|Home | About | Journals | Submit | Contact Us | Français|
Varenicline is a nicotinic acetylcholine receptor (nAChR) agonist used to treat nicotine addiction but a live debate persists concerning its mechanism of action in reducing nicotine consumption. Although initially reported as α4β2-selective, varenicline was subsequently shown to activate other nAChR subtypes implicated in nicotine addiction including α3β4. However, it remains unclear whether activation of α3β4 nAChRs by therapeutically relevant concentrations of varenicline is sufficient to affect the behavior of cells that express this subtype. We used patch-clamp electrophysiology to assess the effects of varenicline on native α3β4* nAChRs (asterisk denotes the possible presence of other subunits) expressed in human adrenal chromaffin cells and compared its effects to those of nicotine. Varenicline and nicotine activated α3β4* nAChRs with EC50 values of 1.8 (1.2-2.7) μM and 19.4 (11.1-33.9) μM, respectively. Stimulation of adrenal chromaffin cells with 10 ms pulses of 300 μM acetylcholine (ACh) in current-clamp mode evoked sodium channel-dependent action potentials (APs). Under these conditions, perfusion of 50 nM or 100 nM varenicline showed very little effect on AP firing compared to control conditions (ACh stimulation alone) but at higher concentrations (250 nM) varenicline increased the number of APs fired up to 436 ± 150%. These results demonstrate that therapeutic concentrations of varenicline are unlikely to alter AP firing in chromaffin cells. In contrast, nicotine showed no effect on AP firing at any of the concentrations tested (50, 100, 250, and 500 nM). However, perfusion of 50 nM nicotine simultaneously with 100 nM varenicline increased AP firing by 290 ± 104% indicating that exposure to varenicline and nicotine concurrently may alter cellular behavior including excitability and neurotransmitter release.
Varenicline is a nicotinic receptor agonist used to treat nicotine addiction. We assessed nicotine and varenicline's effects on cell excitability via endogenously expressed α3β4 nicotinic receptors in human adrenal chromaffin cells. Therapeutic concentrations of nicotine or varenicline failed to evoke action potentials. However, simultaneous exposure to nicotine in the presence of varenicline evoked robust action potential firing. We propose that the use of varenicline and nicotine concomitantly may contribute to some of the side effects associated with varenicline's use.
nAChRs are ligand-gated ion channels that are activated by the neurotransmitters ACh and choline. In the human genome, sixteen different nAChR genes have been identified and are designated α1-α7, α9, α10, β1-β4, δ, ε, and γ (for a review of nAChRs see (Albuquerque et al. 2009)). These subunits can assemble in an impressive number of different combinations with each receptor subtype having distinct pharmacological properties. Most α subunits require co-assembly with either a different α subunit or with a β subunit but some subunits, including α7 and α9, are capable of forming homomeric nAChRs where the receptor is formed from a single gene product (Elgoyhen et al. 1994, Gerzanich et al. 1994).
Varenicline is a nAChR agonist used as a smoking cessation aid. Its mechanism of action for reducing the consumption of nicotine is thought to be mediated by partial agonism of the α4β2 subtype (Rollema et al. 2007). However, it was later shown that varenicline also activates other nAChR subtypes implicated in nicotine addiction including α6β2, α3β4, and α7 (Tammimaki et al. 2012, Mihalak et al. 2006, Capelli et al. 2011). Nicotinic compounds that have activity on off-target subtypes are known to produce significant side effects. This is particularly problematic for compounds that activate the α3β4 subtype due to its prominent expression in ganglionic neurons of the peripheral nervous system (David et al. 2010, Mao et al. 2006, Zhou et al. 2002, Poth et al. 1997, Hone et al. 2012). Notably, the cardiovascular (CV) side effect profile of varenicline suggests that it may be acting on ganglionic α3β4 nAChRs.
Although heterologously expressed α3β4 nAChRs have been shown to be activated by varenicline (Rollema et al. 2014, Stokes & Papke 2012, Mihalak et al. 2006, Campling et al. 2013) sparse information is available concerning its activity on native human α3β4 nAChRs. Thus it would be desirable to evaluate varenicline's effects on a cell population that endogenously expresses α3β4 given this subtype's relevance to nicotine addiction (Slimak et al. 2014, Jackson et al. 2013, George et al. 2012, Dwoskin et al. 2009). The ideal neuronal population in which to study the effects of varenicline may be that of the medial habenula where α3β4 nAChRs are known to be expressed. These neurons are thought to be involved in reward pathways linked to nicotine addiction (Antolin-Fontes et al. 2015, Dani & De Biasi 2013) and thus may be a neuronal population targeted by varenicline. Unfortunately, as with other neuronal populations, medial habenular neurons coexpress multiple nAChR subtypes including α4β2* and α6β2* (Grady et al. 2009, Shih et al. 2014, Scholze et al. 2012) complicating their use for the study of native α3β4 nAChRs in isolation. We recently showed that human chromaffin cells predominantly express the α3β4* subtype and express few, if any, nAChRs with canonical αxβ2 ligand-binding sites (Hone et al. 2015). Importantly, these cells lack α4- and α6-containing nAChRs and thus represent an excellent population in which to assess the effects of varenicline on native α3β4 nAChRs.
