Registration Dossier

Environmental fate & pathways

Bioaccumulation: aquatic / sediment

Currently viewing:

Administrative data

Link to relevant study record(s)

Description of key information

Bioaccumulation: 
BCF: 16200 (lipid normalised, kinetic)
BMF: 0.63 (lipid-normalised steady-state)
Depuration rate constants from BCF study: 0.0294 d-1(1.1 µg/l) and 0.0179 d-1(15 µg/l).

Key value for chemical safety assessment

BCF (aquatic species):
16 200 L/kg ww

Additional information

Steady-state BCF values of 7060 l/kg (1.1 µg/l measured) and 1950 l/kg (15 µg/l measured) and kinetic BCF values of 13300 l/kg (1.1 µg/l measured) and 5250 l/kg (15 µg/l measured) were determined in a reliable study conducted in Pimephales promelas according to an appropriate test protocol, and in compliance with GLP. Lipid normalised (to 5%) values are: BCFss= 8610 l/kg (1.1 µg/l measured) and 2380 l/kg (15 µg/l measured) and BCFk= 16200 l/kg (1.1 µg/l measured) and 6400 l/kg (15 µg/l measured).

Steady-state BCF values of 7060l/kg(1.1 µg/l measured) and 1950l/kg(15 µg/l measured) and kinetic BCF values of 13300l/kg(1.1 µg/l measured) and 5250l/kg(15 µg/l measured) were determined in a reliable study conducted in Pimephales promelas according to an appropriate test protocol, and in compliance with GLP (Drottar, 2005). Lipid normalised (to 5%) values are: BCFss= 8610 l/kg (1.1 µg/l measured) and 2380 l/kg (15 µg/l measured) and BCFk= 16200 l/kg (1.1 µg/l measured) and 6400 l/kg (15 µg/l measured).

A second BCF study (CERI 2010) (only available in Japanese) was conducted using a different species (Cyprinus carpio).Steady-state BCF (BCFss) values were calculated to be 12617 and 12049 for the 1.03µg/l and 0.0981 µg/l treatment groups, respectively.The depuration half-life was estimated by the authors as 22 days for the1.03 µg/l treatment group and 19 days for the 0.0981 µg/l. The lipid content in the fish appears to be presented as 5.45% – 5.96% (average 5.71%). Normalising the BCFssto 5% lipid content would reduce thessSSLto 11050 and 10550 for the 1.03 µg/l and 0.0981 µg/l treatment groupsrespectively.

A kinetic BCF was not determined in this studykto be carried out. Using the sequential method described in OECD TG 305, k2values of 0.0314 d-1(1.03 µg/l) and 0.0355 d-1(0.0981 µg/l) have been estimated from a plot of ln[Cfish] against time for the depuration phase. The uptake rate constants (k1) have been calculated, based on a one compartment, first order kinetic model, as 545 L/kg/d (1.03 µg/l) and 601 L/kg/d(0.0981 µg/l); the BCFkvalues are therefore 17400 L/kg and 16900 L/kg for the 1.03 µg/l and 0.0981 µg/l treatment groups respectively (lipid normalised BCFkof 15240 L/kg and 14800 L/kg respectively).

At both treatment levels, the concentration in fish measured on day 1 of the depuration phase was higher than that measured at the end of the uptake phase. It would therefore appear that steady state had most likely not been reached during the uptake phase. Fish growth rates do not appear to be reported in this study.-1estimated for the uptake phase. If equivalent fish growth took place during the CERI (2010) D5 study, the kinetic BCF may be considerably higher.

A fish feeding study is also available (Drottar, 2006).0.63k1)by the total depuration rate (k2) was determined to be 0.3 When adjusted for lipid concentrations, the kinetic BMF is 1.0. The growth corrected kinetic BMF value (i.e. BMFKg) was calculated and reported to be 1. Therefore, the reported growth corrected values are not considered valid for the determination of bioaccumulation.food in this study was very highly dosed (500 µg/g of 14C-D5 nominal; 458 µg/g mean measured), which may limit the applicability of the values obtained.

 

A second GLP dietary accumulation test (CERI 2011) using D5 was carried out in carp (Cyprinus carpio) and conducted according to OECD TG 305 dietary exposure test (draft version 10 of August 31st2010).Steady-state does not appear to have been reached during the 13 day uptake phase. Steady-state BMF values are not reported by the authors.

The uptake rates (k1) and depuration rates (k2) and kinetic BMF values were determined using Berkeley Madonna software, to fit the data using a first order one-compartment model.

-1was applied.Ak1value of 0.00954 day-1 andk2value ofday-1 were calculated. Thekinetic BMF (K) value was calculated to be. The lipid normalised,kinetic BMF value was calculated to be.

