Evidence for metabotropic function of epithelial nicotinic cholinergic receptors in rat colon

Abstract

Background and Purpose

ACh exerts its actions via nicotinic (nAChR) and muscarinic receptors. In the peripheral nervous system, ionotropic nAChR mediate responses in excitable cells. However, recent studies demonstrate the expression of nAChR in the colonic epithelium, which are coupled to an induction of Cl⁻ secretion via activation of the Na⁺‐K⁺‐pump.

Experimental Approach

In order to find out whether these epithelial nAChR function as ionotropic receptors, intracellular microelectrode and imaging experiments were performed in isolated crypts from rat colon. Apically permeabilized epithelia were used to measure pump current across the basolateral membrane.

Key Results

Imaging experiments with the Na⁺‐sensitive dye SBFI revealed that nicotine induced a decrease in the cytosolic Na⁺ concentration concomitant with a fall in the cytosolic Ca²⁺ concentration in about 50% of the cells. as shown in fura‐2 experiments. Nicotine hyperpolarized the membrane by 6.4 ± 2.1 mV. These observations contradict the assumption that epithelial nAChR function as ligand‐gated non‐selective cation channels. The decrease in the cytosolic Na⁺ concentration was strongly delayed, when the Na⁺‐K⁺‐pump was inhibited by scilliroside. Ussing chamber experiments revealed a strong dependence of the nicotine‐induced pump current on the presence of Ca²⁺, and chelation of cytosolic Ca²⁺ with BAPTA prevented the fall in the cytosolic Na⁺ concentration in SBFI‐loaded crypts. Inhibition of PKC with GF 109203X or Goe 6983 significantly reduced the nicotine‐induced pump current.

Conclusions and Implications

These results suggest that epithelial nAChR activate the Na⁺‐K⁺‐pump via a PKC dependent on a sufficient cytosolic Ca²⁺ concentration.

Abbreviations

I_sc

short‐circuit current

nAChR

nicotinic ACh receptor

TTX

tetrodotoxin

1.

What is already known

• Colonic epithelial cells express nicotinic receptors coupled to the activation of the Na⁺‐K⁺‐pump.

What this study adds

• Epithelial nicotinic receptors function as metabotropic receptors.

• They are coupled to protein kinase C.

What is the clinical significance

• Nicotinic receptors, which are potential drug targets for antitumor therapy, exert important physiological actions.

2. INTRODUCTION

The regulation of physiological functions in the digestive system is controlled by neurons of the autonomic nervous system (Furness, 2006). ACh is the neurotransmitter of parasympathetic and preganglionic sympathetic neurons as well as cholinergic neurons of the enteric nervous system located in the submucosal and the myenteric plexus. ACh mediates its action via nicotinic and muscarinic receptors. The textbook view of cholinergic signalling assumes that Ach is released only from neurons and that expression of nicotinic receptors is restricted to electrically excitable cells. However, several studies have shown expression of the ACh synthesizing enzyme, ChAT in a variety of non‐neuronal cells (Wessler & Kirkpatrick, 2008), especially in cells with barrier and immune functions. The same holds for the expression of nicotinic ACh receptors (nAChR), which is observed in non‐excitable cells, such as astrocytes (Shen & Yakel, 2012), macrophages (see Goverse, Stakenborg, & Matteoli, 2016), or other non‐neuronal cells (Wessler & Kirkpatrick, 2008).

In rat colonic epithelium, the different subunits (α₂, α₄, α₅, α₆, α₇, α₁₀, and β₄) of nicotinic receptors are expressed (Bader & Diener, 2015). Their activation by selective nicotinic receptor agonists evokes a strong secretion of chloride. The ion transport mechanism, which is under the control of epithelial nicotinic receptors, is the basolateral Na⁺‐K⁺‐ATPase, and in contrast to the classical view on cholinergic induced secretion, the action of ACh seems to be mediated by a coactivation of both muscarinic receptors (controlling Ca²⁺‐dependent ion channels) and nicotinic receptors (Bader, Lottig, & Diener, 2017). The basolateral Na⁺‐K⁺‐pump is essential for epithelial Cl⁻ secretion. This pump generates a Na⁺ concentration gradient used by the basolateral Na⁺‐K⁺‐2Cl⁻ cotransporter (NKCC1) responsible for the cellular uptake of Cl⁻. The pump also generates a negative membrane potential, when K⁺ ions recycle across basolateral (and apical) K⁺ channels and thus provides the driving force for Cl⁻ efflux after opening of apical Cl⁻ channels (Barrett & Keely, 2000). Consequently, the basolateral Na⁺‐K⁺‐ATPase, which is regulated by epithelial nAChR (Bader et al., 2017), represents the “motor” for many transport processes across epithelia.