A central goal of this study was to address the running debate on whether or not therapeutic concentrations (≤100 nM) of varenicline are capable of altering the response of a cell that expresses α3β4 nAChRs. The results show that in chromaffin cells, concentrations of varenicline ≤100 nM fail to alter affect AP firing and only concentrations >100 nM were capable of increasing AP firing. In contrast, nicotine showed no effects on AP firing at concentrations ≤500nM. However, nicotine (50 nM) in combination with varenicline (100 nM) did increase AP firing. These findings suggest that varenicline alone is unlikely to affect the behavior of cells that express α3β4 nAChRs but may show effects when combined with nicotine. Our results offer important information in the context of nicotine addiction treatment with varenicline and nicotine replacement therapeutics (NRT) and may help guide the development of improved therapy strategies.
Acetylcholine chloride, (-)-nicotine tartrate, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), amphotericin B, penicillin/streptomycin, protease type XIV, collagenase Type I, poly-D-lysine hydrobromide, Red Blood Cell Lysis solution, dimethylsulfoxide, and all salts were purchase from Sigma-Aldrich (St. Louis, MO, USA). Varenicline tartrate and tetrodotoxin citrate (TTX) were purchased from Tocris Bioscience (Abindgon, UK). Dulbecco's Modified Eagle's Medium (DMEM) and Glutamax were purchased from Life Technologies (Carlsbad, CA, USA). Fetal bovine serum was from LabClinics (Barcelona, Spain) and the D-glucose from Panreac (Barcelona, Spain).
The α-conotoxins (α-Ctxs) BuIA(T5A,P6O) (Azam et al. 2010) and ArIB(V11L,V16D) (Whiteaker et al. 2007) were synthesized as previously described (Cartier et al. 1996). α-Ctx TxID was synthesized using an Apex 396 automated peptide synthesizer (AAPPTec, Louisville, KY) according to previously described methods (Hone et al. 2013).
Human adrenal chromaffin cells were isolated from male and female organ donors and cultured as previously described (Hone et al. 2015). The use of human tissue was approved by the Universidad Autónoma de Madrid's institutional review board and by the review boards of each of the hospitals. Briefly, the cells were isolated by digestion of adrenal medullary tissue using protease type XIV followed by incubation with collagenase type I. The isolated cells were maintained in DMEM culture medium at 37° C in an incubator under an atmosphere of 95% air and 5% CO2 for up to 7 days. The culture medium was changed daily by exchanging approximately 70% of the solution with fresh medium. The age of the male donors was 55.7 ± 7.4 SDM (n=12) years and 47.0 ± 16.8 SDM (n=2) years for the female donors.
To conduct electrophysiology experiments, the cells were gravity perfused with an extracellular solution composed of 145 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM D-glucose, and 10 mM HEPES. The osmolarity was 315 mOsM and the pH was adjusted to 7.4 with NaOH. The flow rate was 1.5 ml/min and was delivered by means of a polyethylene tube with an inner diameter of 0.58 mm. The outlet of this tube was place close to the cell of interest to ensure rapid solution exchange. Patch electrodes were pulled from borosilicate glass capillaries (Kimbal Chase, cat. #3400-99) using a P97 pipette puller (Sutter Instruments, Novato CA, USA). The electrodes had resistances between 1.5 and 3.0 MΩ when filled with an internal electrode solution (145 mM K-glutamate, 10 mM NaCl, 1 mM MgCl2, 10 mM D-glucose, and 10 mM HEPES; pH adjusted to 7.2 with KOH; observed osmolarity 322 mOsM). Under these conditions, the average resting membrane potential (RMP) of the cells was -60.4 ± 0.9 mV (SEM of n=29). To initiate whole-cell recordings, a stock solution of 0.5 mg/ml amphotericin B was prepared daily in dimethyl sulfoxide. Five μl of this stock solution was added to 500 μl of intracellular solution, ultrasonicated, and used to back fill the patch electrodes. All experiments were performed under a sodium lamp for light and at 22-24°C. A HEKA EPC10 amplifier (HEKA Electronik, Lambrect, Germany) was used to record agonist-evoked responses. The signals were sampled at 10 kHz and filtered at 1 kHz through a Bessel filter. The average capacitance of the cells studied was 8.9 ± 3.0 pF (SDM of n=66) and was compensated electronically. Series resistances between 7 and 15 MΩ were obtained and were compensated electronically by 70-80% in voltage-clamp mode. In current-clamp mode, the resistance was compensated by ≥80% using the bridge balance feature of Patch Master. Agonists were applied to the cell in two ways depending on the type of experiment. For all current-clamp experiments, a Picospritzer III (General Valve Corp., Fairfield, NJ, USA) was used to apply ACh to the cell by means of pressure ejection (15 psi) through a glass capillary tube (World Precision Instruments, Sarasota, FL, USA; cat. 1B200F-4) that had been pulled to obtain an opening of 1-2 μm in diameter and filled with extracellular solution containing 300 μM ACh. For agonist studies in voltage-clamp mode, a multi barrel pipette was constructed using polyethylene tubing each with an inner diameter of 0.4 mm. A single polyethylene tube with an inner diameter of 0.28 mm was used for the outlet. The flow rate of the perfusion system was approximately 850 μl/min. The agonist pulses were controlled by a valve controller triggered by the amplifier. For the agonist concentration-response experiments, 500 ms pulses of agonist were applied every 3 min. The cells were first stimulated with 300 μM ACh until steady baseline responses were achieved and then different agonists of interest were applied in ascending concentrations.