The kinetic BMF values and uptake rates (k1) and depuration rates (k2) were also determined using the OECD test guideline (OECD TG 305 Draft, 2010). Growth correction was applied. Fish growth rates were determined independently (via linear regression of natural log wet weight verses time) during the uptake and depuration phases, yielding two statistically different first order growth constants of 0.0272 day-1and 0.0224 day-1, respectively for the test group.Ak1value of 0.00771 day-1 andk2value ofday-1were calculated. Thekinetic BMF (BMFk) value was calculated to be 0.343. The lipid normalised,kinetic BMF value was calculated to be 0.956.

Fish bioconcentration (BCF) studies are most validly applied to substances with log Kowvalues between 1.5 and 6. Practical experience suggests that if the aqueous solubility of the substance is low (i.e. below ~0.01 to 0.1 mg/l) (REACH Guidance R.11; ECHA, 2017), fish bioconcentration studies might not provide a reliable BCF value because it is very difficult to maintain exposure concentrations.Dietary bioaccumulation (BMF) tests are practically much easier to conduct for poorly water-soluble substances, because a higher and more constant exposure to the substance can be administered via the diet than via water.In addition, potential bioaccumulation for such substances may be expected to be predominantly from uptake via feed, as substances with low water solubility and high Kocwill usually partition from water to organic matter.

However, there are limitations with laboratory studies such as BCF and BMF studies with highly lipophilic and adsorbing substances. Such studies assess the partitioning from water or food to an organism within a certain timescale. The studies aim to achieve steady-state conditions, although for highly lipophilic and adsorbing substances such steady-state conditions are difficult to achieve. In addition, the nature of BCF and BMF values as ratio values, means that they are dependent on the concentration in the exposure media (water, food), which adds to uncertainty in the values obtained.

For highlylipophilic and adsorbingsubstances, both routes of uptake are likely to be significant in a BCF study, because the substance can be absorbed by food from the water.

Dual uptake routes can also occur in a BMF study, with exposure occurring via water due to desorption from food, and potential egestion of substance in the faeces and subsequent desorption to the water phase. Although such concentrations in water are likely to be low, they may result in significant uptake via water for highly lipophilic substances.

i.e. kg) can be subtracted from the overall depuration rate constant (k2). In short, the uptake rate constant is divided by the growth-corrected depuration rate constant to give the growth corrected kinetic BCF or BMF value. However, recent scientific discourse on this topic has pointed out that correcting for growth in the depuration phase andnotlikewise accounting for the effects of lack of growth in the uptake phase (i.e.with regards to reduced feeding rate or respiration rate for a non-growing fish), results in an equation where the laws of mass balance are violated (Gobas et al., 2019). Essentially, the uptake parameters of the kinetic BCF or BMF caluculation (i.e. k1) are those of a growing fish, but the depuration parameters are altered to reflect no growth (i.e. k2- kg). Based on this criticisim of the growth dilution correction, these calculations are not considered best practice for the assessment of bioaccumulation (Gobas et al., 2019)

Gosset al. (2013) put forward the use of elimination half-life as a metric for the bioaccumulation potential of chemicals. Using the commonly accepted BMF and TMF threshold of 1, the authors derive a threshold value for keliminationof >0.01 d-1(half-life 70d) as indicative of a substance that does not bioaccumulate.

Depuration rates from BCF and BMF studies, being independent of exposure concentration and route of exposure, are considered to be a more reliable metric to assess bioaccumulation potential than the ratio BCF and BMF values obtained from such studies.

Adepuration rate constantk2– kg)of 0.00939d-1was obtained from the D5 BMF study (Drottar 2006). This value may not be valid due to the very high loading of the food in this study potentially overloading metabolic/elimination pathways. This depuration rate is therefore not taken into account in the assessment of bioaccumulation.

Two further studies (Domoradzi et al., 2015a and 2015b, in publication) report metabolism rates for D4 and D5. In Domoradzki et al., 2015a, a 96-h metabolism study was conducted in rainbow trout whereby a 15 mg [14C] D4 or [14C] D5/kg bw single bolus oral dose was administered via gavage. Of the administered dose, 79% (D4) and 78% (D5) was recovered by the end of the study. Eighty-two percent and 25% of the recovered dose was absorbed based on the percentage of recovered dose in carcass (69% and 17%), tissues, bile and blood (12% and 8%) and urine (1%) for D4 and D5, respectively. A significant portion of the recovered dose (i.e. 18% for D4 and 75% for D5) was eliminated in faeces. Maximum blood concentrations were 1.6 and 1.4µg D4 or D5/g blood at 24 h with elimination half-lives of 39 h (D4) and 70 h (D5). Modelling of parent and metabolite blood concentrations resulted in metabolism rate constants (km) of 0.15 (D4) and 0.17 day-1(D5).

