Fig 2: Conditional KO of AC3 enhances the excitability of DRG neurons and decreases the Kv currents.(A) Image showing an isolated DRG neuron expressing EGFP with the tip of a pipette during patch clamp recording. Scale bar: 20 μm. (B and C) Depolarizing current pulse required to evoke an action potential (AP) in control (B) and AC3-CKO (C) DRG neurons. (D–F) AC3 CKO reduced the rheobase required to evoke AP (D), increased the RMP (more positive; E) and had no difference in the membrane capacitance (F) of recorded neurons. *P < 0.05, ***P < 0.001; 2-tailed Student’s t test; n = 20 Ctrl and 22 CKO (cells). (G) Examples of the AP traces from control and AC3-CKO neurons. (H) AC3 CKO increased the number of APs evoked by current injection. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; 2-way RM ANOVA followed by Bonferroni’s test; n = 20 Ctrl and 22 CKO (cells). (I and J) Pharmacologically separated IA (I) and IK (J) in control and AC3-CKO DRG neurons under different holding voltage. AC3 deficiency robustly attenuated IA (I) and IK (J) densities of DRG neurons. ***P < 0.001; 2-way RM ANOVA followed by Bonferroni’s test; n = 8–11 (cells). (K–M) AC3 CKO did not shift the steady-state activation curve and half-activation voltage of Kv channels (K and M), but it left-shifted the steady-state inactivation curve of Kv channels toward hyperpolarizing direction and decreased half-inactivation voltage (more negative) in DRG neurons (L and M). **P < 0.01; 2-tailed Student’s t test; n = 14–16 (cells).

Fig 3: AC3 mediates KOR-induced analgesia peripherally in mice.(A–D) Lumbar puncture of MOR agonist DAMGO (15 ng) (A and B) and DOR agonist [D-Ala2]-deltorphin II (5 μg) (C and D) elevated paw withdrawal thresholds (PWTs) (A and C) and paw withdrawal latencies (PWLs) (B and D) in both control and AC3-CKO mice. *P < 0.05, **P < 0.01, ***P < 0.001; 2-way RM ANOVA followed by Bonferroni’s test; n = 7 or 8 Ctrl and CKO (mice). (E and F) Lumbar puncture KOR agonist nalfurafine (0.5 μg) significantly enhanced PWTs (E) and PWLs (F) in control mice but had no effect in AC3-CKO mice. *P < 0.05; 2-way RM ANOVA followed by Bonferroni’s test; n = 6 Ctrl and 6 CKO (mice). (G) I.p. injection of KOR agonist ICI204,488 (10 mg/kg) significantly attenuated capsaicin-induced nociceptive responses but had no effect in AC3-CKO mice. *P < 0.05; 2-way RM ANOVA followed by Bonferroni’s test; n = 5 or 6 Ctrl and CKO (mice). (H–K) Lumbar puncture of PTX (0.1 μg) blocked KOR agonist–induced increases in PWTs (H and I) and PWLs (J and K). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; 2-way RM ANOVA followed by Bonferroni’s test; n = 6 Veh and 6 PTX 0.1 μg (mice). Veh, vehicle; L, left paw; R, right paw; DAM, DAMGO; Delt, [D-Ala2]-deltorphin II; Nalf, nalfurafine; NS, normal saline.

Fig 4: Conditional KO of AC3 in L3 and L4 DRGs facilitates nociceptive behavioral responses in mice.(A) Western blot analysis showing a significant decrease in AC3 in L3 and L4 DRGs of CKO mice. The intensity of 3 bands including 180 kDa (glycosylated form), 130 kDa (full-length unglycosylated form), and 70 kDa (monomer form) of AC3 were calculated. *P < 0.05; 2-tailed Student’s t test; n = 5 Ctrl and CKO (mice). (B) AC3-immunoreactivity (AC3-IR, red) colocalized (Co) with EGFP (left) but not with EGFP-T2A-Cre (right) in DRG neurons. Scale bar: 50 μm. (C) Quantitative analysis showing proportion of AC3+ neurons infected with AAV and AC3 knockdown efficiency. (D) Schematic of protocol for virus injection and behavior tests. (E and F) AC3 CKO induced mechanical allodynia in von Frey (E) and paintbrush (F) tests. *P < 0.05, **P < 0.01, 2-tailed Student’s t test; n = 8 Ctrl and CKO (mice). (G) AC3 CKO induced thermal hyperalgesia. **P < 0.01, ****P < 0.0001; 2-way RM ANOVA followed by Bonferroni’s test; n = 7 Ctrl and 11 CKO (mice). (H–K) AC3 CKO facilitated nociceptive responses in 52°C (H) and 55°C (I) hot plate, noxious pinch (J) and intraplantar injection (i.pl.) of 0.1% capsaicin (K). *P < 0.05, **P < 0.01, ***P < 0.001; 2-tailed Student’s t test; n = 9 Ctrl and 10 CKO (mice). (L) I.pl. 2.5% formalin induced a significantly increased nociceptive response at the I phase in AC3-CKO mice. *P < 0.05, ****P < 0.0001; 2-way RM ANOVA followed by Bonferroni’s test (left) and 2-tailed Student’s t test (middle and right); n = 7 Ctrl and CKO (mice). BL, baseline; Ctrl, Control; CKO, AC3 CKO.

Fig 5: AC3 mediates KOR-induced enhancement of Kv channel currents in DRG neurons.(A and B) Dynorphin A significantly increased IA (A) and IK (B) densities in control DRG neurons but not in AC3-CKO DRG neurons. ****P < 0.0001; 2-way RM ANOVA followed by Bonferroni’s test; n = 8–11 (cells). (C and D) PTX (Gαi/o protein irreversible inhibitor, 1 mM) in the intracellular solution completely abolished the dynorphin A–induced increases in IA (C) and IK (D) in naive DRG neurons. *P < 0.05, ****P < 0.0001; 1-way ANOVA; n = 15–16 (cells). Dyn, dynorphin A. (E and F) RNAscope ISH showing colocalization and proportion of DRG neurons positive for AC3 mRNA (Adcy3) with KOR mRNA (Oprk). Scale bar: 50 μm. Arrowheads indicate the colocalization of positive signals for AC3 mRNA and KOR mRNA. (G) Double immunofluorescence of isolated DRG neurons showing the colocalization of AC3-IR with KOR-IR. Scale bar: 10 μm. (H) Proximity ligation assay (PLA) showing close interaction signals of AC3 with KOR in DRG slice (left) and isolated DRG neurons (right). Scale bar: 20 μm (left) and 10 μm (right). (I) Co-IP displaying that AC3 (IB) and KOR (IB) are captured by mouse anti-KOR antibody (IP) in mouse lumbar DRGs. Normal mouse IgG IP was applied as the negative control. n = 3 (mice).