While the structure and mode of action of muscarinic receptors, which use G‐protein mediated signalling cascades for translation of neuronal stimulation into cellular signals, is well investigated within the intestine, the structure and function of nicotinic receptors in the intestinal epithelium remain unknown. Nicotinic receptors in excitable cells are the prototype of ionotropic receptors permeable to Na⁺ and K⁺ (non‐selective cation channels) that can also be permeable to Ca²⁺ depending on the subunit combination (Albuquerque, Pereira, Alkondon, & Rogers, 2009). However, in other cell types, metabotropic mode(s) of action of nAChR have been shown (Grando, 2014). For example, in human monocytes, nicotinic receptor‐induced inhibition of IL‐1β release is independent of a canonical ionotropic function (Richter et al., 2016).

The Na⁺‐K⁺‐ATPase, which is activated after stimulation of epithelial nAChR in rat colon (Bader et al., 2017), is regulated by different intracellular second messenger pathways usually coupled to metabotropic receptors. PKC can stimulate this ATPase by phosphorylating its FXYD subunit (Poulsen, Morth, Egebjerg, & Nissen, 2010). Similary, the thyroid hormone triiodothyronine stimulates Na⁺‐K⁺‐ATPase activity at the tracheal epithelium in a non‐genomic manner by increasing the surface expression of this pump via Src kinases and PI3K (Bhargava, Lei, Mariash, & Ingbar, 2007). On the other hand, the Ca²⁺ or Na⁺ influx via an ionotropic receptor might activate the Na⁺‐K⁺‐ATPase by stimulation of Ca²⁺‐dependent downstream enzymes (such as Ca²⁺‐dependent isoforms of PKC) or simply by an increased substrate concentration, that is, an elevated cytosolic concentration of Na⁺. Therefore, in the present study, we asked the question whether the epithelial nicotinic receptors in the colon use a “classical” ionotropic mode of action or couple as metabotropic receptors to intracellular signalling pathways to control the activity of the Na⁺‐K⁺‐pump. For this purpose, Ussing chamber experiments, imaging techniques with Ca²⁺‐ and Na⁺‐sensitive fluorescent dyes, and microelectrode impalements at colonic crypts isolated from rats were performed.

3. METHODS

3.1. Animals

All animal care and experimental procedures were approved by the named animal welfare officer of the Justus‐Liebig‐University (administrative number 577_M) and performed according to the German and European animal welfare law. Animal studies are reported in compliance with the ARRIVE guidelines (McGrath & Lilley, 2015). Male and female Wistar rats (RRID:RGD_13508588) with an age between 6 and 8 weeks (about 200 g body mass) were used. The animals were bred and housed at the Institute of Veterinary Physiology and Biochemistry of the Justus‐Liebig‐University under specific pathogen‐free conditions. They had free access to a standard rat diet (ssniff ^® R‐Z; Sniff, Soest, Germany) and water until the time of experiment. Animals were housed in macrolone type IV cages using standard bedding material (LT E‐001; Abedd, Vienna, Austria) with three animals per cage at 22°C, 50% air humidity, and a light : dark cycle of 12:12 hr. Animals were killed after CO₂ narcosis by cervical dislocation followed by exsanguination.

3.2. Solutions

Ussing chamber experiments were carried out in a Parsons solution of the following composition (in mmol·L⁻¹): 107 NaCl, 4.5 KCl, 25 NaHCO₃, 1.8 Na₂HPO₄, 0.2 NaH₂PO₄, 1.25 CaCl₂, 1 MgSO₄, and 12.2 glucose. The solution was gassed with 5% (v/v) CO₂ and 95% (v/v) O₂ at 37°C and had a pH of 7.4 (adjusted by NaHCO₃/HCl). For the nominal Ca²⁺‐free buffer, CaCl₂ was omitted and the MgSO₄ concentration was elevated to 5 mmol·L⁻¹. For the Mg²⁺‐free buffer, MgSO₄ was omitted.

For crypt isolation, a Ca²⁺‐ and Mg²⁺‐free HBSS (GE Healthcare, München, Germany) containing 10 mmol·L⁻¹ EDTA was used. The pH was adjusted to 7.4 with Tris. The isolated crypts were stored in a high potassium Tyrode solution consisting of (in mmol·L⁻¹) 100 K gluconate, 30 KCl, 20 NaCl, 1.25 CaCl₂, 1 MgCl₂, 10 HEPES, 12.2 glucose, 5 Na pyruvate, and 1 g·L⁻¹ BSA; pH was 7.4 (adjusted by KOH).

Imaging experiments and microelectrode experiments were performed in Tyrode solution containing (in mmol·L⁻¹) 140 NaC1, 5.4 KC1, 1.25 CaC1₂, 1 MgCl₂, 12.2 glucose, 10 HEPES, adjusted to pH of 7.4 with NaOH.

