Fluoxetine Exposure and Real-time Quantitative PCR (RT-qPCR) verification
A fluoxetine exposure experiment was done to determine if the 15 genes are indeed linked to environmental stress. Fluoxetine is present in the antidepressant Prozac®, the most widely prescribed psychoactive drug in the market. Fluoxetine acts as a selective serotonin reuptake inhibitor (SSRI) and has been frequently reported to have disruptive effects in non-target species [44,45]. After 7 days acclimatizing in aerated natural water, the clams in the experimental groups were exposed to Fluoxetine hydrochloric (TCI, Tokyo Chemical Industry Co., Ltd., Japan; product number F0750, >98%; CAS: 56296-78-7). The final fluoxetine concentrations in the water were 0.05, 0.5 to 5 µg/L which were refer to the concentrations of the fluoxetine in the aquatic environment and correspond to concentrations used in previous studies examining the toxicity of fluoxetine to bivalves [46,47]. Three replicates were prepared for both the control and fluoxetine treated groups. The aquaria were kept at constant temperature (20±1°C), pH (7.8±0.2) and oxygen saturation (96±2%) in 30 L glass aquaria (30 individuals per aquaria).Water was changed daily and fresh doses of fluoxetine were added; clams were fed every 48h with Chlorella vulgaris and Scenedesmus obliquus. There was no observed mortality during the experimental period. After 30 days exposure, the digestive gland in the control and treated groups were collected. At least three independent biological replicates for each sample were harvested. Total RNA was isolated using SV Total RNA Isolation System (Promega, USA) according to the manufacturer’s protocols. The quantity and quality of total RNA was confirmed as before.
Expressions of the 15 identified genes in the control and fluoxetine exposure clams were measured by real time quantitative PCR (RT-qPCR). Primers for each target gene were designed with Primer Express software v3.0.1 (Life Technology, USA). We used β-actin as an internal control [20]. Specificity of primers sets throughout the range of detection was confirmed by the observation of single amplification products of the expected size and Tm, and optimized by performing a standard curve for each primer pair. Over the detection range, the linear correlation (R2) between the mean Ct and the logarithm of cDNA dilution was >0.99 (range from 0.992 to 0.999) in each case, and efficiencies were between 93.1% to 106.8% [48]. The sequences, PCR product sizes, annealing temperatures and PCR efficiencies for each primer pair are shown in Table 1. The cDNA was synthesized according to manufacturer’s instructions from 2 µg of total RNA treated with RQ1 DNase (Promega, USA), using random hexamers and M-MLV reverse transcriptase (Promega). A total of 2 μg RNA from each tissue was reverse transcribed in a final volume of 25 μL and incubated for 1 h at 37°C after an initial denaturation step at 70°C for 10 min. The cDNAs were stored at -20°C until further use. qPCR amplification was performed using an ABI 7500 real-time quantitative PCR system (Life Technology, USA) in a total volume of 25 µL, consisting of the GoTaq® SYBR Green qPCR Master Mix (Promega, USA). The thermal cycle parameters used were as follows: 2 min at 95°C, 40 cycles of 15 s at 95°C and 1 min at 60°C. To confirm that only one PCR product was amplified and detected, a melting curve analysis of amplification products was performed at the end of each PCR reaction. Results were analyzed based on the delta-delta Ct method [49]. Experiments were performed in triplicate and repeated three times with similar results. Statistical analysis was performed with the SPSS (version 16.0; USA) and OriginPro® (version 8.0). One-way ANOVA (p<0.05) was performed using OriginPro® to test the differences of gene expressions between control and fluoxetine treated clams.
10.1371/journal.pone.0079516.t001 Table 1  Analyzed genes and their specific primers.

R