Complete methods for conducting two-electrode voltage-clamp electrophysiology of Xenopus laevis oocytes expressing nAChRs have been previously described (Hone et al. 2009). Briefly, to express α3β4 nAChRs the oocytes were injected with water containing 1 ng of cRNA encoding human α3 and 1 ng encoding β4 subunits. To express α4β2 and α7 nAChRs, 10 ng of cRNA for both α4 and β2 subunits were injected and 41 ng of cRNA for α7. The oocytes were incubated for 1-3 days at 17°C prior to use. To conduct electrophysiology experiments, the oocytes were mounted in a 30 μl perfusion chamber and stimulated with 1 sec pulses of 300 μM ACh once every 60 sec until a steady baseline response was achieved. The control solution was then changed to one containing α-Ctx TxID and the responses to ACh monitored for inhibition. For the evaluation of TxID and BuIA(T5A,P6O) on the α7 subtype, the toxin was applied in a static bath for 5 min and the responses in the presence of the toxin compared to control responses in the absence of toxin. Agonist concentration-response curves were obtained as previously described (Azam et al. 2015).
Agonist concentration-response analyses were performed using GraphPad Prism (La Jolla, CA, USA). The responses were normalized according to the following procedures: the current amplitudes for ACh-evoked responses for each cell were fit with the Hill equation to obtain the calculated plateau value for activation. The difference between the calculated plateau value and the observed response obtained at 300 μM was ≤ 3% in all cells indicating that 300 μM ACh elicited a maximal response. The observed ACh-evoked responses were then normalized to the calculated plateau value to obtain a percent response. Since 300 μM ACh was determined to produce a maximal response, in subsequent experiments the responses of all other agonists were normalized to those obtained by 300 μM ACh in the same cell. Agonist concentration-response experiments were conducted in the presence of the selective α7 nAChR antagonist α-Ctx ArIB(V11L,V16D) (Innocent et al. 2008, Whiteaker et al. 2007) to ensure that only α3β4* nAChRs were activated.
For the evaluation of the effects of TTX, α-Ctxs, nicotine, and varenicline on ACh-evoked APs in current-clamp experiments, a response was considered an AP when the membrane depolarization exceeded 0 mV, the approximate Erev for nAChR channels (Chavez-Noriega et al. 2000, Stauderman et al. 1998), to distinguish the depolarization component produced by VGICs from that produced by the nAChRs. An analysis of variance (ANOVA) and a Fischer's least significant difference (LSD) was used to determine statistical significance for the effects of varenicline and nicotine on AP firing and a t-test was used to compare changes in the RMP. Significance was determined at the 95% confidence interval. The numbers of cells per experiment (n) are reported as the SEM unless otherwise indicated. GraphPad Prism was also used to determine the IC50 values for inhibition of the ACh-evoked responses in adrenal chromaffin cells and Xenopus oocytes by TxID.
The EC50 value for activation of heterologously expressed human α3β4 nAChRs by varenicline has been reported to be in the range of 2.0-25.0 μM and 10.0-200.0 μM for nicotine (Campling et al. 2013, Rollema et al. 2014, Tammimaki et al. 2012, Stokes & Papke 2012). We confirmed the activity of varenicline on human α3β4 and α4β2 nAChRs expressed in Xenopus oocytes where it was observed to be more potent but less efficacious on α4β2 than α3β4 nAChRs (Supplemental Fig. 1). The EC50 value for activation of the α4β2 subtype was 54.3 (30.3-97.4) nM but was only 7 ± 0.4% as efficacious compared to ACh. The EC50 value for the α3β4 subtype was determined to be 26.3 (21.0-32.9) μM and showed a 96 ± 4% efficacy relative to ACh. These data confirm varenicline's partial agonism of human α4β2 nAChRs and full agonism of the α3β4 subtype.
