*3.1. Cytotoxicity Assay*

Cell viability was expressed as a percentage of the absorbance value obtained from the media only treated control group (Figure 1). Even though there were statistically significant differences in Caco-2 cell viability between control and *R. acetosa* treated groups at concentrations of 20, 10, 2 and 1 μg/mL, the cell viability values were high enough to study (96.4% ± 1.3%, 95.0% ± 2.1%, 95.0% ± 1.2% and 95.5% ± 2.1% at concentrations of 20, 10, 2 and 1 μg/mL, respectively). It is suggested that there is a negligible cytotoxic effect of *R. acetosa* on the Caco-2 cells at the concentration range of 1 to 100 μg/mL.

There was no significant difference on the cell viability on HEK293 cells at the concentration ranges of 1 to 50 μg/mL. The cytotoxic effect of *R. acetosa* was only detected on HEK293 cells at a concentration of 100 μg/mL with the value of 80.9% ± 11.7%. This result suggests a dose window of *R. acetosa* extract for the experiment using HEK293 cells. It also indicates that the inhibitory effect of *R. acetosa* on fexofenadine uptake discussed in Section 3.2 was not due to the cytotoxic effects of *R. acetosa* on HEK293 cells at the concentration range tested.

**Figure 1.** Cytotoxicity of *R. acetosa* extract in (**A**) Caco-2 cells and (**B**) HEK293 cells (*n* = 6). Con—media only treated control; Veh—vehicle treated group; \*—*p* < 0.05 compared to media only treated control group.

#### *3.2. P-gp Inhibition Test of Anthraquinones and R. acetosa Extract*

To determine inhibitory effect of anthraquinones on *P*-gp, an MDR kit was used. The accumulated amount of fluorescent dye in the cells represented the *P*-gp inhibition activity. The measured fluorescence intensity is expressed as a percentage of the fluorescence intensity in the control group and is shown in Figure 2. The verapamil, chrysophanol-8-*O*-β-<sup>d</sup>-glucoside and emodin treated groups displayed significantly different fluorescence intensities in comparison to those of the control group. However, the chrysophanol-8-*O*-β-<sup>d</sup>-glucoside and emodin treated groups showed significantly higher fluorescence intensities than the control group, with average values of 121.4% ±2.3% and 147.2% ±12.4%, respectively (mean ± standard deviation). This result suggests that chrysophanol-8-*O*-β-<sup>d</sup>-glucoside and emodin affect the efflux of fluorescent dye from Caco-2 cells through *P*-gp inhibition. This is consistent with previous findings that emodin inhibits *P*-gp [18]. It is thus reasonable to sugges<sup>t</sup> that herbal drug containing chrysophanol-8-*O*-β-<sup>d</sup>-glucoside and emodin may also inhibit *P*-gp.

**Figure 2.** P-gp inhibitory effect of anthraquinones in Caco-2 cells. Cells were treated with 10-μM anthraquinones or 100-μM verapamil (*n* = 3). Con—vehicle treated control; Ver—verapamil; 1—chrysophanol; 2—chrysophanol-8-*O*-β-<sup>d</sup>-glucoside; 3—emodin; 4—emodin-8-*O*-β-<sup>d</sup>-glucoside; 5—physcion; 6—physcion-8-*O*-β-<sup>d</sup>-glucoside; \*—*p* < 0.05 compared to control group.

The effects of *R. acetosa* extract on the *P*-gp were also assessed using an MDR kit. The measured fluorescence intensity is expressed as a percentage of the fluorescence intensity in the control group and is shown in Figure 3. There was no significant difference in fluorescence intensity between the control and the *R. acetosa* extract treated group. The significant inhibitory effect at the 95% confidence interval was only detected in the verapamil group used as a positive control. Although *R. acetosa* extract contains chrysophanol-8-*O*-β-<sup>d</sup>-glucoside and emodin at concentrations of 0.77% ± 0.12% and 0.94% ± 0.15% (*w*/*w*), respectively [27], inhibitory effects on *P*-gp could not be detected from *R. acetosa* extract at the concentrations tested in this assay. The concentrations of emodin and chrysophanol-8-*O*-β-<sup>d</sup>-glucoside in *R. acetosa* extract may not be high enough to inhibit *P*-gp in Caco-2 cells.

**Figure 3.** P-gp inhibition test of *R. acetosa* extract using an MDR kit in Caco-2 cells (*n* = 6). Con—vehicle treated control; Ver—100-μM verapamil; \*—*p* < 0.05 compared to control group.

#### *3.3. Fexofenadine Uptake Test with OATP1A2*/*SLCO1A2 Transfected HEK293 Cells*

The decreased fexofenadine uptake in the OATP1A2/SLCO1A2 transfected HEK293 cells represented the inhibitory effects on OATP1A2. The accumulated amount of fexofenadine in the OATP/SLCO1A2 transfected cells (the control group) was higher than that of untransfected cells (untransfected control group), which means that the OATP1A2 gene was transfected and expressed sufficiently in the control group. In addition, the fexofenadine uptake was lower in the verapamil cotreated group than in the transfected control group. In the *R. acetosa* extract co-treated group, the uptake amounts of fexofenadine in the OATP1A2 transfected HEK293 cells were significantly lower than that in the control group (Figure 4). This result suggests that *R. acetosa* extract could affect the absorption of fexofenadine through the inhibition of OATP1A2.

**Figure 4.** Inhibitory effect of *R. acetosa* extract on fexofenadine uptake in OATP1A2/SLCO1A2 transfected HEK293 cells (*n* = 6). Con—OATP1A2/SLCO1A2 transfected control; Nov—untransfected control; Ver—100-μM verapamil; \*—*p* < 0.05 compared to control group.
