*2.5. Cell viability Assays*

β Reduced cell viability compromises the production and secretion of cellular products. Therefore, MTT assays were performed to verify whether incubation with the P-gp inhibitors verapamil and nicardipine at the final experimental concentrations (1–30 µM), would affect CHO-APP or SK-N-SH cell viability. Figure 5a shows that CHO-APP cell viability was not affected by the inhibitors. There was a marginal effect of high verapamil concentrations on SK-N-SH cell viability. However, this was not statistically significant (*p* > 0.05 compared to control) (Figure 5b).

μ

**Figure 5.** P-gp inhibitors do not affect cell viability. (**a**) CHO-APP and (**b**) SK-N-SH cells were incubated with verapamil or nicardipine at varying concentrations, DMSO, or untreated full culture medium (control) for 24 h to simulate the full experiment duration. The following day, cells were washed and treated with 0.5 mg/mL MTT in serum-free culture medium for two hours at 37 ◦C. After washing with PBS, cells were lysed with DMSO. Absorbance was measured at 550 nm. Data are presented as mean ± SEM of three independent experiments, each performed with six replicates.

### *β* β *2.6. Inhibition of P-gp Reduces Aβ Secretion from CHO-APP and SK-N-SH Cells*

β β β β β β β μ β β β μ β β To investigate whether P-gp is involved in the export of Aβ, we assessed the effect of chemical inhibition of P-gp activity on the secretion of Aβ<sup>40</sup> and Aβ<sup>42</sup> peptides into cell media. Initial experiments were conducted using CHO-APP cells, which overexpress human APP and exhibit ample P-gp expression. Considering these cells produce large quantities of Aβ, we were able to use in-house ELISAs to quantify these peptides in the supernatant. Control (untreated) cells in our experiments secreted on average 3.3 and 1.1 ng/mL Aβ<sup>40</sup> and Aβ42, respectively. Treatment of CHO-APP with verapamil for 24 h significantly reduced secretion of Aβ<sup>40</sup> (Figure 6a) and Aβ<sup>42</sup> (Figure 6b) into the media compared to control in a concentration-dependent manner. Results were most pronounced with verapamil 30 µM, which reduced Aβ<sup>40</sup> and Aβ<sup>42</sup> secretion by approximately half, compared to control. Treatment with nicardipine similarly yielded a dose-dependent effect on Aβ<sup>40</sup> secretion, with 10 µM reducing Aβ<sup>40</sup> levels to 48 ± 7.6% of that of control (Figure 6a); reductions in Aβ<sup>42</sup> levels were also significant. However, a dose-dependent relationship could not be confirmed due to variability in response to the highest nicardipine concentration (Figure 6b).

Our observations in CHO-APP are in line with findings from other groups that also report the involvement of P-gp in Aβ export in vitro, utilising human embryonic HEK293 cells transfected with APP695 [26], P-gp-transfected Lewis lung carcinoma cells [63] and LS-180 human colon adenocarcinoma cells [64]. However, the relationship between P-gp and Aβ has, until now, not been investigated in neurons where these peptides are predominantly generated. Therefore, we applied the same experimental conditions to SK-N-SH human neuroblastoma cells.

β β β β β ≤ ≤ ≤ **Figure 6.** Inhibition of P-gp reduces Aβ efflux from CHO-APP and SK-N-SH cells. CHO-APP cells were treated with verapamil, nicardipine, or DMSO, or left untreated, for 24 h before supernatant was collected and analysed for (**a**) Aβ<sup>40</sup> and (**b**) Aβ<sup>42</sup> content by in-house ELISA. Data are displayed as mean ± SEM of three independent experiments for each peptide, and were adjusted for protein concentration. Each experiment was conducted with three biological replicates, each with additional three technical replicates. SK-N-SH cells were similarly treated and analysed for (**c**) Aβ<sup>40</sup> (mean ± SEM, n = 2) and (**d**) Aβ<sup>42</sup> (mean ± SEM of one experiment run in triplicate cultures) secretion into media but using commercial ELISA kits. \* *p* ≤ 0.05, \*\* *p* ≤ 0.01, \*\*\* *p* ≤ 0.001.

β β β β β μ μ Since SK-N-SH cells secrete considerably smaller quantities of peptides (approximately 100-fold less than CHO-APP cells), the cell supernatants were analysed using commercial ELISA kits. Control SK-N-SH cells were found to secrete on average 29.0 pg/mL Aβ<sup>40</sup> and 3.6 pg/mL Aβ42. As seen in Figure 6c, pharmacological inhibition of P-gp significantly and dose-dependently reduced Aβ<sup>40</sup> secretion from SK-N-SH cells. Results were most pronounced with verapamil 30 µM and nicardipine 10 µM, which reduced levels to 42 ± 6.1% and 43 ± 4.0% of that of control, respectively. Verapamil similarly yielded a dose-dependent reduction in Aβ<sup>42</sup> secretion (Figure 6d). Interestingly, the same observation as seen in CHO-APP (Figure 6b) was also observed in SK-N-SH, with the higher nicardipine concentration producing an unexpected increase in Aβ42. Verapamil has been shown not to affect cellular Aβ<sup>40</sup> or Aβ<sup>42</sup> production [65], suggesting that the observed reductions in Aβ secretion can be attributable to reduced P-gp-mediated export. As anticipated, SK-N-SH cells consistently secreted higher proportions of Aβ<sup>40</sup> compared to Aβ<sup>42</sup> (approximately 8:1 ratio), which corresponds with previously reported in vitro data as well

as physiological ratios observed in human AD brains [52,66]. Overall, chemical inhibition using P-gp-specific inhibitors was demonstrated to suppress Aβ peptide secretion from both CHO-APP and SK-N-SH cells.