3.3. Tissue preparation

The distal colon, which is a well‐established model to study epithelial anion secretion, was dissected, and its lumen was flushed several times with ice‐cold Parsons solution before it was mounted on a thin plastic rod. A circular incision was made near the distal end with a blunt scalpel. The serosa and the tunica muscularis were stripped off manually in order to obtain a mucosa–submucosa preparation. Two segments of the distal colon of each rat were used for the experiments. One segment was treated with inhibitors of different cellular signal cascades; the other segment was treated only with the solvent of the respective inhibitor in the subsequently performed Ussing chamber experiments. Segments were randomly distributed by a technician; blinding did not seem to be appropriate for these in vitro experiments.

For isolation of intact crypts, the mucosa–submucosa preparations of distal colon were mounted on a holder with tissue adhesive (cyanoacrylate) and incubated for 7 min in EDTA‐containing solution at 37°C. After the incubation time, the holder was fixed on a vibrating machine (Chemap, Volketswil, Switzerland) and vibrated once for about 30 s in order to release crypts that were collected in high K⁺ Tyrode buffer.

The isolated crypts were fixed on poly‐l‐lysine (0.1 mg·ml⁻¹; Biochrom, Berlin, Germany) coated cover slips with a diameter of 22 mm for imaging experiments. For the microelectrode studies, the isolated crypts were attached to poly‐l‐lysine coated petri dishes with a diameter of 35 mm.

3.4. Ussing chamber experiments

The mucosa–submucosa preparations were fixed in a modified Ussing chamber and bathed with a volume of 3.5 ml on each side. The tissue was incubated at 37°C and short‐circuited by a computer‐controlled voltage‐clamp device (Ingenieur Büro für Mess‐ und Datentechnik Mussler, Aachen, Germany) with correction for solution resistance. The exposed surface of the tissue was 1 cm². The electrodes used for voltage measurement and current application were Ag/AgCl electrodes in 3 mol·L⁻¹ KCl, which were separated from the chamber lumen by agar bridges (46.7 g·L⁻¹ agar in the standard bathing solution). Short‐circuit current (I_sc) was continuously recorded, and tissue conductance (G_t) was measured every minute by applying a current pulse of ±50 μA·cm⁻² with a duration of 200 ms. I_sc is expressed as μEq·hr⁻¹·cm⁻², that is, the flux of a monovalent ion per time and area with 1 μEq·hr⁻¹·cm⁻² = 26.9 μA·cm⁻². Drugs were administered after an equilibration period of about 60 min. All experiments were performed in the presence of tetrodotoxin (TTX; 10⁻⁶ mol·L⁻¹ at the serosal side) in order to prevent actions of nicotinic receptor stimulation on enteric neurons by blocking voltage‐dependent Na⁺ channels (Catterall, 1980). The apical membrane was permeabilized by mucosal administration of nystatin (100 μg·ml⁻¹) dissolved in DMSO (final concentration 0.2% v/v). Nystatin was ultrasonified immediately before use. The I_sc response to the ionophore was measured in the absence of a K⁺ gradient, that is, with identical bathing solutions at the mucosal and the serosal side. Due to the missing driving force for K⁺ flux across basolateral K⁺ channels, I_sc is carried under these conditions by the activity of the basolateral 3Na⁺‐2K⁺‐ATPase, which is activated due to the influx of mucosal Na⁺ across the nystatin pores in the apical membrane (Schultheiss & Diener, 1997).

The maximal increase in I_sc evoked by an agonist is given as the difference from the baseline value just prior administration of the drug. In those experiments, where the I_sc did not stabilize, that is, when drugs were administered during the decaying phase of the nystatin‐induced I_sc, the theoretical course of I_sc was calculated by linear regression analysis as described previously (Schultheiss & Diener, 1997). To do so, the I_sc 3 min prior administration of the drug (30 data points, as I_sc was registered every 6 s) was used to calculate the regression line. This regression served to extrapolate the decay of I_sc in the absence of nicotine, which was subtracted from the maximal increase in I_sc evoked by nicotine.

3.5. Fura‐2 measurements

Isolated crypts were fixed on poly‐l‐lysine coated cover slips and loaded for 60 min with 4 μmol·L⁻¹ fura‐2 acetoxymethylesther (fura‐2/AM) in the presence of 1.2 g·L⁻¹ pluronic F‐127 (Life Technologies, Darmstadt, Germany) at room temperature. After the loading period, the dye not taken up by the cells was washed away. The cover slip was transferred into a chamber with a volume of 2 ml, and the experiment was carried out in Tyrode solution at room temperature. The excitation wave lengths were 340 and 380 nm; the emission was measured at wave length >440 nm. The fura‐2 ratio (emission at the excitation wave length of 340 nm per emission at the excitation wave length 380 nm) was used to monitor changes in cytosolic Ca²⁺ concentration.