Despite a good amount of available information concerning the pharmacology of heterologously expressed human nAChRs, very little information is available for native human α3β4 nAChRs. To determine if human adrenal chromaffin cell α3β4* nAChRs display a pharmacology similar to heterologously expressed α3β4 nAChRs, we assessed the ability of varenicline and nicotine to evoke currents in patch-clamped human adrenal chromaffin cells and compared their potencies to that of ACh. Varenicline was found to be 10-fold more potent and a full agonist (103 ± 7%) relative to ACh (Fig. 1A). Nicotine was only slightly more potent (~2-fold) and behaved as a partial agonist (80 ± 3%) relative to ACh (Fig. 1A). The maximal efficacy of varenicline and nicotine appeared to be limited by open-channel block as evidenced by currents that, in some cells, showed a slight increase in amplitude at the end of the drug application (not shown). Importantly, therapeutically relevant concentrations (100 nM) of both agonists failed to evoke substantial whole-cell currents (Fig. 1A-C). A summary of the data obtained for ACh, nicotine, and varenicline is presented in Table 1.
The experiments shown in Figure 1 provided basic parameters for the potency and efficacy of varenicline and nicotine on native α3β4* nAChRs. Significant whole-cell currents evoked by varenicline and nicotine were only observed at concentrations ≥1 μM. Nevertheless, activation of nAChRs by low concentrations of agonists may alter cell excitability. We therefore sought to determine if lower concentrations of varenicline and nicotine showed effects on AP firing in human adrenal chromaffin cells. A stimulation protocol using brief puffs of ACh to evoke APs was employed to mimic the natural stimulus the chromaffin cell would receive in vivo. This protocol consisted of stimulating the cells with 10 or 25 ms puffs of 300 μM ACh at 0.2 Hz. The perfusion solution outlet was placed as close as possible to the cell such that the ACh response was terminated in ~500 ms. Figure 2A shows the responses evoked by this stimulation protocol in a cell voltage-clamped at -60 mV (near the RMP). Figure 2B shows the responses to the same ACh stimulation protocol but in current-clamp mode at the RMP. This protocol evoked APs that were sodium channel-dependent as evidenced by the fact that the depolarization of the membrane potential triggered by ACh could be reduced to 60 ± 4% (n=4) of control values by the sodium channel antagonist TTX (Figure 2C). Similarly, after washout of the TTX, changes in the depolarization of the membrane could also be reduced and the generation of APs prevented by simultaneous perfusion of the α3β4* nAChR antagonist α-Ctx BuIA(T5A,P6O) (1 μM) and the α7 antagonist ArIB(V11L,V16D) (100 nM) (D). In the presence of the α-Ctxs, the ACh-induced change in membrane potential was reduced to 33 ± 8% (n=4) of control amplitudes. This residual response indicated that a fraction of the nAChRs remained uninhibited. To determine the fraction that was not blocked, we reassessed the actions of BuIA(T5A,P6O) in voltage-clamp mode (Vh= -60 mV) in the same cells. The cells were perfused with extracellular solution for >20 min in order to wash out the α-Ctxs from the previous application. Afterwards, the cells were again exposed to BuIA(T5A,P6O) (1 μM) which inhibited 97 ± 0.3% (n=3) of the ACh-evoked response leaving on average -178 ± 0.7 pA (n=3) of current (Fig. 3). Higher concentrations were not tested due to the amount of BuIA(T5A,P6O) that would have been required. Therefore, to determine if the remaining response was potentially mediated by other nAChRs in addition to α3β4* nAChRs, we tested the recently described α-Ctx TxID. This peptide has been shown to potently inhibit heterologously expressed rat α3β4 but shows essentially no inhibition of α7 nAChRs at 10 μM (Luo et al. 2013). To ensure that this selectivity is conserved between rat and human, we first tested TxID on Xenopus oocytes expressing either human α3β4 or α7 nAChRs. TxID was found to inhibit oocyte expressed human α3β4 nAChRs with an IC50 value of 8.7 (7.8-9.7) nM (n=4, Fig. 4A). In contrast, very little inhibition by TxID of the ACh-evoked response was observed in oocytes expressing α7 nAChRs (Fig. 4A). The α7-mediated responses to ACh were 93 ± 5% (n=5) of control values after a 5 min application of 30 μM TxID. BuIA(T5A,P6O) was also tested on oocytes expressing α7 nAChRs and no inhibition of the ACh-evoked response was observed at 10 μM (99 ± 3 %, n=5; data not shown). Thus, the expectation was that the portion of the whole-cell current mediated by α3β4* nAChRs in human adrenal chromaffin cells would be completely inhibited and any residual currents would be from activation of other non α3β4* subtypes. TxID was tested on human adrenal chromaffin cells and was found to inhibit the ACh-evoked responses with an IC50 value of 24.1 (20.1-28.5) nM (n=4; Fig. 4C). At 1 μM the α-Ctx nearly completely inhibited (99 ± 0.3%, n=4) the ACh-evoked responses (Fig. 4D). The absence of residual currents indicate that under the conditions utilized in this study, the ACh-evoked APs in human adrenal chromaffin cells are mediated by α3β4* nAChRs.
