38 Appendix 8, Regulation (EU) 2017/735

In forcesince 14 Feb 2017

The dietary method is included in this test method for the bioaccumulation testing of substances that cannot in practice be tested using the aqueous exposure method. The aqueous exposure method gives a bioconcentration factor, whereas the dietary method leads directly to information on feeding biomagnification potential. In many chemical safety regimes information on aquatic bioconcentration is required (for example in risk assessment and the Globally Harmonization System of Classification). Hence there is a need to use the data generated in a dietary study to estimate a bioconcentration factor that is comparable to tests conducted according to the aqueous exposure method [In the wild the route leading to greatest exposure in aqueous environments is likely to be through ingestion for very hydrophobic substances and so an estimated BCF is not strictly representative of such a substance's bioaccumulation potential.]. This section explores approaches that may be followed to do this, while recognising the shortcomings that are inherent in the estimations.

The dietary study measures depuration to give a depuration rate constant, k2. If an uptake rate constant can be estimated with the available data for the situation where the fish had been exposed to the test substance via the water, then a kinetic BCF could be estimated.

The estimation of an uptake rate constant for water exposure of a test substance is reliant on many assumptions, all of which will contribute to the estimate's uncertainty. Furthermore, this approach to estimating a BCF assumes that the overall rate of depuration (including contributory factors like distribution in the body and individual depuration processes) is independent of the exposure technique used to produce a test substance body burden.

The main assumptions inherent in the estimation approach can be summarised as follows.

Depuration following dietary uptake is the same as depuration following aqueous exposure for a given substance

Uptake from water would follow first order kinetics

Depending on the method used to estimate uptake:

uptake can be correlated with fish weight alone

uptake can be correlated with the substance's octanol-water partition coefficient alone

uptake can be correlated with a combination of fish weight and the substance's octanol-water partition coefficient

factors that can affect uptake in an aqueous exposure study in practice such as substance bioavailability, adsorption to apparatus, molecular size etc. have little effect

and, crucially:

The database (“training set”) used to develop the uptake estimation method is representative of the substance under consideration

Several publications in the open literature have derived equations relating uptake from water in fish via the gills to a substance's octanol-water partition coefficient, fish weight (1) (2) (3) (4), volume and/or lipid content, membrane permeation/diffusion (5) (6), fish ventilation volume (7) and by a fugacity/mass balance approach (8) (9) (10). A detailed appraisal of such methods in this context is given in Crookes & Brooke (11). A publication by Barber (12) focussed on modelling bioaccumulation through dietary uptake is also useful in this context as it includes contributions from gill uptake rate models. A section of the background document to the 2004 dietary protocol (13) was also devoted to this aspect.

Most of these models seem to have been derived using limited databases. For models where details of the database used to build the model are available, it appears that the types of substances used are often of a similar structure or class (in terms of functionality, e.g. organochlorines). This adds to the uncertainty in using a model to predict an uptake rate constant for a different type of substance, in addition to test-specific considerations like species, temperature, etc.

A review of available techniques (11) highlighted that no one method is “more correct” than the others. Therefore, a clear justification should be given for the model used. Where several methods are available for which the use can be justified, it may be prudent to present several estimates of k1 (and so BCF) or a range of k1 values (and BCF) according to several uptake estimation methods. However, given the differences in model types and datasets used to develop them, taking a mean value from estimates derived in different ways would not be appropriate.

Some researchers have postulated that BCF estimates of this sort require a bioavailability correction to account for a substance's adsorption to dissolved organic carbon (DOC) under aqueous exposure conditions, to bring the estimate in line with results from aqueous exposure studies (e.g. (13) (14)). Howeverl this correction may not be appropriate given the low levels of DOC required in an aqueous exposure study for a ‘worst case’ estimate (i.e. ratio of bioavailable substance to substance as measured in solution). For highly hydrophobic substances uptake at the gill may become limited by the rate of passive diffusion near the gill surface; in this case it is possible that the correction may be accounting for this effect rather than what it was designed for.