The imaging experiments were carried out on an inverted microscope (Olympus IX‐50; Olympus, Hamburg, Germany), equipped with an epifluorescence set‐up and an image analysis system (Till Photonics, Martinsried, Germany). Twelve regions of interest, each with the size of about one individual cell, were selected in each crypt with an equal ratio chosen for the surface region, the middle part, and the fundus region of the crypt. Data were sampled at 0.2 Hz. The baseline in the fluorescence signal was measured for several minutes before any drug was added. A response to nicotine was accepted by definition when two conditions were fulfilled simultaneously: (a) The amplitude of the change exceeded the fourfold SD of the scattering in the fura‐2 ratio during the control period just prior to addition of the drug. (b) The amplitude of the change in the fura‐2 ratio exceeded an absolute value of 0.1. At the end of each experiment, cyclopiazonic acid (10⁻⁵ mol·L⁻¹), an inhibitor of sarcoplasmic endoplasmic reticulum Ca2+ ATPase (SERCAs), was administered as viability control.

3.6. SBFI measurements

Isolated crypts (fixed on poly‐l‐lysine coated cover slips as described for fura‐2 measurements) were loaded for 100 min with 10 μmol·L⁻¹ SBFI (Na⁺‐binding benzofuran isophthalate) acetmethoxyesther (SBFI/AM; Life Technologies) in the presence of 0.3 g·L⁻¹ pluronic acid. The excitation took place at a wave length of 340 and 380 nm, and the emission was measured at wave length >440 nm. Changes in cytosolic Na⁺ concentration were presented as the ratio of the emission at the excitation wave length of 340 and 380 nm. A response to nicotine was accepted by definition when two conditions were fulfilled simultaneously: (a) The amplitude of the change exceeded the fourfold SD of the scattering in the SBFI ratio during the control period just prior to addition of the drug. (b) The amplitude of the change in the SBFI ratio exceeded an absolute value of 0.03.

3.7. Microelectrode experiments

Microelectrodes were pulled on a horizontal puller (P‐97 Flaming/Brown type micropipette puller, Sutter Instruments, Novato, CA) from filamented borosilicate glass capillary tubes with an outer diameter of 1 mm to an averaged tip resistance of 30 MΩ. The microelectrodes were backfilled with 3 mol·L⁻¹ KCl and connected via an Ag–AgCl pelleted holder to an amplifier (World Precision Instruments, Berlin, Germany). Crypts isolated from rat distal colon were mounted on petri dishes by the aid of poly‐l‐lysine. The impalement took place in the lower third of the crypt under visual control of an inverted microscope (×20 objective, Nikon Ts2R; Nikon, Düsseldorf, Germany). The micropipette was directed by a water hydraulic micromanipulator (Narishige International, London, UK). A silver wire (covered with AgCl) in the petri dish served as ground electrode. Data were sampled at 1 Hz using an A‐D converter (Digidata 1322A; Axon Instruments, Foster City, CA) and Clampex 8.2 software (Axon Instruments).

All experiments were carried out in Tyrode solution at room temperature. Impalement was accepted when the measured potential difference dropped rapidly to a value between −30 and −50 mV and remained constant for 60 s. Furthermore, rapid retraction of the pipette at the end of the recording period had to be followed by a return of the measured potential difference to zero before a data set was finally accepted.

3.8. Data and statistical analysis

The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015). Results are given as mean ± SEM with the number (n) of investigated tissues. For all Ussing chamber experiments, a group size of n = 6–7 was designed. In the imaging and in the microelectrode experiments n refers to the number of measured cells in crypts that were prepared from at least three different individual animals for each experimental series.

For the comparison of two groups, either Student's t test or Mann–Whitney U‐test was applied. An F test decided which test method had to be used. Statistical comparison of the number of responders and non‐responders for imaging experiments was calculated by using a χ ² test. Statistical comparisons were performed using the statistical software Winstat 2012.1 (R. Fitch Software). Linear regressions were calculated with Excel 2010 (RRID:SCR_016137; Microsoft).