A series of experiments were conducted to evaluate the effects of varenicline and nicotine on AP firing in human adrenal chromaffin cells. The cells were stimulated using the protocol described in Figure 2 with the exception that the duration of the ACh pulse was reduced to 10 ms. Approximate EC50 concentrations of each compound were chosen to determine if intrinsic differences on AP firing could be observed. Under these conditions, perfusion of 4 μM varenicline reduced the number of ACh-evoked APs to 34 ± 10% (n=6) of control values (Fig. 5A). After washout by perfusion with control solution (Fig. 5D), the cells were again exposed to 4 μM varenicline for 1 min (Fig. 5E). In the absence of ACh stimulation, varenicline evoked 8 ± 3 APs in 5 cells and no APs in 1 cell. Next we tested an approximate EC50 concentration of nicotine (20 μM) and observed effects on AP firing similar to those observed with varenicline. Nicotine evoked a burst of APs in 4/6 cells and in all cells completely abolished the ability of ACh to evoke APs after a ~10-15 second exposure (Fig. 6A-C). The number of ACh-evoked APs evoked during the 1 min perfusion with nicotine was reduced to 38 ± 9% (n=6) compared to control conditions (Fig. 6B). Nicotine was also capable of evoking APs in the absence of ACh stimulation in 5/7 cells. In these 5 cells, nicotine elicited 8 ± 2 APs during the 1 min perfusion (Fig. 6E). Interestingly, in both the ACh stimulated and non-stimulated conditions the membrane potential remained depolarized for the 1 min duration of exposure to the agonists. The membrane potential at the end of the 1 min perfusion of 4 μM varenicline was -17 ± 1 mV (n=6; Fig. 5E) and -15 ± 2 mV (n=6) in the presence of 20 μM nicotine (Fig. 6E). In a separate group of cells, we tested nicotine and varenicline in voltage-clamp mode (Vh= -60 mV) and found that 4 μM varenicline reduced the ACh-evoked currents to 8 ± 3% (n=4) of control values and 20 μM nicotine reduced currents to 13 ± 3% after a 1 min perfusion (Fig. 7). These results suggest that activation of only a fraction of the α3β4* nAChRs is required to maintain the membrane in a depolarized state.
In the previous experiments, we found that ~EC50 concentrations of varenicline and nicotine altered the cells ability to fire APs in response to ACh and were capable of evoking APs in the absence of ACh stimulation. However, EC50 concentrations of these compounds are not achieved clinically in the plasma of humans. Concentrations of varenicline achieved in human plasma following a daily 1-2 mg dose are approximately 50-100 nM (Faessel et al. 2006, Kikkawa et al. 2011, Ravva et al. 2009). Nicotine concentrations rise to 50-500 nM in tobacco users but in the clinical setting, only ~90-100 nM concentrations are achieved from the use of nicotine patches (Gorsline et al. 1992, Sobue et al. 2005). To determine if clinically relevant concentrations of varenicline affected the chromaffin cell's response to stimulation with ACh, we perfused the cells first with 50 nM followed by 100 nM of the drug. Higher non-clinical concentrations of 250 nM and 500 nM varenicline were also tested to compare responses obtained with the same concentrations of nicotine. The cells were continuously perfused over the entire range of concentrations and each concentration was perfused for a 1 min duration each. Under these conditions, no significant increases in the number of APs fired were observed in the presence of 50 nM or 100 nM varenicline (125 ± 25%, n=6, P> 0.05, Fig. 8B and 204 ± 99%, n=6, P> 0.05, Fig. 8C, respectively) compared to control conditions (Fig. 8A). However, robust increases were observed in the presence of 250 nM varenicline (436 ± 150%, n=6, ***P< 0.001, Fig. 8D). At a concentration of 500 nM varenicline, the number of APs fired was 225 ± 59% (P> 0.05, Fig. 8E) of control values. These results suggest that therapeutically relevant concentrations of varenicline are unlikely to alter the adrenal chromaffin cells' response to ACh. Nevertheless, in the absence of ACh stimulation, fluctuations in the membrane potential of the cell were observed in the presence of 50 nM and 100 nM varenicline (Fig. 8G, H) indicating that varenicline produced low levels of α3β4* nAChR activation at these concentrations. These fluctuations did not result in a statistically significant number of APs fired (1.8 ± 1.8; P > 0.05, n=6 and 0.25 ± 0.25, P > 0.05, n=6, for 50 nM and 100 nM, respectively, Fig. 8F-H). Additionally, no changes in the RMP were observed in the presence of 50 nM or 100 nM varenicline compared to controls (-62.4 ± 0.9 and -60.6 ± 1.5 vs -65.2 ± 1.4, respectively; P > 0.5; n=5; Fig. 8F-H). However, 250 nM and 500 nM varenicline depolarized the membrane such that the threshold for the firing of APs was achieved. The membrane potential of the cell in the presence of 250 nM and 500 nM varenicline was depolarized to -36.3 ± 3.0 mV (n=6) and -35.3 ± 5.6 mV (n=6) which resulted in the cell firing 130 ± 15 (**P < 0.01, n=6) and 49 ± 16 (P > 0.05, n=6) APs, respectively (Fig. 8I, J).
