Equations for estimating binary mixture toxicity: Methyl-2-chloroacetoacetate-containing combinations.

Mixture toxicity was determined for 30 A+B combinations. Chemical A was the reactive soft electrophile methyl-2-chloroacetoacetate (M2CA), and chemical B was one of 30 reactive or non-reactive agents. Bioluminescence inhibition in Allovibrio fischeri was measured after 15-, 30-, and 45-minutes of exposure for A, B, and the mixture (MX) with EC(x) (i.e., EC(25), EC(50), and EC(75)) values being calculated. Concentration-response curves (CRCs) were developed for A and B at each exposure duration and used to create predicted CRCs for the concentration addition (CA) and independent action (IA) mixture toxicity models. Likewise, MX CRCs were generated and compared with model predictions, along with the calculation of additivity quotient (AQ) and independence quotient (IQ) values. Mixture toxicity vs. the models showed various combined effects, including toxicity that was slightly greater than IA and/or CA, consistency with CA, IA or both models, effects that were less toxic than expected for either model and antagonism. Simple linear regression analyses of time-dependent toxicity (TDT) data showed very strong correlations (r(2) ≥ 0.997) for B-TDT vs. the average TDT for A and B. Likewise, for both CA and IA, multiple linear regression analyses showed strong correlations (r(2) > 0.960) between MX EC(x) and either CA EC(x) and AQ(x) or IA EC(x) and IQ(x) values at each exposure duration. The results show that analyses of binary mixture toxicity data produced linear relationships resulting in equations that can effectively predict such toxicity.