3.9. Materials

If not indicated otherwise, all compounds used in these experiments were acquired from Sigma (Taufkirchen, Germany). TTX (CAS 4368‐28‐9) was dissolved in 2·10–2 mol·L⁻¹ citrate buffer. Forskolin (Tocris, Bristol, UK, CAS 66575‐29‐9) was dissolved in ethanol (final ethanol concentration 0.3 ml·L⁻¹). Nystatin (CAS 400‐61‐9) was dissolved in DMSO (final DMSO concentration 2 ml·L⁻¹). BAPTA (Calbiochem, Schwalbach, Germany, CAS 126150‐97‐8), calmidazolium (CAS 57265‐65‐3), cyclopiazonic acid (CAS 18172‐33‐3), GF 109203X (2‐[1‐(3‐dimethylaminopropyl)‐1H‐indol‐3‐yl]‐3‐(1H‐indol‐3‐yl)maleimide; CAS133052‐90‐1), Goe 6983 (2‐[1‐(3‐dimethylpropyl)‐5‐methoxyindol‐3‐yl]‐3‐(1H‐indol‐3‐yl)maleimide; Calbiochem, CAS 133053‐19‐7), KN‐62 (CAS 133053‐19‐7), LY 294002 (CAS 934389‐88‐5), PP2 (4‐amino‐3‐(4‐chlorophenyl)‐1‐(t‐butyl)‐1H‐pyrazolo[3,4‐d]pyrimidine; CAS 172889‐27‐9), staurosporine (CAS 62996‐74‐1), trifluoperazine (CAS 440‐17‐5), U‐73122 (Calbiochem, CAS 142878‐12‐4), and wortmannin (CAS 19545‐26‐7) were dissolved in DMSO (final maximal DMSO concentration 4 ml·L⁻¹). The cyclic depsipeptide YM‐254890 (Biomol, Hamburg, Germany, CAS 568580‐02‐9) was dissolved in distilled water containing 0.1% (w/v) BSA. Scilliroside (Sandoz, Basel, Switzerland, 507‐60‐8) was dissolved in methanol. ACh chloride (CAS 60‐31‐1) and nicotine ditartrate dihydrate (CAS 65‐31‐6) were dissolved in distilled water.

3.10. Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Harding et al., 2018), and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 (Alexander, Christopoulos et al., 2017; Alexander, Fabbro et al., 2017; Alexander, Peters et al., 2017).

4. RESULTS

4.1. Missing evidence for ionotropic function of nicotinic receptors in the colonic epithelium

In neuronal tissues as well as in the neuromuscular endplate, nicotinic receptors function as ligand‐gated ion channels (Albuquerque et al., 2009). Binding of ACh or nicotine leads to a conformational change of the receptor resulting in an inward flux of cations, predominantly Na⁺ (and Ca²⁺), and membrane depolarization. In order to find out whether the epithelial nicotinic receptors share a similar mode of action, isolated colonic crypts were loaded with SBFI, a Na⁺‐sensitive dye (see Naftalin & Pedley, 1970). Nicotine (2·10⁻⁴ mol·L⁻¹), however, did not cause an increase in the SBFI ratio signal as one might expect, if nicotine would evoke an influx of Na⁺. Instead, a prompt fall of the SBFI signal was observed after administration of nicotine (Figure 1a). Not all cells responded to nicotine. When all control groups from the series of experiments described in Table 1 were pooled, 238 out of 476 cells, that is, 50% of the cells investigated, reacted with a marked fall in the SBFI signal when exposed to nicotine. The decrease in the SBFI signal is caused by a stimulation of the Na⁺‐K⁺‐pump as shown previously in Ussing chamber experiments (Bader et al., 2017). Preincubation of the crypts with scilliroside (10⁻⁴ mol·L⁻¹), a known inhibitor of rat Na⁺‐K⁺‐ATPase (Robinson, 1970), clearly decreased the slope of the nicotine‐induced fall of the SBFI ratio signal (Figure 1c) from −3.4 ± 0.4 per 1,000 s (n = 71) to −2.2 ± 0.4 per 1,000 s (n = 96), whereas the amplitude of the nicotine response was only slightly reduced (Table 1). As a proof of principle, crypts were permeabilized at the end of each experiment with nystatin (100 μg·ml⁻¹) which led to a prompt increase in the SBFI signal caused by an influx of extracellar Na⁺ via the nystatin pores (Figure 1).

Figure 1.

Effect of nicotine (2·10⁻⁴ mol·L⁻¹) on the cytosolic Na⁺ concentration in the absence (a) or presence (b) of scilliroside (10⁻⁴ mol·L⁻¹). The SBFI ratio signal (emission at excitation wave length 340 nm per emission at excitation wave length 380 nm) was measured in isolated colonic crypts (as indicated by the schematic drawing). (c) Overlay of both curves for better comparison of the time course of the nicotine response under both conditions. At the end of the experiment, nystatin (100 μg·ml⁻¹) was administered in order to permeabilize the membrane which leads to a massive influx of extracellular Na⁺ and thereby an increase in the SBFI ratio signal. Values are means (symbols) ± SEM (grey area), n = 12 cells. For statistics, see Table 1

Table 1.

Effect of a range of inhibitors on the response of cytosolic Na+ to nicotine in colonic crypt cells.