In contrast to the results observed with varenicline, nicotine showed very little effect on AP firing. No significant increases in the number of APs fired were observed at any of the nicotine concentrations tested (50, 100, 250, and 500 nM; P> 0.05; n=6; Fig. 9A-E) nor changes in the RMP (-62.0 ± 2.3, -61.7 ± 2.5, -58.2 ± 5.2 and -60.2 ± 3.0 vs -61.3 ± 2.7, respectively; P > 0.5; n=6; Fig. 9F-J). Nevertheless, nicotine was not completely without effect as evidenced by fluctuations of the membrane potential observed in the presence of 250 nM and 500 nM nicotine. These fluctuations were particularly evident in traces where nicotine was perfused alone in the absence of ACh stimulation (Fig. 9I, J).
Figure 2 demonstrates that varenicline is ~10-fold more potent at activating chromaffin cell α3β4* nAChRs than nicotine and this may account for the relative lack of effect observed with nicotine (Fig. 9) compared to varenicline (Fig. 8) when perfused at the same concentrations. Nevertheless, nicotine in the presence of varenicline may have additive effects on AP firing. Such a condition often occurs clinically during the initial stages of varenicline treatment when patients continue to use tobacco or when NRTs are used as an adjunct therapy. When we perfused the cells with 50 nM nicotine together with 100 nM varenicline we observed a 290 ± 104% (*P < 0.05, n=5; Fig. 10B) increase in the number of APs fired compared to control conditions (Fig. 10A). However, a statistical comparison of the values obtained under these conditions to those obtained by perfusion with 100 nM varenicline alone (Fig. 8B) indicated no additive effects of nicotine on AP firing (P > 0.05). Increasing the concentrations of nicotine further to 100 nM, 250 nM, and 500 nM resulted in responses that were 155 ± 72% (P> 0.05, n=5, Fig. 10C), 120 ± 33% (P > 0.05, n=5, Fig. 10D), and 155± 22% (P > 0.05, n=5, Fig. 10E), respectively, of control conditions (Fig. 10A). Repeating this protocol, but without ACh stimulation, perfusion of 50 nM nicotine together with 100 nM varenicline resulted in the cells firing an average of 61 ± 31 APs (***P < 0.001, n=4; Fig. 10G) compared to control conditions (Fig. 10F). Further increases in nicotine concentrations to 100 nM, 250, nM, and 500 nM evoked 16 ± 9, 11 ± 4, and 23 ± 9 APs respectively, but these values were not statistically significant (P > 0.05, n=4; Fig. 10H-J). We also compared the number of APs fired in response to perfusion with 50 nM nicotine together with 100 nM varenicline to perfusion with 100 nM varenicline alone (Fig. 8G) and found no statistical difference (P > 0.05) again suggesting no additive effects of nicotine of AP firing. Lastly, a statistically significant change in the steady-state membrane potential of the cells during perfusion with 50 nM, 100 nM, 250 nM and 500 nM nicotine in the presence of 100 nM varenicline was observed. The RMPs under these conditions were -29.0 ± 3.4 mV (***P < 0.001, n=4; Fig. 10G), -32.5 ± 4.8 mV (**P < 0.005, n=4; Fig. 10H), -37.3 ± 6.6 mV (*P < 0.05, n=4; Fig. 10I), and -37.8 ± 6.3 mV (*P < 0.05, n=4; Fig. 10J), respectively, compared to control values (-58.3 ± 1.3 mV, Fig. 10F).
