Interaction of bupropion with muscle-type nicotinic acetylcholine receptors in different conformational states

Abstract

To characterize the binding sites and the mechanisms of inhibition of bupropion on muscle-type nicotinic acetylcholine receptors (AChRs), structural and functional approaches were used. The results established that bupropion (a) inhibits epibatidine-induced Ca(2+) influx in embryonic muscle AChRs, (b) inhibits adult muscle AChR macroscopic currents in the resting/activatable state with approximately 100-fold higher potency compared to that in the open state, (c) increases the desensitization rate of adult muscle AChRs from the open state and impairs channel opening from the resting state, (d) inhibits binding of [(3)H]TCP and [(3)H]imipramine to the desensitized/carbamylcholine-bound Torpedo AChR with higher affinity compared to the resting/alpha-bungarotoxin-bound AChR, (e) binds to the Torpedo AChR in either state mainly by an entropy-driven process, and (f) interacts with a binding domain located between the serine (position 6') and valine (position 13') rings, by a network of van der Waals, hydrogen bond, and polar interactions. Collectively, our data indicate that bupropion first binds to the resting AChR, decreasing the probability of ion channel opening. The remnant fraction of open ion channels is subsequently decreased by accelerating the desensitization process. Bupropion interacts with a luminal binding domain shared with PCP that is located between the serine and valine rings, and this interaction is mediated mainly by an entropy-driven process.

Fig. 1

Effects of bupropion on agonist-induced macroscopic currents in HEK293 cells expressing mouse α1β1εδ (adult) AChRs. A: Bupropion effects from the open state. Left: Ensembled mean currents obtained from outside-out patches activated in absence (control) or simultaneous application of ACh and bupropion, without preincubation with bupropion (protocol −/+; open state). Each trace represents the average of 4–8 applications of agonist. Curves from right to left correspond to: control and recovery, 50, 100, and 200 µM bupropion. The calculated decay time constants (τ_d) are 25 ms for control and recovery curves, and 11.3, 6.3, and 4.5 ms, for 50, 100, and 200 µM bupropion, respectively. Membrane potential: −50 mV. Right: Concentration-response curve for the decrease in the decay time constant (n = 5). The calculated IC₅₀ and n_H values are summarized in Table 1. B: Effects of bupropion from the resting/activatable state. Left: Effect of bupropion application protocol +/− (resting/ activatable state): 2 min pre-incubation of bupropion following ACh application. Each trace represents the average of 4–8 applications of agonist. Curves from right to left correspond to: control and recovery, 0.25, 0.5, 1, and 5 µM bupropion, respectively. The peak current decreases with increased bupropion concentrations. Membrane potential: −50 mV. Right: Concentration-response curve for the decrease in the peak current on the resting/activatable state. The calculated IC₅₀ and n_H values are summarized in Table 1. C:Effects of bupropion application protocol on macroscopic currents. Superimposed currents responses to 300 µM ACh and bupropion concentrations correspond to IC₅₀ for the resting/activatable state (~0.4 µM; see Table 1) and for the open state (~40 µM bupropion; see Table 1) using different protocols. From right to left curves correspond to control condition (−/−protocol; τ_d = 13.4 ms), simultaneous 300 µM ACh/40 µM bupropion application without preincubation with bupropion (−/+ protocol; τ_d = 6.5 ms; peak current 99% of the control), ACh application following 2 min pre-incubation with 0.4 µM bupropion (+/− protocol; τ_d = 13.9 ms; peak current = 50.4% of the control), and simultaneous 300 µM ACh/40µM bupropion application after pre-incubation with 0.4 µM bupropion (+/+ protocol; τ_d = 6.3 ms; peak current = 47% of the control).

Fig. 2

Bupropion-induced inhibition of [³H]TCP binding to Torpedo AChRs in the desensitized (CCh-bound) (■) and resting (α-BTx-bound) (□) states. AChR-rich membranes (0.3 µM) were equilibrated (2 h) with 7 nM [³H]TCP, 1 mM CCh (■) or 1 µM α-BTx (□), and increasing concentrations of bupropion (i.e., 1 nM-200 µM). Nonspecific binding was determined in the presence of 50 (■) or 100 µM PCP (□), respectively. Each plot is the combination of two separated experiments each one performed in triplicate. From these plots the IC₅₀ and n_H values were obtained by nonlinear least-squares fit according to eq. 3. Subsequently, the K_i values were calculated using eq. 4. The calculated K_i and n_H values are summarized in Table 2.