	Δ SBFI ratio (340 nm/380 nm) in cells with a response to nicotine	Number of cells responding to nicotine
− BAPTA	−7.6·10−2 ± 0.6·10−2	36 (out of 97 cells)
+ BAPTA	−4.8·10−2 ± 0.3·10−2 *	10 (out of 80 cells)*
− GF 109203X	−12.3·10−2 ± 1.0·10−2	40 (of 53 cells)
+ GF 109203X	+12.1·10−2 ± 0.3·10−2 *	80 (out of 80 cells)*
+ Ca2+	−17.1·10−2 ± 1.4·10−2	39 (of 66 cells)
− Ca2+	−14.6·10−2 ± 1.5·10−2	46 (of 93 cells)
− Scilliroside	−12.6·10−2 ± 0.9·10−2	58 (of 78 cells)
+ Scilliroside	−11.0·10−2 ± 0.7·10−2	69 (of 104 cells)
− PP2	−8.1·10−2 ± 1.8·10−2	7 (of 44 cells)
+ PP2	−6.7·10−2 ± 1.0·10−2	23 (of 92 cells)
− Staurosporine	−6.6·10−2 ± 0.8·10−2	44 (of 80 cells)
+ Staurosporine	−7.1·10−2 ± 0.3·10−2	68 (of 108 cells)
− YM‐254890	−4.6·10−2 ± 0.6·10−2	14 (of 58 cells)
+ YM‐254890	−5.6·10−2 ± 0.5·10−2	24 (of 67 cells)

Effect of nicotine (2 x·10⁻⁴ mol·L⁻¹) on the SBFI ratio signal (as indicator for the cytosolic Na⁺ concentration) in the absence (−) or presence (+) of different inhibitors. The concentrations of the inhibitors were BAPTA‐AM (10⁻⁴ mol·L⁻¹), GF 109203X (10⁻⁵ mol·L⁻¹), scilliroside (10⁻⁴ mol·L⁻¹), PP2 (10⁻⁵ mol·L⁻¹), staurosporine (10⁻⁶ mol·L⁻¹), and YM‐254890 (10⁻⁷ mol·L⁻¹). For the definition of responders/non‐responders, see Section 3. Values are given as differences from the baseline SBFI ratio immediately before administration of nicotine (ΔSBFI) and are means ± SEM.

^* P < 0.05, significantly different from the response to nicotine in the absence of the inhibitor or the presence of Ca²⁺, respectively.

When ionotropic nicotinic receptors are activated in excitable cells, a depolarization of the membrane based on cation influx is evoked. In order to find out whether this might also be the case in the colonic epithelium, intracellular microelectrode recordings were performed in isolated colonic crypts. Nicotine did not induce a membrane depolarization. In contrast, in eight out of eight successful impalements (verified by an instantaneous return of the measured potential difference to zero when the microelectrode was retracted from the crypt at the end of the recording), nicotine (2·10⁻⁴ mol·L⁻¹) induced a transient hyperpolarization (Figure 2a). In average, membrane potential hyperpolarized from −38.2 ± 3.3 mV to −44.5 ± 2.1 mV (n = 8) in the presence of nicotine (Figure 2b).

Figure 2.

(a) Original recording of the hyperpolarization induced by nicotine (2·10⁻⁴ mol·L⁻¹) in isolated colonic crypts (as indicated by the schematic drawing). As viability control, forskolin (10⁻⁵ mol·L⁻¹) was added, which evokes a slow depolarization of the membrane. Typical tracing from a total number of eight experiments. (b) Basal membrane potential before administration of nicotine (left white bar) or forskolin (right white bar) and membrane potential in the presence of nicotine (2·10⁻⁴ mol·L⁻¹) or forskolin (10⁻⁵ mol·L⁻¹). Values are means ± SEM, n = 8. ^* P < 0.05, significantly different from baseline membrane potential just prior administration of the respective agonist

As a viability control, after the administration of nicotine, cells were exposed to forskolin (5·10⁻⁶ mol·L⁻¹), an activator of adenylate cyclase(s), which opens apical Cl⁻ channels and thereby induces a strong secretion of chloride followed by a depolarization of the membrane (Figure 2). In average, forskolin depolarized the membrane by 13.1 ± 1.8 mV (Figure 2b; n = 8). Consequently, these experiments at isolated crypts do not support the assumption that the epithelial nicotinic receptors exert an ionotropic mode of action.