Nicotine is one of the most widely used drugs of abuse and the ubiquitous expression of nAChRs throughout the central and peripheral nervous systems means that multiple physiological processes may be affected by exposure to this alkaloid. Likewise, these systems may also be affected by administration of smoking cessation therapeutics such as varenicline. Varenicline's mechanism of action for reducing the consumption of nicotine is believed to be mediated by partial agonism of central nervous system α4β2 nAChRs (Coe et al. 2005, Rollema et al. 2010). However, varenicline has also been shown to activate other human nAChR subtypes implicated in nicotine addiction including α6β2β3, α3β4, and α7 (Stokes & Papke 2012, Bordia et al. 2012, Capelli et al. 2011, Campling et al. 2013). The reported potency and efficacy of varenicline on heterologously expressed human α3β4 nAChRs ranges from 2-25μM (Campling et al. 2013, Rollema et al. 2014, Stokes & Papke 2012). We found that varenicline activated native α3β4* nAChRs in human adrenal chromaffin cells with an EC50 of ~2 μM. We also found that varenicline behaved as a full agonist of native α3β4* nAChRs (Fig. 1A, C) similar to the activity reported by others (Campling et al. 2013, Rollema et al. 2014, Jensen et al. 2014) and in this study (Fig. 1) for heterologously expressed α3β4 nAChRs. Other reports suggest that varenicline acts as a partial agonist of α3β4 nAChRs (Stokes & Papke 2012, Arias et al. 2015) and its effects can be influenced by the particular subunit present in the “auxiliary” 5th position of the receptor complex. For example, the presence of an α5 subunit has been reported to reduce efficacy (Rollema et al. 2014, Stokes & Papke 2012, Tammimaki et al. 2012). Reports of nicotine efficacy also vary but it has generally been shown to be a partial agonist of heterologously expressed human α3β4 nAChRs (George et al. 2012, Campling et al. 2013, Stokes & Papke 2012, Wang et al. 1996) in agreement with the results presented here for native α3β4* nAChRs in human adrenal chromaffin cells (Fig. 1A, B). Nicotine's efficacy also can be influenced by the particular subunit present in the 5th position. As with varenicline, the presence of an α5 subunit reduces efficacy while a β2 subunit increases efficacy relative to ACh (Stokes & Papke 2012). Interestingly, α5 and β2 subunit mRNA transcripts have been detected in human adrenal chromaffin cells leading to the possibility that more than one α3β4 nAChR subtype may be expressed by these cells (Hone et al. 2015).
While both varenicline and nicotine activated whole-cell currents in human adrenal chromaffin cells, the concentrations at which these currents were evoked exceeded those that are normally achieved in human plasma during routine administration. Peak nicotine concentrations found in the plasma of smokers can reach ~500 nM (Benowitz et al. 2009, Henningfield et al. 1993, Gourlay & Benowitz 1997) but concentrations of varenicline generally only reach ~50-100 nM. Though unable to evoke substantial whole-cell currents at nM concentrations, low levels of nAChR activation by varenicline and nicotine may lower the threshold for firing APs by increasing the excitability of the cell membrane. We tested this hypothesis by examining the effects of varenicline and nicotine on human adrenal chromaffin cell's ability to fire APs either in response to ACh or by perfusion of the agonists alone. Stimulation of the cells in current-clamp mode by ms puffs of ACh evoked APs that could be inhibited by the sodium channel antagonist TTX or by nAChR antagonists (Fig. 2). These APs were mediated by the α3β4* subtype as evidenced by experiments that showed that the α3β4 nAChR antagonist α-Ctx BuIA(T5A,P6O) inhibited >97% of the ACh-evoked currents in these cells (Fig. 3). Additional experiments were performed using the potent and highly selective α3β4 antagonist α-Ctx TxID. This α-Ctx inhibited ~99% of the ACh-evoked currents in both human chromaffin cells and in oocytes heterologously expressing human α3β4 nAChRs (Fig. 4A-D). These experiments confirmed our previous report that α3β4* nAChR are the predominant subtype expressed by human adrenal chromaffin cells. (Hone et al. 2015)
The effects of low level activation of nAChRs on membrane excitability are not easily measured in whole-cell voltage-clamp experiments. Voltage-clamp experiments measure the amount of current applied by the amplifier to keep the voltage of the membrane constant at an experimentally set value. This is essential for isolating the activation of ligand-gated ion channels from the responses of VGICs. Current-clamp electrophysiology, on the other hand, can measure the response of the cell without artificially setting the voltage of the membrane and allows the cell to respond in a more physiological manner upon exposure to a particular stimulus. The response may include the recruitment of VGICs and may assess more closely the physiological response of the cell compared to whole-cell voltage-clamp experiments. To assess the effects of varenicline and nicotine on membrane excitability, the cells were stimulated with 10 ms puffs of ACh to evoke APs and then perfused with either varenicline or nicotine over a range of concentrations in order to determine if therapeutically relevant concentrations produced an effect. For comparison, we first perfused varenicline or nicotine at high concentrations (~EC50 values) and observed that varenicline and nicotine initially evoked a burst of APs but then inhibited the ability of the cell to fire APs in response to ACh by maintaining the cell membrane at depolarized potentials (Fig. 5 and and6).6). Interestingly, the voltage of the membrane remained substantially depolarized throughout the 1 min exposure to the agonists. In voltage-clamp mode, perfusion of the same concentrations of varenicline or nicotine inhibited the ACh-evoked currents by >90% after a 1 min exposure (Fig. 7). These observations, combined with those in Figs. 2 and and3,3, suggest that activation of less than <10% of the α3β4* nAChRs present in human adrenal chromaffin cells is sufficient to substantially affect membrane potential and aligns with previous results using ACh (Perez-Alvarez & Albillos 2007). Similar effects have been observed in Xenopus laevis oocytes where continuous perfusion of varenicline (1 μM) resulted in steady-state currents mediated by α4β2 nAChRs that lasted for the duration of agonist exposure (Papke et al. 2011).
