*2.1. CHS Stimulates ATP Hydrolysis by Wild-Type (WT) ABCG5*/*G8*

α

Despite the known physiological role of ABCG5/G8 in biliary and intestinal cholesterol secretion, only indirect evidence of sterol-coupled transporter activity was detected by using steroid mimetics, such as androstane or bile acids [32,33]. In this study, we investigated a direct sterol-coupled ATPase activity by using CHS, a cholesterol mimetic that is more soluble in aqueous solution. First, to overcome low sensitivity of detecting the ABCG5/G8 ATPase activity using previous protocols, we optimized the ATPase assay for ABCG5/G8 by adopting a previous assay [34] and a colorimetric bismuth citrate-based detection approach [35]. As described and explained in Section 4, this optimized assay significantly reduces the background noise due to cloudiness by phospholipid/cholate/DDM mixtures, which improves the detecting sensitivity of liberated inorganic phosphate within the first few minutes and allows us to calculate more accurate rates of ATP hydrolysis. We show here that CHS can significantly stimulate ABCG5/G8-mediated ATP hydrolysis when co-incubated with sodium cholate (a bile acid) and *Escherichia coli* polar lipids (Figure 2). Using 5 mM ATP, the basal activity of ABCG5/G8 was calculated as 160 ± 15 nmol/min/mg (*n* = 4), similar to reported values, whereas, in the presence of CHS, the specific ATPase activity of ABCG5/G8 reached 565 ± 30 nmol/min/mg (*n* = 8), 3–4-fold higher than that in the absence of CHS (Figure 3A,B). Absence of cholate was unable to activate the ATP hydrolysis, consistent with the previous studies (data not shown) [33]. In addition, the activity was inhibited either by orthovanadate, an ATPase inhibitor [36] (Figure 3C), or by a catalytically deficient mutant ABCG5WT/G8G216D (G8-G216D) [18], which displayed no ATP hydrolysis (Figure 3A,B). The specific activity of ATP hydrolysis by ABCG5/G8 is by far the highest in comparison with the previously reported values [33]. −

**Figure 2.** (**A**) Chemical structures of cholesterol, cholesteryl hemisuccinate (CHS), and cholic acid (cholate). Source: PubChem. (**B**) Schematic illustration of sterol-coupled ATPase activity of ABCG5/G8. Dodecyl-maltoside (DDM)-purified ABCG5/G8 (light/dark-gray surface) is preincubated with phospholipids and cholate. Addition of CHS (four-ringed steroid structure) stimulates hydrolysis of ATP to ADP and inorganic phosphate (Pi) in the presence of the divalent magnesium ions (Mg2+). Using the colorimetric and bismuth citrate-based assay, the liberated Pi is then captured by ammonium molybdate in the presence ascorbic acid. The color is developed upon mixing with bismuth citrate and sodium citrate, and the absorbance was measured at 695 nm. See details in Section 4.

**Figure 3.** ATPase activity of ABCG5/G8. The ATP hydrolysis was used as a measure of ABCG5/G8 ATPase activity at 37 ◦C in conditions with 5 mM ATP and 4.1 mM CHS. The protocol is entailed in Section 4. (**A**) Data points are presented as the means ± standard deviations from 4–8 independent experiments using 2–4 independently purified proteins; where not visible, the error bars are covered by the plot symbols. A linear regression, plotted from the first 12 min, is used to calculate the specific activities. (**B**) Bar graphs show the specific activities of ATP hydrolysis by wild type (WT) in the presence and absence of CHS and the catalytically deficient mutant G8-G216D in the presence of CHS. The specific activity of WT in the absence of CHS is regarded as the basal ABCG5/G8 ATPase activity. (**C**) Bar graphs represent the percentage inhibition of ABCG5/G8 ATPase activity by 0.015 mM orthovanadate, where a *p*-value of <0.0001 (marked as \*\*\*\*) was obtained using ordinary one-way ANOVA (Prism 8).