4.2. The stimulation of Na⁺‐K⁺‐pump current by nicotine is inhibited by blockers of Ca²⁺‐signalling pathways

In order to find out whether metabotropic pathways known to be involved in the regulation of the Na⁺‐K⁺‐ATPase such as, for example, PLC, PKC, Ca²⁺/calmodulin kinase II, PI3K, or the tyrosine kinase Src (Grando, 2014) might be involved in the response to nicotine, Ussing chamber experiments were performed. All experiments were performed in the presence of TTX (10⁻⁶ mol·L⁻¹ at the serosal side). This was intended to prevent actions of nicotinic receptor stimulation on enteric neurons by blocking voltage‐dependent Na⁺ (Na_v) channels with this neurotoxin, although actions on TTX‐insensitive Na_v channels such as Na_v1.9 expressed by rat enteric neurons (Padilla et al., 2007) are not excluded by this strategy. The epithelia were apically permeabilized with nystatin (100 μg·ml⁻¹). The influx of mucosal Na⁺ via the nystatin pores evokes a prompt increase in I_sc (Figure 3a). In the absence of a K⁺ gradient (i.e., in the absence of a driving force for K⁺ flux across basolateral K channels), this I_sc is a Na⁺‐dependent current carried by the Na⁺‐K⁺‐ATPase (Bader et al., 2017), which slowly fades over time (Schultheiss & Diener, 1997). When nicotine (10⁻⁴ mol·L⁻¹ at the serosal side) was administered during this decaying phase, it induced a prompt, transient increase in I_sc indicating a stimulation of electrogenic ion transport across the basolateral Na⁺‐K⁺‐ATPase (Figure 3a). The I_sc response evoked by nicotine was tested in the absence and presence of putative inhibitors of different intracellular signalling pathways (see Figure 3b as an example). When Ca²⁺ was omitted from the serosal buffer solution, there was a non‐significant trend towards reduction of the nicotine‐induced pump current by about 50% (Table 2). However, omission of Ca²⁺ from both the serosal and the mucosal medium (elevating the concentration of Mg²⁺ to 5 mmol·L⁻¹ in order to maintain the integrity of the tight junctions) almost totally suppressed the stimulation of the Na⁺‐K⁺‐pump current by nicotine (Table 2). As these results indicate the involvement of Ca²⁺‐dependent processes in the regulation of the Na⁺‐K⁺‐ATPase by nicotinic receptors, inhibitors of different enzymes involved in classical Ca²⁺ signalling pathways were screened for a possible interaction with the nicotine‐induced pump current.

Figure 3.

Effect of nicotine (10⁻⁴ mol·L⁻¹ at the serosal side) in the presence of tetrodotoxin (TTX; 10⁻⁶ mol·L⁻¹ at the serosal side) on current mediated by the Na⁺‐K⁺‐pump (mucosal and serosal solution: 107 mmol·L⁻¹ NaCl/4.5 mmol·L⁻¹ KCl as indicated by the schematic inset). The effect of nicotine was tested in the presence of (a) DMSO (the solvent for GF 109203X) or in the presence of (b) GF 109203X (10⁻⁵ mol·L⁻¹ at the serosal side). The apical membrane was permeabilized by nystatin (100 μg·ml⁻¹ at the mucosal side). Mucosa–submucosa preparations from rat distal colon. Values are means (symbols) ± SEM (lines), n = 7 in each group. For statistics, see Table 2

Table 2.

Effect of a range of inhibitors on the Na⁺‐K⁺‐ATPase current response to nicotine in colonic crypt cells.▪

Drug/solution	Δ Isc (μEq·hr−1·cm−2)
	Without drug/solution	With drug/solution	n
Ca2+‐free solution serosal	5.77 ± 1.45	2.72 ± 0.63	6/8
Ca2‐free solution mucosal and serosal	5.45 ± 1.06	0.24 ± 0.26*	6/6
Calmidazolium	4.59 ± 0.45	4.26 ± 0.44	7/7
Trifluoperazine	3.36 ± 0.82	1.33 ± 0.33*	6/7
KN‐62	6.10 ± 1.07	4.90 ± 0.91	6/6
U‐73122	4.16 ± 0.69	3.08 ± 1.09	7/7
GF 109203X	6.13 ± 1.21	1.05 ± 0.41*	7/7
Goe 6983	4.25 ± 0.68	1.88 ± 0.30*	5/6
Staurosporine	5.17 ± 1.17	2.78 ± 1.06	7/7
Wortmannin	6.86 ± 1.66	4.65 ± 0.67	6/6
LY 294002	4.64 ± 1.04	2.23 ± 0.56	7/6
PP2	6.66 ± 1.25	2.26 ± 0.60*	7/6
YM‐254890	7.10 ± 1.43	2.65 ± 0.68*	7/7
Mg2+‐free solution mucosal and serosal	4.74 ± 0.67	4.47 ± 0.86	7/7

Effect of nicotine (10⁻⁴ mol·L⁻¹ at the serosal side) on the current across the basolateral membrane carried by the Na⁺‐K⁺‐ATPase in the presence (right) and in the absence (left; only solvent for the putative inhibitors present) of different inhibitors or in the absence of Ca²⁺ or Mg²⁺, respectively. The concentrations of inhibitors (all administered at the serosal side) were calmidazolium (5·10⁻⁷ mol·L⁻¹), trifluoperazine (10⁻⁵ mol·L⁻¹), KN‐62 (4·10⁻⁶ mol·L⁻¹), U‐73122 (5·10⁻⁶ mol·L⁻¹), GF 109203X (10⁻⁵ mol·L⁻¹), Goe 6983 (10⁻⁶ mol·L⁻¹), staurosporine (10⁻⁶ mol·L⁻¹), wortmannin (10⁻⁷ mol·L⁻¹), LY294002 (10⁻⁵ mol·L⁻¹), PP2 (10⁻⁵ mol·L⁻¹), and YM‐254890 (10⁻⁷ mol·L⁻¹). The apical membrane was permeabilized with nystatin (100 μg·ml⁻¹ at the mucosal side). Values are given as differences from the extrapolated baseline current just prior administration of nicotine (ΔI_sc; see Section 3) and are means ± SEM. ^* P < 0.05, significantly different from I_sc in the absence of the inhibitor. Differences in group size are due to exclusion of tissues which were damaged during the preparation.