In general, when therapeutically relevant concentrations (50 nM or 100 nM) of varenicline were perfused, no effect on the adrenal chromaffin cell's response to ACh stimulus was observed (Fig. 8B, C). However, when the cells were exposed to 250 nM varenicline, 83% (5/6) of the cells responded robustly with increased AP firing (Fig. 8D). In contrast, cells perfused with nicotine up to 500 nM generally showed no alteration in their response to ACh (Fig. 9A-E). Nevertheless, nicotine was not completely without effect as evidenced by increased fluctuations of the membrane potential which were particularly pronounced at 500 nM (Fig. 9J). Thus, even in the absence of direct effects on ACh-evoked AP firing, low concentrations of varenicline and nicotine may increase the excitability of the cell membrane via activation of α3β4 nAChRs and may influence different cellular processes. Further experiments were performed to determine if nicotine exposure in the presence of varenicline would produce additive effects on the cells response to ACh. Such a circumstance might occur during concomitant use of varenicline and NRTs such as nicotine patches. We found that perfusion of nicotine (50 nM) in the presence of varenicline (100 nM) increased AP firing relative to control conditions (290 ± 104%) (Fig. 10A, B) but no statistical differences were found when these values were compared to those obtained by perfusion with 100 nM (204 ± 99%) varenicline alone (Fig. 8C).
Overall, the results obtained in this study suggest that at therapeutic concentrations, varenicline alone are unlikely to substantially alter AP firing in chromaffin cells. Concentrations >100 nM (Fig. 8C), however, did evoke a significant number of APs over control conditions (Fig. 8A). This observation seems to correlate with clinical data in that doses of varenicline that produce plasma concentrations >100 nM produce significant side effects. We also note that concentrations >100 nM would produce very little if any increase in α4β2 receptor activation (Supplemental Fig. 1). This observation is supported by data that demonstrate that a single 0.5 mg dose of varenicline produces concentrations in the brain that saturate α4β2 receptor binding sites (Lotfipour et al. 2012). Thus, concentrations >100 nM or the addition of another nicotinic receptor agonist such as nicotine may not have added benefits in terms of α4β2 receptor activity. Furthermore, clinical data examining NRTs such as gum or lozenges containing nicotine or nicotine patches in combination with varenicline are ambiguous with respect to smoking cessation rates (Koegelenberg et al. 2014, Baker et al. 2016).
Meta-analyses of clinical data suggest that varenicline use is associated with increased CV events in a small percentage of patients (Singh et al. 2011, Harrison-Woolrych et al. 2012) but the exact mechanisms by which these side effects are produced remain unknown. We observed that nM concentrations of varenicline caused fluctuations in the membrane potential of adrenal chromaffin cells suggesting an increase in membrane excitability as a result of low levels of nAChR activation. It's tempting to speculate that treatment with varenicline may increase chromaffin cell excitability, and hence catecholamines release, sufficiently to elevate circulating catecholamine levels. Another possibility is that nicotine in combination with varenicline may increase cell excitability sufficiently to increase catecholamine release and thus cause CV side effects. This circumstance may occur in patients who begin treatment with varenicline and continue to use nicotine containing products. However, to our knowledge, direct effects of varenicline in combination with nicotine, either from smoking or NRTs, on circulating catecholamine levels in humans have not been examined. Thus, it is unclear whether potential alterations in the release of catecholamines produced by combination therapy would elevate plasma levels sufficiently to produce CV effects. More data is needed to determine if varenicline use is associated with elevated catecholamine levels and whether or not such increases would be significant enough to produce CV side effects.
We thank Dr. Paul Whiteaker at the Barrow Neurological Institute for critical review of the manuscript. This work was supported by a Marie Curie Fellowship grant [NRHACC-329966 to A.J.H], and by grants from the Spanish Ministerio de Ciencia y Tecnología [BFU2012-30997 and BFU2015-69092 to A.A.] and the U.S. National Institutes of Health [GM103801 and GM48677 to J.M.M].
ARRIVE guidelines have been followed:
=> if No, skip complete sentence
=> if Yes, insert “All experiments were conducted in compliance with the ARRIVE guidelines.”
Conflicts of interest: none
=> if ‘none’, insert “The authors have no conflict of interest to declare.”
=> otherwise insert info unless it is already included