Fig. 3

Bupropion-induced inhibition of [³H]imipramine binding to the AChR in the desensitized (CCh-bound) (○) and resting (α-BTx-bound) (●) states. AChR-rich membranes (0.3 µM) were equilibrated (2 h) with 7 nM [³H]imipramine, 1 mM CCh (○) or 1 µM α-BTx (●), and increasing concentrations of bupropion (i.e., 1 nM-200 µM). Nonspecific binding was determined in the presence of 100 µM PCP (○) or 200 µM tetracaine (●), respectively. Each plot is the combination of three separated experiments each one performed in triplicate. From these plots the IC₅₀ and n_H values were obtained by nonlinear least-squares fit according to eq. 3. Subsequently, the K_i values were calculated using eq. 4. The calculated K_i and n_H values are summarized in Table 2.

Fig. 4

Chromatograhic elution of bupropion from the CMAC-Torpedo AChR column. (A) Bupropion is eluted from the column with ammonium acetate buffer (10 mM, pH 7.4) and 15% methanol as the mobile phase, at 0.2 mL/min and 20°C. The dashed line represents the elution of bupropion from the CMAC-Torpedo AChR column pretreated with a-BTx (the immobilized AChR is mainly in the resting state), and the straight line represents the elution of bupropion from the CMAC-Torpedo nAChR column pretreated with epibatidine (the immobilized AChR is mainly in the desensitized state). (B) Bupropion is eluted from the CMAC-Torpedo AChR column pretreated with epibatidine (predominantly desensitized state) at different temperature (from right to left: 12, 16, 20, and 25°C).

Fig. 5

Van’t Hoff plots of bupropion determined by elution from the CMAC-Torpedo AChR column at different temperatures (see Fig. 4B). Ammonium acetate buffer (10 mM, pH 7.4) and 15% methanol were used as the mobile phase (0.2 mL/min) to elute bupropion from the CMAC-Torpedo AChR column at different temperatures (10–25°C). The column was pretreated with epibatidine (□) (the AChR is mainly in the desensitized state) or with α-BTx (□) (the AChR is mainly in the resting state), respectively, before bupropion elution. The plots are the results from three experiments (n = 3), where the SD error bars are smaller than the symbol size. The observed r² values are 0.995 (□) and 0.998 (○), indicating that the plots are perfectly linear. The ΔH° and ΔS° values were determined using the slope (ΔH° = −Slope. R) and y-intercept (ΔS° = y-intercept R) values from the plots, according to eq. 10, where R is the gas constant (8.314 J K⁻¹ mol⁻¹).

Fig. 6

Complex formed between R-bupropion and the Torpedo AChR (a and b) and between S-bupropion and the mouse muscle AChR (c) ion channel obtained by molecular docking. (a) Side view of the lowest energy complex showing four Torpedo subunits rendered in secondary structure mode, whereas the ligand in the neutral form is rendered in element color coded ball mode. Side views of the Torpedo (b) and mouse muscle (c) AChR subunits rendered in semitransparent surface with visible secondary structure and explicit CPK atoms of residues forming the valine (position 13’ in green) and serine (position 6’ in red) rings. The ligand in the neutral form is rendered in stick mode with hydrogen atoms not shown explicitly. On both pictures the δ subunit was removed for clarity, and the order of remaining subunits is (from left to right) α1, γ, α1, and β1.

Fig. 6

Complex formed between R-bupropion and the Torpedo AChR (a and b) and between S-bupropion and the mouse muscle AChR (c) ion channel obtained by molecular docking. (a) Side view of the lowest energy complex showing four Torpedo subunits rendered in secondary structure mode, whereas the ligand in the neutral form is rendered in element color coded ball mode. Side views of the Torpedo (b) and mouse muscle (c) AChR subunits rendered in semitransparent surface with visible secondary structure and explicit CPK atoms of residues forming the valine (position 13’ in green) and serine (position 6’ in red) rings. The ligand in the neutral form is rendered in stick mode with hydrogen atoms not shown explicitly. On both pictures the δ subunit was removed for clarity, and the order of remaining subunits is (from left to right) α1, γ, α1, and β1.