Nicotine stimulates ion transport via metabotropic β4 subunit containing nicotinic ACh receptors

Background and Purpose Mucociliary clearance is an innate immune process of the airways, essential for removal of respiratory pathogens. It depends on ciliary beat and ion and fluid homeostasis of the epithelium. We have shown that nicotinic ACh receptors (nAChRs) activate ion transport in mouse tracheal epithelium. Yet the receptor subtypes and signalling pathways involved remained unknown. Experimental Approach Transepithelial short circuit currents (ISC) of freshly isolated mouse tracheae were recorded using the Ussing chamber technique. Changes in [Ca2+]i were studied on freshly dissociated mouse tracheal epithelial cells. Key Results Apical application of the nAChR agonist nicotine transiently increased ISC. The nicotine effect was abolished by the nAChR antagonist mecamylamine. α‐Bungarotoxin (α7 antagonist) had no effect. The agonists epibatidine (α3β2, α4β2, α4β4 and α3β4) and A‐85380 (α4β2 and α3β4) increased ISC. The antagonists dihydro‐β‐erythroidine (α4β2, α3β2, α4β4 and α3β4), α‐conotoxin MII (α3β2) and α‐conotoxin PnIA (α3β2) reduced the nicotine effect. Nicotine‐ and epibatidine‐induced currents were unaltered in β2−/−mice, but in β4−/− mice no increase was observed. In the presence of thapsigargin (endoplasmatic reticulum Ca2+‐ATPase inhibitor) or the ryanodine receptor antagonists JTV‐519 and dantrolene there was a reduction in the nicotine‐effect, indicating involvement of Ca2+ release from intracellular stores. Additionally, the PKA inhibitor H‐89 and the TMEM16A (Ca2+‐activated chloride channel) inhibitor T16Ainh‐A01 significantly reduced the nicotine‐effect. Conclusion and Implications α3β4 nAChRs are responsible for the nicotine‐induced current changes via Ca2+ release from intracellular stores, PKA and ryanodine receptor activation. These nAChRs might be possible targets to stimulate chloride transport via TMEM16A.


| INTRODUCTION
Strict regulation of transepithelial ion transport is essential for effective mucociliary clearance. Mucociliary clearance represents a crucial primary innate defence mechanism of the airways (Hollenhorst, Richter, & Fronius, 2011;Knowles & Boucher, 2002). This is facilitated by beating of the cilia of ciliated epithelial cells and depending on tightly controlled level and viscosity of the periciliary liquid assured by appropriate regulation of transepithelial ion transport.
Several chronic airway diseases may be attributed to impaired mucociliary clearance, including primary ciliary dyskinesia, cystic fibrosis, asthma and chronic obstructive pulmonary disease (COPD) (Dransfield et al., 2013;Knowles & Boucher, 2002). Disruption of transepithelial ion transport is involved in severe pulmonary phenotypes including impaired function of the cystic fibrosis transmembrane conductance regulator in cystic fibrosis, chronic obstructive pulmonary disease and cigarette smoke-induced chronic bronchitis (Clunes et al., 2012;Dransfield et al., 2013;O'Sullivan & Freedman, 2009).
In mouse tracheal epithelium, nicotine binding to apical nAChRs activates apical chloride secretion driven by basolateral potassium secretion (Hollenhorst, Lips, Weitz, et al., 2012). However, the nAChR subtype, the ionotropic or metabotropic nature of these receptors and the ion channel subtypes facilitating the secretion of Cl − are unknown. Here, we aim to identify the nAChR subtype as well as the nAChR-activated channels responsible for the ion transport changes and discuss whether this is a suitable target to activate chloride secretion to restore mucociliary clearance.

| Compliance with requirements for studies using animals
Animal studies are reported in compliance with the ARRIVE guidelines (Percie du Sert et al., 2020) and with the recommendations made by the British Journal of Pharmacology . All animal care and experimental procedures were approved by the German and European animal welfare committee and performed according to the German and European animal welfare law. Adult wild-type (WT, C57Bl/6J, RRID:IMSR_JAX:000664) and β2 −/− (Picciotto et al., 1995) or β4 −/− nAChR mice (Kedmi, Beaudet, & Orr-Urtreger, 2004) aged between 10 and 15 weeks of both sexes were used throughout the study. The β2 −/− mice were backcrossed to C57Bl/6J for 12 generations and the β4 −/− mice were backcrossed to C57Bl/6J for six generations after germline transmission. The animals were randomly chosen once they had the appropriate age. The WT mice were bred and What is already known • Non-neuronal nicotinic ACh receptors activate chloride secretion in mouse tracheal epithelium.

