Effect-based methods are key. The European Collaborative Project SOLUTIONS recommends integrating effect-based methods for diagnosis and monitoring of water quality

Compilation of a battery of bioassays

A wide range of EBMs has been applied successfully for both diagnostic and monitoring purposes to assess the likelihood of impacts of chemical pollution, most of them in a scientific development context for establishing robust and meaningful EBM-tools. These activities provided substantial progress towards the compilation of a useful battery of bioassays. First, a comprehensive analysis of about 1000 typical water contaminants identified 31 major MoA categories while for a substantial fraction (37%) of the compounds no information on MoAs was available [9]. Second, MoA-specific in vitro assays fit for the purpose of environmental diagnosis and monitoring are available for receptor-mediated endocrine effects, genotoxicity and mutagenicity, activation of metabolism, adaptive stress responses, photosynthesis inhibition and cell line-specific cytotoxic effects [10–12]. Thus, in vitro assays address well-described MoAs with known environmental relevance as proxies for long-term effects, although not all potentially relevant effects are covered with present test systems. To also cover chemicals with unknown and non-specific MoAs as well as with MoAs that cannot be addressed with existing MoA-specific in vitro assays and to detect specific impacts on the WFD-Biological Quality Elements, it is recommended to complement these assays with apical short-term toxicity bioassays representing at least fish (fish embryo toxicity), invertebrates (immobilization of daphnia) and algae (inhibition of cell multiplication), which represent BQEs for pelagic communities in WFD (Fig. 1). Amongst the MoA-specific in vitro assays, priority of application should be given to endocrine disruption and mutagenicity. Dioxin-like effects should be analyzed particularly in sediments [13], biota [14] and equilibrium passive samplers [15], since typical drivers of these effects are very hydrophobic and accumulate in these matrices.

Standardization and utility of test systems

In SOLUTIONS and the NORMAN network, we proposed a test battery of in vitro and in vivo bioassays and published standard operating procedures [12, 16]. The utility of EBMs is found in both the diagnosis and assessment of impacts on ecological status (cf. WFD Annex II) and the monitoring water quality status and trends (WFD Annex V).

Availability of robust enrichment tools

Solid-phase extraction (SPE) was found to be a suitable sample preparation method for environmental water samples that are to be tested in the aforementioned bioassays, with effect recovery by current SPE methods similar to recovery of individual chemicals by chemical analysis [17]. While sample enrichment is always restricted to an application domain with respect to the physicochemical properties of the chemicals, the “effect recovery” experiments indicated that for the typically applied co-polymer sorbents this domain is sufficiently broad to extract a large share of the overall toxicity of organic chemicals in water [17]. Metals and other inorganic chemicals are not addressed and need to be monitored separately. A robust mobile large-volume SPE has been developed for the use in the field, which avoids the transportation of large water volumes to the laboratory for enrichment [18] and allows for time-integrated as well as event-based sampling. Equilibrium passive sampling may be useful to concentrate hydrophobic chemicals in a biomimetic manner for subsequent EBM application [15]. For screening purposes, samplers for more hydrophilic compounds can also be used [15, 19].

Demonstration and evaluation in case studies

In SOLUTIONS, EBMs were applied in a series of case studies, where it was possible to characterize the likelihood that complex mixtures present in water systems pose specific (MoA-related) harm to the Biological Quality Elements, along a river stretch [20], around wastewater treatment plants [21, 22] and close to inflows of untreated wastewater [23]. For the selected types of example sites, mutagenic, estrogenic, androgenic and anti-androgenic effects could be established as markers for the likelihood that treated and untreated wastewater affects aquatic life. In addition, the methods allowed the impact of wastewater effluents on surface water quality to be estimated and the overall effects of chemical pollution on aquatic life and thus water quality to be assessed. The methods helped identify damage and associated causes (diagnosis, Annex II) in support of water quality management. Examples are the detection of strong anti-androgenic effects in the River Holtemme (Germany) and the identification of the fluorescence dye coumarin 47 as the cause of this effect [24], the detection of mutagenicity in the Rivers Mulde and Rhine and the identification of diaminophenazines [22] and synergistic effects of aromatic amines with natural alkaloids [25] as mutagenicity drivers. These examples may also underline how monitoring (Annex V) with EBM’s can help evaluate status and trends.

Quality/performance criteria for the benchmarking of estrogenicity bioassays have been recently investigated in an inter-laboratory comparison study [26]. In a Europe-wide demonstration program supported by SOLUTIONS, the NORMAN network, the Swiss Centre for Applied Ecotoxicology and the Joint Research Centre of the European Commission, the reliability of EBMs for screening of estrogenic compounds was analyzed to harmonize monitoring and data interpretation methods, and to contribute to the current WFD review process. Surface water and wastewater samples were collected across Europe and analyzed using chemical analyses and EBMs. The study demonstrated that the inclusion of effect-based screening methods into monitoring programs for estrogens in surface waterbodies is a valuable complement to chemical analysis because of the lower LODs of the EBMs in comparison to chemical analysis [27, 28]. Based on the results and achievements of SOLUTIONS and the NORMAN network, such comprehensive case studies should also be performed for other modes of action.

Development of effect-based trigger values (EBT)

Effect-based trigger values (EBT) have been developed for many EBMs. EBTs are expressed as bioanalytical equivalent concentrations (BEQ) and can be read across from existing EQS values for single chemicals. EBTs basically define an acceptable level of effect (translated into EBT-BEQ), in close alignment with the WFD protection goals and concentration-based Environmental Quality Standards (EQS), which proved to be useful for interpreting EBM-results in relation to the likelihood to pose harm [28, 29]. Bioassay-specific EBTs were derived by translating individual annual average (AA)-EQS for single dominant chemicals such as estrogens into EBT-BEQs [26, 28, 29], by ecological considerations and application of species sensitivity distributions [30] or by reading across from all existing EQSs using a transparent algorithm that does not require any user assumptions or judgements about the data [29]. The latter EBT-derivation method targets undefined mixtures acting according to a specific MoA. In contrast to EQSs, EBTs consider all chemicals in a mixture contributing to measured effect. Thus, this approach does not require individual guideline values for all mixture components of a mixture. Bioassay-specific EBTs are key for the interpretation of results from water quality assessment, as effects below the corresponding EBT indicate a low likelihood that the chemical mixtures pose harm whilst exceedance implies increasingly clear indications for harm to aquatic life. Importantly, the proposed approach can be applied to any bioassay provided there are sufficient effect data available.