*2.2. [11C]Tariquidar PET*

We used a previously developed PET protocol [28] which involved PET scans with the dual ABCB1/ABCG2 substrate [11C]tariquidar with co-administration of unlabeled tariquidar (12 mg/kg) to saturate cerebral ABCB1 activity and thereby selectively measure ABCG2 activity, without and with partial ABCG2 inhibition with the ABCG2 inhibitor Ko143 (5 mg/kg) [34]. PET summation images of [11C]tariquidar in APP/PS1-21 and wild-type mice are shown in Figure 3. Without Ko143 pretreatment, brain radioactivity concentrations in both mouse strains were markedly lower than in most of the surrounding head region, while in Ko143-treated animals, brain radioactivity concentrations approached the concentrations in the surrounding head region. The corresponding time-activity-curves (TACs) in whole brains are shown in Figure 4. In both mouse strains, mean TACs were higher under conditions of partial ABCG2 inhibition than under conditions without ABCG2 inhibition. Brain-to-plasma radioactivity concentration ratios (*K*p,brain) were determined as a parameter for the brain distribution of [11C]tariquidar (i.e., the ratio of PET-derived radioactivity concentration at the last time point and the radioactivity concentration in plasma measured at the end of the PET scan) [28]. In Figure 5, *K*p,brain values are shown for the two mouse strains for the three examined brain regions (hippocampus, cortex and cerebellum) without and with ABCG2 inhibition. No significant differences in *K*p,brain values were found between APP/PS1-21 and wild-type mice in any of the investigated brain regions, either for scans without or for scans with partial ABCG2 inhibition. For both mouse strains, *K*p,brain values in the hippocampus and cortex were significantly higher after partial ABCG2 inhibition (Figure 5a,b). In the cerebellum, *K*p,brain was only significantly increased after partial ABCG2 inhibition in APP/PS1-21 mice but not in wild-type mice (Figure 5c). In all three brain regions, the percentage increase in *K*p,brain of [11C]tariquidar following ABCG2 inhibition was not significantly different between APP/PS1-21 and wild-type mice (APP/PS1-21: 36–52%, wild-type: 26–41%).

**Figure 3.** Sagittal (median axis) PET summation images (0–60 min) of 6-months-old APP/PS1-21 mice and age-matched wild-type mice pretreated i.v. with tariquidar (TQD, 12 mg/kg) at 2 h and Ko143 vehicle solution or Ko143 (5 mg/kg) at 1 h prior to [11C]tariquidar PET. Whole brain region is outlined with a red broken line. All images are set to the same intensity scale (0–3 standardized uptake value, SUV).

**Figure 4.** Time-activity curves (mean <sup>±</sup> SD) of [11C]tariquidar in whole brains of (**a**) wild-type mice and (**b**) APP/PS1-21 mice pretreated with unlabeled tariquidar (TQD, 12 mg/kg) at 2 h and Ko143 vehicle solution (open circles, wild-type: *n* = 6, APP/PS1-21: *n* = 5) or Ko143 (5 mg/kg, closed circles, wild-type: *n* = 7, APP/PS1-21: *n* = 7) at 1 h prior to PET acquisition.

**Figure 5.** Regional brain-to-plasma radioactivity concentration ratios (*K*p,brain) in (**a**) hippocampus, (**b**) cortex and (**c**) cerebellum at the end of the [11C]tariquidar PET scan in 6-months-old APP/PS1-21 mice and age-matched wild-type mice pretreated with unlabeled tariquidar (TQD, 12 mg/kg) at 2 h and Ko143 vehicle solution (wild-type: *n* = 6, APP/PS1-21: *n* = 5) or Ko143 (5 mg/kg, wild-type: *n* = 7, APP/PS1-21: *n* = 7) at 1 h prior to the start of the PET scan. (ns, not significant; \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; compared to Ko143 vehicle treated animals; one-way ANOVA followed by Tukey's multiple comparison test).
