Quantifying Synergy: A Systematic Review of Mixture Toxicity Studies within Environmental Toxicology

Abstract

Cocktail effects and synergistic interactions of chemicals in mixtures are an area of great concern to both the public and regulatory authorities. The main concern is whether some chemicals can enhance the effect of other chemicals, so that they jointly exert a larger effect than predicted. This phenomenon is called synergy. Here we present a review of the scientific literature on three main groups of environmentally relevant chemical toxicants: pesticides, metal ions and antifouling compounds. The aim of the review is to determine 1) the frequency of synergy, 2) the extent of synergy, 3) whether any particular groups or classes of chemicals tend to induce synergy, and 4) which physiological mechanisms might be responsible for this synergy. Synergy is here defined as mixtures with minimum two-fold difference between observed and predicted effect concentrations using Concentration Addition (CA) as a reference model and including both lethal and sub-lethal endpoints. The results showed that synergy occurred in 7%, 3% and 26% of the 194, 21 and 136 binary pesticide, metal and antifoulants mixtures included in the data compilation on frequency. The difference between observed and predicted effect concentrations was rarely more than 10-fold. For pesticides, synergistic mixtures included cholinesterase inhibitors or azole fungicides in 95% of 69 described cases. Both groups of pesticides are known to interfere with metabolic degradation of other xenobiotics. For the four synergistic metal and 47 synergistic antifoulant mixtures the pattern in terms of chemical groups inducing synergy was less clear. Hypotheses in terms of mechanisms governing these interactions are discussed. It was concluded that true synergistic interactions between chemicals are rare and often occur at high concentrations. Addressing the cumulative rather than synergistic effect of co-occurring chemicals, using standard models as CA, is therefore regarded as the most important step in the risk assessment of chemical cocktails.

Incorporating the joint toxicity of co‐applied pesticides into the ecological risk assessment process

“…For example, Coors and Frische (2011) evaluated the predictability of mixture toxicity by CA of commercially sold pesticide products for which heterogeneity of the input data was likely high. Deviation from the CA model (higher percentage of MDR values >2) was found to be higher than reported in previous review papers focused on peerreviewed literature (Belden et al 2007a;Cedergreen 2014).…”

“…In addition, toxicants could interact, meaning that one of the toxicants influences the other toxicant through an indirect mechanism. The most frequently described examples of interaction in ecotoxicology are changes in toxicokinetics in which 1 toxicant interferes with the biotransformation of the other (Cedergreen 2014). On the basis of this framework, we can describe mixture toxicity in practical terms within 3 categories: concentration addition (CA), independent action (IA), or simple interaction (SI) (Belden et al 2007a).…”

“…4). In contrast, most ecotoxicological efforts have promoted CA (Belden et al 2007a;EFSA 2013;Cedergreen 2014). In order to demonstrate how method selection can impact the results, we provide a theoretical example.…”

“…In the last 2 decades, numerous studies have extensively tested the predictability of these models for many species and contaminants. In addition, meta-analyses evaluating the accuracy of the modeling attempts have been conducted (Belden et al 2007a;Cedergreen 2014). It is clear from this research that mixtures of pesticides applied at equipotent concentrations typically result in more toxicity than would be expected from either of the individual pesticides by itself.…”

“…It is clear from this research that mixtures of pesticides applied at equipotent concentrations typically result in more toxicity than would be expected from either of the individual pesticides by itself. It is also clear that predictive models are relatively good at estimating toxicity, and cases of synergy, defined here as more than additive toxicity, are very infrequent (Belden et al 2007a;Cedergreen 2014). Thus, there is a workable approach for incorporating joint action of co-applied pesticides into the risk assessment process based on existing models and research.…”