming, e.g. the controls not swimming as much as usual due to poorer condition. No effect on EROD, ECOD or transketolase activity was found. Presumably these enzyme activities were measured in liver, although the tissue analyzed was not identified.

Only a very small amount of the test material was taken up during the 120 -day exposure phase. Uptake was estimated to be 0.02 –0.13% of the dose after 120 days of exposure based on the muscle concentrations of the total hexa-to DecaBDE isomers. Uptake of BDE-209 (e.g. the decabrominated congener) was estimated at only 0.005% of the dose, and declined significantly during depuration. No evidence of debromination of the test article to BDE 47, 99 or 100 was found, and the authors concluded “no evidence of debromination to these congeners was found in this study" (pg 1616). Kierkegaard et al. also stated that no evidence of brominated dibenzofurans or methoxylated PBDEs fromation was found.

Measured concentrations in liver and muscle were reported for only BDEs 47, 99, 100, 153, 154 and 209. Meausred concentrations of the remaining BDE congeners discussed in the publication were not provided.

Some hexa-, hepta-, octa- and nonaBDPO congeners’ concentrations increased with exposure duration in liver and muscle. Some of these congeners were not detectable in the technical product (Fig 2) and Kierkegaard et al. speculated that their presence might be the result of a metabolic process or a more efficient absorption of trace amounts initially present in the food or test article. Kierkegaard et al. was not able to distinguish between these two possibilities. A third possibility, not considered in Kierkegaard et al., is that these hexa-, hepta-, congeners were present in the test article but not detected, and slowly increased in fish tissue over time to detectable levels over the 120 d test period as a result of slow elimination. (Fig 2 indicates octa- and nonaBDEs in the technical product and treated cod chips plus heptas in the treated cod chips.) This third possibility is considered more likely. BDE congeners are known to have increasing bioaccumulation potential and half-lives with decreasing numbers of bromine atoms.

In reference to BDE 153 and 154, the authors' claim that "HxBDES did not decrease during the depuration period" is not supported by the data (pg 1615). When corrected for concentrations reported in control fish, BDE 154 and 153 clearly decreased after 71 d deperation compared to d49 concentrations in muscle. In liver, BDE154 decreased after depuration and BDE153 remaining approximately constant. Further, it is important to note that the measured concentrations were very low <~1 ng/g muscle and <1 0~0.1ng/g liver.

Complicating interpretation of this study is that Fig 2 (pg 1614) did not include chromatograms from control fish on days 49 and 120. Only day 16 is shown. Thus, it it not known if the reported increases in hexas, heptas, and octas in treated fish were contributed, at least in part, by the cod used as fish food. Notably, control fish were shown on day 16 to contain tetra-, penta- and hexaBDEs.

The results of this bioaccumulation study are consistent with previous work showing insignificant bioaccumulation of DecaBDE in fish, do not provide evidence that DecaBDE is debrominated metabolically, and indicate that metabolic debromination of DecaBDE is not the source of tetra- and pentaBDE congeners detected in wild-caught fish.

Validity criteria fulfilled:

not specified

Conclusions:

After 120 days of feeding DecaBDE-treated food at 7.5 -10 mg DecaBDE/kg bw/d, uptake by trout was approximately 0.005% of the dose. Adverse effects of treatment were not detected. The results of this bioaccumulation study are consistent with previous work showing insignificant bioaccumulation of DecaBDE in fish, do not provide evidence that DecaBDE is debrominated metabolically, and indicate that metabolic debromination of DecaBDE is not the source of tetra- and pentaBDE congeners detected in wild-caught fish.

Executive summary:

Kierkegaard et al. (1999) investigated the uptake in trout of Dow FR-300-BA following administration in food. Rainbow trout were force-fed homogenized cod containing the suspended test article for a period of 16, 49 and 120 d. Doses ranged between 7.5 and 10 mg/kg/d.

Dow FR-300 -BA has not been manufactured since the 1980s. The published composition was described as 77.4% of the DecaBDE, 21.8% NonaBDEs and 0.8% OctaBDEs (Norris et al 1973, 1974, 1975). Kierkegaard et al. did not provide the composition of the mixture tested, but stated that the technical product contained "detectable amounts of non- (NoBDE), octa- (OcBDE), and heptabromodiphenyl ethers (HpBDE)" (pg 1612). Fig 2, pg 1614, inidcates peaks corresponding to Nona- and OctaBDEs in the "DeBDE technical product". Additionally, very small peaks corresponding to the heptaBDEs can be seen the "DeBDE prepared cod chips".