It is advised to focus on methods that require inputs for which data will be readily available for substances tested according to the dietary study described here (i.e. log KOW, fish weight). Other methods that require more complex inputs may be applied, but may need additional measurements in the test or detailed knowledge on the test substance or fish species that may not be widely available. In addition, choice of model may be influenced by the level of validation and applicability domain (see (11) for a review and comparison of different methods).

It should be borne in mind that the resulting k1 estimate, and estimated BCF, are uncertain and may need to be treated in a weight-of-evidence approach along with the derived BMF and substance parameters (e.g. molecular size) for an overall picture of a substance's bioaccumulation potential. Interpretation and use of these parameters may depend on the regulatory framework.

(1)

Sijm D.T.H.M., Pärt P. and Opperhuizen A. (1993), The influence of temperature on the uptake rate constants of hydrophobic compounds determined by the isolated perfused gills of rainbow trout (Oncorhynchus mykiss). Aquat. Toxicol. 25: 1-14.

(2)

Sijm D.T.H.M., Verberne M.E., Part P. and Opperhuizen A. (1994), Experimentally determined blood and water flow limitations for uptake of hydrophobic compounds using perfused gills of rainbow trout (Oncorhynchus mykiss): Allometric applications. Aquat. Toxicol. 30: 325-341.

(3)

Sijm D.T.H.M., Verberne M.E., de Jonge W.J., Pärt P. and Opperhuizen A. (1995), Allometry in the uptake of hydrophobic chemicals determined in vivo and in isolated perfused gills. Toxicol. Appl. Pharmacol. 131: 130-135.

(4)

Barber M.C. (2003), A review and comparison of models for predicting dynamic chemical bioconcentration in fish. Environ. Toxicol. Chem. 22: 1963-1992.

(5)

Opperhuizen A. (1986), Bioconcentration of hydrophobic chemicals in fish, in Aquatic Toxicology and Environmental Fate, STP 921, Poston, T.M. and Purdy, R., Editors. American Society for Testing and Materials, Philadelphia, PA, USA: 304-315.

(6)

Arnot J.A. and Gobas F.A.P.C. (2004), A food web bioaccumulation model for organic chemicals in aquatic ecosystems. Environ. Toxicol. Chem. 23: 2343-2355.

(7)

Thomann R.V. (1989), Bioaccumulation model of organic chemical distribution in aquatic food chains. Environ. Sci. Technol. 23: 699-707.

(8)

Hendriks A.J., van der Linde A., Cornelissen G. and Sijm D.T.H.M. (2001). The power of size. 1. Rate constants and equilibrium ratios for accumulation of organic substances related to octanol-water partition ratio and species weight. Environ. Toxicol. Chem. 20: 1399-1420.

(9)

Campfens J. and Mackay D. (1997), Fugacity-based model of PCB bioaccumulation in complex aquatic food webs. Environ. Sci. Technol. 31: 577-583.

(10)

Arnot J.A. and Gobas F.A.P.C. (2003), A generic QSAR for assessing the bioaccumulation potential of organic chemicals in aquatic food webs. QSAR Comb. Sci. 22: 337-345.

(11)

Crookes M. and Brooke D. (2010), Estimation of fish bioconcentration factor (BCF) from depuration data. Draft Report. Environmental Agency, Bristol, UK.

(12)

Barber M.C. (2008), Dietary uptake models used for modelling the bioaccumulation of organic contaminants in fish. Environ. Toxicol. Chem. 27: 755-777

(13)

Anonymous (2004), Background document to the fish dietary study protocol, document submitted to the TC-NES WG on PBT.

(14)

Gobas F. and Morrison H. (2000), Bioconcentration and biomagnification in the aquatic environment, in Handbook of property estimation methods for chemicals, Boethling, R.S. and Mackay, D., Editors. Lewis Publishers, Boca Racton, FL, USA: 189-231.