In the patch clamp experiments, the DRG cells initially rested in the whole-cell voltage clamp mode at a constant -80 mV holding potential. When depolarized to +10 mV (100 ms), VDCCs opened to allow Ca²⁺ ions to flow into the cell inducing an inward current. NFEPP treatment dose-dependently decreased VDCC currents at both pH values (6.5 and 7.4) but to different degrees. At pH 6.5, the most effective NFEPP dose (100 µM) attenuated Ca²⁺currents by 41.8 %. This effect was significantly stronger than the effect of NFEPP at pH 7.4 (28.8%) (Figure 2a). In contrast to NFEPP, the dose-response curves upon fentanyl treatment were not different and the most effective fentanyl dose (100 µM) attenuated +10 mV-induced Ca²⁺ currents to the same degree (~39%) at both pH 7.4 and pH 6.5 (Figure 2b). These findings show that NFEPP is more effective at low pH, whereas the conventional opioid agonist fentanyl is equally effective at both acidic and physiological pH values. Naloxone dose-dependently recovered the +10 mV-induced VDCC currents during NFEPP treatment. The effects of NFEPP were completely reversed by the highest naloxone dose (100 µM) at both pH values (Figure 3a). This dose of naloxone also abolished fentanyl effects at both pH values (Figure 3b).

To investigate the role of G-protein subunits in detail, we initially used PTX which prevents the replacement of GDP by GTP at the G_αi subunits and precludes subsequent G_αβγ dissociation (Lu and Ikeda, 2016). PTX blocked NFEPP’s effect in a dose-dependent manner (Figure 4a), fully reversing the +10 mV-induced VDCC currents at the highest PTX dose (100 ng/ml). The effect of fentanyl was also blocked in a dose-dependent manner after PTX incubation (Figure 4b). Because PTX does not allow to distinguish between the involvement of G_α versus G_βγsubunits (Lu and Ikeda, 2016), we sought to selectively block G_βγ subunits by gallein (Lu and Ikeda, 2016). Gallein dose-dependently decreased NFEPP-induced effects and fully recovered +10 mV-induced Ca²⁺ currents at the highest dose at both pH values (Figure 5a). Fentanyl’s effects were also decreased and fully blocked by the most effective dose of gallein at both pH values (Figure 5b). Together, these results indicate that the effects of both opioid agonists on VDCCs are mediated via activation and subsequent dissociation of heterotrimeric G-protein complexes at both pH values.

We then performed Ca²⁺ imaging experiments to corroborate our electrophysiological results. Under our recording conditions, both NFEPP and fentanyl decreased the Fura ∆ ratio 340/380, i.e. the Ca²⁺ signals were reduced. Naloxone (100 µM) completely reversed the effect of fentanyl (Figure 6b, c). However, in NFEPP-treated neurons, the Ca²⁺ signal was significantly lower at pH 6.5 when compared to pH 7.4 (Figure 6a, c). We also examined whether the extracellular pH by itself had a role in the decrease of Ca²⁺ signals. During short (60 s) (Figure 7a) or long (20 min) (Figure 7b) term pH changes, there was no difference in the Fura ∆ ratio 340/380 at either pH 6, 6.5, 7 or 7.4. This indicates that the pH difference by itself does not influence Ca²⁺ signals in the cells and that NFEPP and fentanyl were solely responsible for the decrease. The control experiments using high potassium indicated that the signals indeed derived from the influx of extracellular Ca²⁺ into neurons (Figure 7).

Finally, we examined the effects of NFEPP on MOR phosphorylation. MOR expressing HEK293 cells were treated with either NFEPP or fentanyl at pH 6.0 - 8 (Figure 8a). The strongest NFEPP-mediated phosphorylation signal was detected at pH 6.0. This signal diminished with increasing pH values (Figure 8b). In contrast, fentanyl-induced phosphorylation was similar at all pH values (Figure 8c).