Results of physiologic and biochemical tests from control and exposed fish were compared using Student's t test by day of treatment except for those analyzed at depuration day 71.

Effects of treatment were not observed on condition or mortality. Although measured, body weights, lenghts and liver weights were not reported. An increase in the liver to body weight was reported in treated compared to control fish on d 120. However, the authors note that "control fish showed signficantly decreased liver weight and condition factor (especially male fish), probably as a result of water temperature" from days 16 -49 (pg 1613). Inspection of Table 1 indicates the DecaBDE-treated fish on d 16, 49 and 120 had better body condition factors than the control fish (although not statistically significant). The better condition of the DecaBDE-treated fish, coupled with the decrease in condition and liver weight of control fish from days 16 -49, explains the statistical difference in d 120 liver to body weight ratio in DecaBDE-treated fish. DecaBDE likely did not, as the publication infers, increase liver weight; rather the control fish were in poor condition.

Differences compared to controls in blood lactate (treated higher) and lymphocyte (treated lower) levels or numbers at d 120 were reported. Information on historical values for these endpoints in fish were not provided, and thus it is difficult to assign biological significance to the statistical result. Blood lactate is not a typical endpoint used to assess toxicology. Lactate levels could have been higher in treated fish due to their better condition if condition was a factor in the speed/amount of swimming, e.g. the controls not swimming as much as usual due to poorer condition. No effect on EROD, ECOD or transketolase activity was found. Presumably, these enzyme activities were measured in liver, although the tissue analyzed was not identified.

Only a very small amount of the test material was taken up during the 120 -day exposure phase. Uptake was estimated to be 0.02 –0.13% of the dose after 120 days of exposurebased on the muscle concentrations of the total hexa-to DecaBDE isomers. Uptake of BDE-209 (e.g. the decabrominated congener) was estimated at only 0.005% of the dose, and declined significantly during depuration. No evidence of debromination of the test article to BDE 47, 99 or 100 was found, and the authors concluded “no evidence of debromination to these congeners was found in this study" (pg 1616). Kierkegaard et al. also stated that no evidence of brominated dibenzofurans or methoxylated PBDEs fromation was found.

Measured concentrations in liver and muscle were reported for only BDEs 47, 99, 100, 153, 154 and 209. Meausred concentrations of the remaining BDE congeners discussed in the publication were not provided.

Some hexa-, hepta-, octa- and nonaBDPO congeners’ concentrations increased with exposure duration in liver and muscle. Some of these congeners were notdetectable in the technical product (Fig 2) and Kierkegaard et al. speculated that their presence might be the result of a metabolic process or a more efficient absorption of trace amounts initially present in the food or test article. Kierkegaard et al. was not able to distinguish between these two possibilities. A third possibility, not considered in Kierkegaard et al., is that these hexa-, hepta-, congeners were present in the test article but not detected, and slowly increased in fish tissue over time to detectable levels over the 120 d test period as a result of slow elimination. (Fig 2 indicates octa- and nonaBDEs in the technical product and treated cod chips plus heptas in the treated cod chips.) This third possibility is considered more likely. BDE congeners are known to have increasing bioaccumulation potential and half-lives with decreasing numbers of bromine atoms.

In reference to BDE 153 and 154, the authors' claim that "HxBDES did not decrease during the depuration period" is not supported by the data (pg 1615). When corrected for concentrations reported in control fish, BDE 154 and 153 clearly decreased after 71 d deperation compared to d49 concentrations in muscle. In liver, BDE154 decreased after depuration and BDE153 remaining approximately constant. Further, it is important to note that the measured concentrations were very low <~1 ng/g muscle and <1 0~0.1ng/g liver.

Complicating interpretation of this study is that Fig 2 (pg 1614) did not include chromatograms from control fish on days 49 and 120. Only day 16 is shown. Thus, it it not known if the reported increases in hexas, heptas, and octas in treated fish were contributed, at least in part, by the cod used as fish food. Notably, control fish were shown on day 16 to contain tetra-, penta- and hexaBDEs.

The results of this bioaccumulation study are consistent with previous work showing insignificant bioaccumulation of DecaBDE in fish, do not provide evidence that DecaBDE is debrominated metabolically, and indicate that metabolic debromination of DecaBDE is not the source of tetra- and pentaBDE congeners detected in wild-caught fish.