Wha this study adds
• α3β4 nicotinic receptors are essential for activation of apical chloride secretion in the airways.
• Activation of α3β4 nicotinic receptors mediates Ca 2+ release from intracellular stores and PKA-dependent signalling.
What is the clinical significance • α3β4 nAChR-dependent activation of Ca 2+ -dependent chloride channel TMEM16A might be beneficial in cystic fibrosis.
• Tracheal nicotinic ACh receptors might serve as possible pharmaceutical targets to stimulate chloride transport. housed in IVC cages in the animal facility of the Institute of Experimental Surgery of the Saarland University under standardized 12-h day-night cycles with free access to food and water. β2 −/− or β4 −/− nAChR mice were bred and held under specific pathogen-free conditions in the animal facility of the Medical University of Vienna. After shipping, these mice were housed in the animal facility of the Institute of Experimental Surgery of the Saarland University for 2 weeks for acclimatization and quarantine before the experiments were performed.

| Ussing chamber experiments
For Ussing chamber measurements, mice were exposed to an overdose of the narcotic isoflurane and killed by aortic exsanguination.
Then they were removed from the solution and the solution was cen- Analyses were performed with the NIS-Elements software (Nikon Instruments, Amsterdam, Netherlands).

| Experimental design and statistical analysis
The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis et al., 2018). In the Ussing chamber experiments, the values were calculated for 1-cm 2 tissue area and reported as mean ± SEM with the number (n) of the investigated tracheas and animals. In all Ussing chamber experiments, a group size of n = 5-7 animals was designed. According to our previous experience, this was identified as a suitable size to evaluate statistical significances Hollenhorst, Lips, Weitz, et al., 2012;Hollenhorst, Lips, Wolff, et al., 2012). For the calcium imaging experiments, coverslips with cells isolated from at least five different animals were measured with "N" denoting the number of animals and "n" the number of cells evaluated. All data were first analysed for normal distribution with the Kolmogorov-Smirnov test.
Afterwards, the paired or unpaired Student's t-test was applied when data passed the normality test. Datasets that did not pass the normality test were analysed with the Mann-Whitney U test. Statistical significance was assigned for P < 0.05.

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in the IUPHAR/BPS Guide to PHARMACOL-OGY http://www.guidetopharmacology.org and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20 (Alexander, Fabbro, et al., 2019.

| Nicotine activates transepithelial ion current in WT mice
First, we characterized the basic effect of nicotine on transepithelial ion transport. We applied nicotine on the apical side of the epithelium, in order to investigate possible non-neuronal effects of nAChRs. Apical application of 100 μmolÁL −1 nicotine transiently increased I SC (20.09 ± 2.89 μAÁcm −2 , Figure 1a,b). The nicotine effect was dose dependent with an EC 50 of 19.83 μmolÁL −1 ( Figure 1c). When nicotine (100 μmolÁL −1 , apical) was applied two times repeatedly, the second nicotine-induced peak was significantly smaller (of in total 6.11 ± 3.11 μAÁcm −2 , Figure

| Nicotine acts on heteromeric αβ nAChR
In the murine tracheal epithelium, we previously detected transcripts for several nAChR subunits, comprising α3, α4, α5, α7, α9, α10, β2 and β4 (Hollenhorst, Lips, Weitz, et al., 2012). However, it remained elusive, which receptor subtype is responsible for the observed ion transport changes. Therefore, we here analysed the abundancy of the different α nAChR subunits in every single epithelium of 18 different mice via RT-PCR. The α3 and α10 subunits were present in 17 out of 18 epithelia, making them highly probable candidates for being involved in the nicotine effect. Also, the α4 subunit with 14 out of 18 epithelia and the α7 subunit with 12 out of 18 epithelia showed high distribution, while the α5 and α9 subunits were present only partially (5 of 18 and 7 of 18, respectively) making them less likely candidates (Table 2).
In indicating that the activated nAChRs were more sensitive to epibatidine than to nicotine. In the presence of 1 μmolÁL −1 dihydroβ-erythroidine, an antagonist for α4β4, α4β2, α3β2 and α3β4 nAChRs with decreasing affinity (Harvey & Luetje, 1996), the nicotine effect was similar to control conditions (Figure 2d). Application of 10 μmolÁL −1 dihydro-β-erythroidine reduced the nicotine effect by 69.01% (Figure 2d), hinting to an involvement of α3 containing nAChRs rather than α4 containing nAChRs in the nicotine effect.
The α3β2 antagonist α-conotoxin MII (Cartier et al., 1996)  In contrast to this, in β4 −/− mice, the effect on ion transport induced by 100 μmolÁL −1 nicotine and by 1 μmolÁL −1 epibatidine, both applied apically, was completely abolished (Figure 3e-h), demonstrating that the nAChR responsible for the ion transport changes contains the β4 subunit and that this subunit is essential for nicotine effect.
3.3 | The nicotine effect is associated with Ca 2+and PKA-dependent intracellular signalling To investigate if activation of tracheal epithelial nAChR leads to a release of Ca 2+ from intracellular stores, we performed experiments with thapsigargin, an inhibitor of the Ca 2+ ATPase in the endoplasmatic reticulum (Thastrup, Cullen, Drobak, Hanley, & Dawson, 1990). In the presence of 1 μmolÁL −1 thapsigargin, applied apically, the nicotine effect was significantly reduced by 81.40% ( Figure 4a), indicating that the activation of the nicotine-induced ion current changes is due to a release of Ca 2+ from intracellular stores.
Inhibition of KCNQ1 with chromanol 293B (100 μmolÁL −1 , basolateral) significantly reduced the nicotine effect by 87% (Figure 6e). This indicates that the nicotine-induced apical chloride secretion is driven by activation of the basolateral potassium channel KCNQ1.

