Surfactants, which are comprised of non-ionic, anionic, cationic and zwitterionic structures, are a group of chemicals for which it is difficult to obtain reliable partitioning (log Kow) or bioconcentration factor (BCF) data for inclusion in current models used for performing environmental risk assessments. This issue affects a number of inter-industry organisations such as ERASM, EOSCA and ISOPA. The difficulties revolve largely around the intrinsic properties of surface active substances to adsorb to surfaces and to accumulate at phase interfaces. Despite the limitations of current methods for estimating bioaccumulation potential for a surfactant, submission of log Kow data for surfactants for the purpose of environmental risk assessments is required under REACH, though is not accepted for surfactants under the Harmonised Offshore Chemical Notification (OSPAR HOCNF) guidelines. Reported experimental BCF data for surfactants are limited [1,2], whereas QSARs based on log Kow data for estimating a surfactant BCF are necessarily based on either unreliable or unrepresentative log Kow data, since no surfactant data will be in the training and validation sets.
Most BCF models for organic contaminants take the lipid phase as a single entity and Kow as single parameter to estimate BCF in fish. It has been proposed that, for ionic compounds, distribution coefficients of the neutral and charged form to phospholipids (membranes), neutral lipids (storage lipids) and proteins are needed in BCF models . Distribution coefficients to neutral lipids may still be estimated from Kow, but sorption to membranes include more specific interactions and cannot be modelled via Kow [4-6]. Information about membrane partitioning of organic compounds is even more relevant for interpreting effect concentrations because the cell membrane is the target site for many chemicals. For compounds that elicit their toxicity via narcosis, it has been shown that the internal concentration in the cell membrane of neutral non-polar and polar compounds is relatively constant [7-9]. The development of a model or refinement of an existing model (e.g. BIONIC v2 ) is required for use in a tiered approach leading to more refined calculations of BCF values for the different classes of surfactants. The applicability domain of the model should cover a sufficiently wide range and number of surfactants. The surface activity of surfactants is not considered to be a problem when determining Klipw so long as the concentrations are kept below the Critical Micelle Concentration (CMC) . Measurement of Klipw data for typical surfactants using different methods (e.g. liposomes, IAM) and also for neutral lipids and proteins storage lipids with available experimental BCF values data are required which would allow for comparisons of the Klipw properties between the classes of surfactants and possible extrapolations to homologue series and analogue structures. Careful selection of test substances that reflect the range of the surfactant structures available and which are relatively data rich with respect to hydrophobicity data availability (e.g. log Kow, CMC, BCF, etc.) is required.
- Generation of liposome-water partition coefficients for a selection of surfactant structures. This dataset would be used to (a) extend the applicability domain of a Klipw QSAR f to surfactants and (b) once tested be used to derive estimation of BCFs for such compounds in comparison with experimental BCF data.
- Comparison of the experimental Klipw data generated in this work with other Klipw data based on the mechanistic model COSMOmic , as well as comparison of Klipw data with other hydrophobicity data available, e.g. Kow, KIAM, Kfw.
- Incorporation of Klipw data and predicted BCF data into a tiered approach which equates with key BCF triggers , ≤500, ≥2000, ≥5000 in an integrated testing strategy (ITS)
- Regular interactions/dialogue with inter-industry contacts in ERASM and EOSCA regarding other ongoing activities as well as with regulatory agencies, such as ECHA and CEFAS, over the acceptability of predicted BCF data for surfactants based on Klipw data.
- To review the available current surfactant BCF data and select suitable compounds (non-ionic and ionisable) with high quality BCF data for further Klipw determination (and subsequent comparison of experimental BCF data with BCF predictions using the new/revised model.
- To develop robust methods for measuring partitioning (Klipw) of surfactants structures to liposomes (e.g. phosphatidylcholine; POPC, storage lipids (triglycerides)) and to proteins.
- To assess the relative importance of membranes, storage lipids and proteins for uptake of selected surfactant structures from the aqueous phase into aquatic organism tissues. In particular the potential importance of the different types of phospholipids should be considered, e.g. the use of acidic/negatively charged phospholipids, such as phosphatidylserine, for cationic substances versus neutral phospholipids for zwitterionic substances .
- To measure in-vitro biotransformation rates for the selected test compounds using S9 preparations from rainbow trout liver
- Using the data to determine predicted BCF values for the selected surfactants and compare these against experimental BCF data.
Download here the full version of the RfP LRI-ECO37.