**3. Results**

#### *3.1. Single Toxicity of BPA and Heavy Metals*

Before the examination of the binary–mixture toxicities of BPA and heavy metals, the single toxicities of BPA, Cr, Cd, Pb, As, Hg, and Ni to Q67 were tested (Figure 1). Considering that the acute toxicity of the aqueous BPA solution toward bioluminescent bacteria was detected for the first time, the quantities of solutions of aqueous BPA and

BPA dissolved in methanol at the same concentrations were tested using HPLC and then compared. The results showed that the aqueous BPA and methanol dissolved BPA solutions exhibited similar retention times and peak areas (Figure S2). In terms of toxicity detection, most of these concentration–inhibition data could be fitted by logistic and dose–response models. However, the concentration–inhibition data of Hg were fitted better than others by the bi–dose–response model. These models suggested good quality, and the values of R2 were over 0.99 for all fitting models. The reduced chi–sqr values ranged from 1.57 to 30.85. The EC50 value of each contaminant was calculated by the fitting function, which indicated the toxicity of each contaminant.

**Figure 1.** The acute (**A**) and chronic (**B**) single toxicity of BPA and Cr, Cd, Pb, As, Hg, and Ni. The concentration–inhibition data were fitted using the logistic and dose–response models. The error bars indicate the standard deviations from three independent experiments.

For the acute toxicity (Figure 1A), BPA and Cd displayed the strongest and weakest toxicity to Q67, respectively. The EC50 values of these contaminants ranged from 0.62 to 13.27 mg/L. The toxicities of these contaminants were in the following order: BPA > Hg > Cr > As > Ni > Pb > Cd. In the case of chronic toxicity, Pb and Ni revealed no obvious toxicity effects to Q67 in the range of the experimental concentration (data not shown). In addition, the chronic toxicity of these contaminants suggested a time–dependent nature (Figure 1B). The chronic toxicities of BPA, Cr, Cd, and Hg decreased with the increase in exposure time. All the concentration–inhibition data fit well with the logistic model for 6 h exposure but not for 12 h exposure; thus, the exposure time for chronic exposure was 6 h in other studies. The EC50 values of these contaminants (6 h exposure time) ranged from 2.81 to 8.79 mg/L. The chronic toxicities of these contaminants were as follows: Hg > BPA > Cd > As > Cr. The difference in the acute and chronic toxicities of these contaminants was significant (*p* < 0.05).

## *3.2. Toxicities of Binary–Mixtures of BPA and Heavy Metals*

Based on the single toxicities of these contaminants, the toxicities of binary mixtures of BPA and heavy metals were analyzed at equitoxic and non–equitoxic ratios. Concentration– inhibition data were fitted using same mathematic models as the single toxicity test. For all models developed in this work, the values of R2 (all over 0.99) and reduced chi–sqr (ranged from 0.70 to 15.86) indicated the good fits. In terms of acute toxicity of the mixtures, the EC50 of these binary mixtures revealed that their toxicities were as follows: BPA + Hg > BPA + Cr > BPA + As > BPA + Ni > BPA + Pb > BPA + Cd at equitoxic ratios (Figure 2A), which was the same as the single toxicities of heavy metals. In order to validate the toxicities of mixtures of BPA + Cd and BPA + As, the toxicities of mixtures of BPA + Cd and BPA + As at non–equitoxic were measured (Figures S3 and S4). The EC50

values of BPA + Cd ranged from 0.91 to 17.42 mg/L. Indeed, the toxicities of mixtures of BPA + Cd decreased with the increase in the ratio of lg(BPA/Cd) (Figure 2B), and the same conclusion was obtained from BPA + As (Figure 2C). These results indicate that the toxicities of mixtures of BPA + Cd and BPA + As were affected by the ratio of Cd and As in the mixture, respectively.

**Figure 2.** Toxicities of mixtures of BPA and heavy metals. (**A**) BPA and heavy metals at equitoxic ratios in acute toxicity tests; (**B**,**C**) BPA and Cd, BPA, and As at non–equitoxic ratios in the acute toxicity test, respectively; (**D**) BPA and heavy metals at equitoxic ratios in the chronic toxicity test; (**E**,**F**) BPA and Cd, BPA, and As at non–equitoxic ratios in the chronic toxicity test, respectively. The error bars indicate the standard deviations from three independent experiments.

In the case of chronic mixture toxicity, the toxicities of these mixtures were shown to be BPA + Hg > BPA + Cd > BPA + As > BPA + Cr at equitoxic ratios (Figures 2D and S5), which is the same as the single toxicities of Hg, Cd, As, and Cr. Furthermore, toxicities of mixtures of BPA + Cd and BPA + As at non–equitoxic ratios were measured (Figures S6 and S7). The EC50 values of BPA + Cd at non–equitoxic ratios ranged from 3.02 to 5.86 mg/L. When the ratio of lg(BPA/Cd) was under 0, toxicities of mixtures of BPA + Cd decreased with the ratio of lg(BPA/Cd) increasing, while it increased with the increase in the ratio of lg(BPA/Cd) when the ratio of lg(BPA/Cd) was over 0 (Figure 2E); the same conclusion could be obtained from BPA + As (Figure 2F). The differences between the acute and chronic toxicities of mixtures of BPA + Cr, BPA + Cd, BPA + As, and BPA + Hg were statistically significant at equitoxic ratios as well as non–equitoxic ratios (*p* < 0.05).