| DISCUSSION
We have recently shown that ACh plays an important role as a nonneuronal autocrine and paracrine signalling molecule by increasing mucociliary clearance in the mouse tracheal epithelium in response to bacterial molecules . This is of considerable importance and clinically relevant, as this mechanism provides an effective response to remove inhaled pathogens. Using Ussing chamber experiments, we have previously shown that in the mouse tracheal epithelium functional nAChRs are expressed and that their activation leads to a transient apical chloride secretion which is dependent on a basolateral potassium secretion (Hollenhorst, Lips, Weitz, et al., 2012). In the present study we have confirmed the findings that application of nicotine leads to a transient current increase, observing an EC 50 of 19.83 μmolÁL −1 . This is in agreement with observations in monkey bronchial epithelial cells, where the EC 50 for nicotine was 26.5 μmolÁL −1 (Xiao, Lindstrom, & Spindel, 2009  Our observation that α-bungarotoxin, α-conotoxin ImI and ACV-1, antagonists for α7, α9 and α9α10 nAChR, did not alter the nicotine effect shows that these nAChR subtypes are not involved in the nicotine effect and that rather mixed αβ heteromeric receptors are responsible for the effect. Nevertheless, the α7 nAChR is expressed in mouse tracheal epithelium as we have shown here and previously (Hollenhorst, Lips, Weitz, et al., 2012). This receptor was found to influence transepithelial ion transport by other mechanisms as it was functionally coupled to cystic fibrosis transmembrane conductance regulator activation (Maouche et al., 2013). The complete inhibition of the nicotine effect by pretreatment with the non-selective nAChR antagonist mecamylamine observed in present study supports the conclusion that the nicotine-induced effect was indeed due to activation of heteromeric αβ nAChRs. Consistent with these observations, desensitization was observed upon a second agonist application for the αβ heteromeric nAChRs but not for the α7 nAChR (Chavez-Noriega et al., 1997).
Indeed, the α3β2, α4β2, α4β4 and α3β4 nAChR agonist epibatidine and the α4β2 and α3β4 agonist A-85380 mimicked the nicotine effect in our study. In accordance with a study from Sullivan et al. (1996), A-85380 was more potent than nicotine but less potent nAChR and 0.19 μM for the α4β4 nAChR (Harvey & Luetje, 1996).
The residual effect observed at 10 μmolÁL −1 and the fact that besides α3β4 nAChRs all other α3 or α4 containing nAChR subtypes should have already been blocked at a concentration of 1 μmolÁL −1 point towards the α3β4 nAChR being responsible for the nicotine effect.
Also, our observation that the α3 subunit was present in almost all epithelia analysed by PCR points to the α3 subunit as being responsible for the nicotine effect. Supportively, the α3β2 antagonists, α-conotoxin MII and α-conotoxin PnIA only attenuated the nicotine effect in doses much higher than their IC 50 of 0.5 nM for MII (Cartier et al., 1996) and 9.56 nM for PnIA (Luo et al., 1999), indicating that they might also act on similar nAChR, such as the α3β4 receptor. This is supported by our findings in β2 −/− and β4 −/− nAChR mice that clearly demonstrate the β4 subunit is responsible for the nicotine effect. The complete abolishment of the nicotine effect in β4 −/− nAChR mice is not due to reduced reaction caused by a deterioration of epithelial integrity as the epithelia still reacted to ATP that was used to finale experiments as a viability control (data not shown). Taken together, these results show that the α3β4 nAChR is the subtype that activates transepithelial ion transport in the mouse tracheal epithelium. In support of this, this receptor subtype has recently been shown to transiently activate ciliary beat (Perniss, Latz, et al., 2020), indicating that it is essential for the regulation of F I G U R E 4 Downstream signalling involved in the nicotine-induced activation of transepithelial ion transport