In Domoradzki et al., 2015b, Whole Body Autoradiography (WBA) was conducted in conjunction with dietary bioaccumulation studies with [14C] D4 and [14C] D5 to assess distribution and metabolism in rainbow trout. Comparison of radioactivity to D4 parent concentrations in liver extracts indicated the presence of one or more metabolites. WBA conducted during the depuration phase showed that the highest amounts of radioactivity were present in the gastrointestinal (GI) or digestive tract, liver and fat (adipose) tissue, with lesser amounts in other tissues. There was a significant amount of radioactivity in the gall bladder, with moderate amounts in the liver and contents of the digestive tract. Continued presence of parent D4 in the digestive tract contents at 42 days depuration demonstrates elimination of parent D4 via enterohepatic circulation. Similarly, a [14C] D5 bioaccumulation study analysis of radioactivity concentration and WBA provide evidence that D5 is metabolized and eliminated via the digestive tract over time. WBA findings support in vivo 96-h metabolism studies with rainbow trout.

A conservative km value of ≥0.01 day-1may be estimated from these study data and is indicative of a substance that does not bioaccumulate.

Burkhard et al., 2012 has described fugacity ratios as a method to compare laboratory and field measured bioaccumulation endpoints. By converting data such as BCF and BSAF (biota-sediment accumulation factor) to dimensionless fugacity ratios, differences in numerical scales and units are eliminated.

Fugacity is an equilibrium criterion and can be used to assess the relative thermodynamic status (chemical activity or chemical potential) of a system comprised of multiple phases or compartments (Burkhard et al., 2012). At thermodynamic equilibrium, the chemical fugacities in the different phases are equal. A fugacity ratio between an organism and a reference phase (e.g. water) that is greater than 1, indicates that the chemical in the organism is at a higher fugacity (or chemical activity) than the reference phase.

The fugacity of a chemical in a specific medium can be calculated from the measured chemical concentration by the following equation:

f = C/Z

Where f is the fugacity (Pa), C is concentration (mol/m3) and Z is the fugacity capacity (mol(m3.Pa)).

The relevant equation for calculating the biota-water fugacity ratio (Fbiota-water) is:

Fbiota-water= BCFWD/LW/ Klwx ρl/ ρB

where BCFWD/LW is ratio of the steady-state lipid-normalised chemical concentration in biota (µg-chemical/kg-lipid) to freely dissolved chemical concentration in water (µg-dissolved chemical/l-water), Klw is the lipid-water partition coefficient, ρl is the density of lipid and ρB is the density of biota.

It can be assumed that n-octanol and lipid are equivalent with respect to their capacity to store organic chemicals, i.e. Klw= Kow. For some substances with specific interactions with the organic phase, this assumption is not sufficiently accurate. Measurement of Klw values for siloxane substances is in process. Initial laboratory work with olive oil as lipid substitute indicates that the assumption that Klw= Kow is appropriate (Reference: Dow Corning Corporation, personal communication). However, the calculated fugacity ratios presented here should be used with caution at this stage.

The table below presents fugacity ratios calculated from the BCF data for D5, using Kow for Klw. BMF values do not require adjustments because these values are already equivalent to fugacity-based values.

Table: Calculated biota-water fugacity ratios

Endpoint

Exposure concentration

BCF Value

Fbiota-water*

BCFss

1.1 µg/l

7060

1.51E-03

BCFss

15 µg/l

1950

4.17E-04

BCFk

1.1 µg/l

13300

2.84E-03

BCFk

15 µg/l

5250

1.12E-03

*Using log Kow8.03

The fugacity-based BCF directly reflect the thermodynamic equilibrium status of the chemical between the two media included in the ratio calculations. The fugacity ratios calculated are all below 1, indicating that the chemical in the organism tends to be at a lower fugacity (or chemical activity) than in the water. It should be noted however, that the BCF study may not have reached true steady-state in the timescale of the laboratory studies. The fugacity ratio indicates that uptake may be less than expected on thermodynamic grounds, suggesting that elimination is faster than might be expected on grounds of lipophilicity alone.

Trophic Magnification Factors (TMF) values for D5 have been evaluated in several aquatic systems (Powell D. E.et al. 2009 (Lake Pepin), Powell D. E.et al. 2010 (Inner and Outer Oslofjord), Powell et al.,2012 (Tokyo Bay), Borgå et al., 2012a (Lake Mjosa), McGoldricket al., 2014 (Lake Erie), Borgå K.et al., 2013 (Lake Mjosa and Lake Randsfjorden), Jia, H.,et al.,2015 (Chinese Yellow Sea), NIVA 2016 (Inner Oslofjord). Further information and interpretation of the TMF data is available in Annex 3 of the CSR (see attached file).