in mouse trachea. (a) Application of 1-μM thapsigargin (THA, apical, n = 5), an inhibitor of the Ca 2+ -ATPase in the endoplasmatic reticulum, significantly reduced the nicotine effect (100 μM, apical, ΔI SC , *P < 0.05). (b) In the presence of the ryanodine receptor antagonist JTV519 and dantrolene (each 10 μmolÁL −1 , apical), nicotine (100 μmolÁL −1 , apical) had no effect on I SC (ns, not significant; n = 5). (c) The protein kinase C (PKC) inhibitor chelerythrine chloride (CC, 5 μmolÁL −1 , apical and basolateral, n = 7) did not influence the nicotine effect (100 μmolÁL −1 , apical, ΔI SC ; ns, not significant). (d) In the presence of the protein kinase A (PKA) inhibitor H-89 (10 μmolÁL −1 , apical and basolateral, n = 5), the current induced by nicotine (100 μmolÁL −1 , apical, ΔI SC , *P < 0.05) was significantly reduced.
Additionally, in the present experiments we have elucidated the nAChR downstream signalling cascades leading to the nicotine effect.
Previously, we have shown that the nicotine effect in the mouse tracheal epithelium is mediated only to a small extend by Ca 2+ influx from extracellular sources into the cell and Na + influx played no role (Hollenhorst, Lips, Weitz, et al., 2012). In rat colonic epithelium, metabotropic nAChRs that activate Na + /K + -ATPase currents have been reported (Bader, Lottig, & Diener, 2017 Our findings of metabotropic nAChR add to the increasing evidence for metabotropic nAChR signalling (Kabbani & Nichols, 2018).
In rat colonic epithelial cells, Lottig, Bader, Jimenez, and Diener (2019) recently found that nAChRs activate the Na + /K + ATPase via cytosolic increased Ca 2+ and PKC. However, in the tracheal epithelium, we could not find evidence for the involvement of PKC in response to nicotine. Further, evidence for metabotropic activity of nAChRs interfering with changes of [Ca 2+ ] i was found in rat alveolar macrophages (Mikulski et al., 2010). In these cells, nAChRs reduced ATP-induced Ca 2+ release and this was also independent of the presence of extracellular Ca 2+ .
We have previously found an involvement of adenylyl cyclase (AC) in the nicotine effect  is further supported by our findings that KCNQ1 is involved in the nicotine effect, as this channel is a phosphorylation target of PKA downstream of cAMP activation (Marx et al., 2002).
Effectors of nAChR activation in mouse tracheal epithelium are channels involved in chloride secretion (Hollenhorst, Lips, Weitz, et al., 2012), but the type of channels involved remains unknown.
Interestingly, in the present study we observed the activation of the Ca 2+ -activated chloride channel TMEM16A via nAChR signalling. This is of particular importance, because targeting this channel has been discussed as an alternative drug target to restore Cl − secretion in cystic fibrosis patients due to defective cystic fibrosis transmembrane conductance regulator function (Danahay et al., 2020). The observed transient increase in TMEM16A observed in our study might also be beneficial in cystic fibrosis, since one study hypothesized that a transient stimulation of chloride secretion is impaired in cystic fibrosis (Song et al., 2009). Interestingly, non-neuronal cholinergic signalling was down-regulated in cystic fibrosis patients (Wessler et al., 2007), further underlining the putative beneficial effects of targeting the nAChR signalling in airway epithelia, as investigated in our study.
Taken together, our study identifies α3β4 nAChRs as the main subunits responsible for the nicotine effect. The activation of which leads to an increase of [Ca 2+ ] i released from the endoplasmatic reticulum and PKA activation. This [Ca 2+ ] i increase in turn activates the TMEM16A chloride channel. These nAChRs might represent a novel pharmacological target to restore defective anion secretion in conditions such as cystic fibrosis, chronic obstructive pulmonary disease and cigarette smoke-induced chronic bronchitis and to improve mucociliary clearance.