Preincubation of the tissue with calmidazolium (5·10⁻⁷ mol·L⁻¹ at the serosal side), a specific calmodulin antagonist (see Adkins et al., 2000; Yap et al., 2000), and KN‐62 (4·10⁻⁶ mol·L⁻¹ at the serosal side), an inhibitor of Ca²⁺/calmodulin kinase II (see MacLeod & Hamilton, 1999), did not affect the stimulation of the pump current by nicotine. Only trifluoperazine (10⁻⁵ mol·L⁻¹ at the serosal side), an non‐specific calmodulin antagonist (Taylor & Broad, 1998), significantly reduced the nicotine‐induced pump current in comparison to a control only pretreated with the solvent for the antagonist (Table 2). Blockade of PLC with U‐73122 (5·10⁻⁶ mol·L⁻¹ at the serosal side) as well as inhibition of PI3K with wortmannin (10⁻⁷ mol·L⁻¹ at the serosal side) or LY 294002 (10⁻⁵ mol·L⁻¹ at the serosal side) did not inhibit the nicotine‐induced pump current, whereas PP2 (10⁻⁵ mol·L⁻¹ at the serosal side), a Src kinase inhibitor (Zachos, Lee, Kovbasnjuk, Li, & Donowitz, 2013), significantly reduced the response to nicotine by about 65% in comparison to solvent‐treated control tissues (Table 2).

A significant inhibition of the pump current induced by nicotine (10⁻⁴ mol·L⁻¹ at the serosal side) was observed, when PKCs were inhibited by GF 109203X (10⁻⁵ mol·L⁻¹ at the serosal side), a subtype‐independent blocker of PKCs (Mroz & Keely, 2012). In the presence of this inhibitor, the nicotine‐induced pump current was reduced by more than 80% (Figure 3). Inhibition was mimicked by Goe 6983 (10⁻⁶ mol·L⁻¹ at the serosal side; Table 2), another broad spectrum inhibitor of PKCs (Liu, Wang, Zhang, Shen, & Xu, 2017). There was a non‐significant trend towards inhibition of the nicotine‐induced I_sc by staurosporine (10⁻⁶ mol·L⁻¹ at the serosal side), a nonselective inhibitor of different protein kinases, which reduced the response to the nicotinic agonist by about 45% without reaching statistical significance (Table 2).

In order to find out whether G‐proteins might be involved in the presumed metabotropic mode of action of colonic epithelial nicotinic receptors, tissues were pretreated with a cyclic depsipeptide blocker of G_q proteins (Takasaki et al., 2004), YM‐254890 (10⁻⁷ mol·L⁻¹ at the serosal side), which indeed inhibited significantly the nicotine‐induced pump current. However, in a Mg²⁺‐free buffer solution, which should equilibrate with the cytosol in nystatin‐permeabilized cells leading to an intracellular Mg²⁺ depletion, the nicotine reponse remained unaffected (Table 2), making a contribution of G‐proteins, which depend on the availability of cytosolic Mg²⁺ (Birnbaumer & Zurita, 2010), unlikely.

4.3. Nicotine, however, does not evoke an increase in the cytosolic Ca²⁺ concentration

Due to the strong Ca²⁺ dependeny of the nicotine‐induced pump current observed in Ussing chamber experiments, we investigated whether nicotine might induce an increase in the cytosolic Ca²⁺ concentration. Therefore, isolated crypts were loaded with the Ca²⁺‐sensitive fluorescent dye, fura‐2. Unexpectedly, nicotine (2·10⁻⁴ mol·L⁻¹) induced a fall in the cytosolic Ca²⁺ concentration in 50% (26 out of 52) of cells indicated by a decrease in the fura‐2 ratio signal (Figure 4a). Only 33% of the cells showed either an early (reached within 3 min after administration of the agonist; observed in 10 out of 52 cells) or a late (reached within 10 min after administration of the agonist; observed in seven out of 52 cells) increase of the cytosolic Ca²⁺ concentration (Figure 4b). Seventeen percent (nine out of 52) of measured cells did not respond to nicotine with a measurable change in the fura‐2 ratio. The subsequent administration of ACh (10⁻⁴ mol·L⁻¹), the agonist of both muscarinic and nicotinic receptors, or of the SERCA inhibitor cyclopiazonic acid (10⁻⁵ mol·L⁻¹), applied as viability control, evoked a consistent increase in the fura‐2 signal in all cells tested (Figure 4).

Figure 4.

Effect of nicotine (2·10⁻⁴ mol·L⁻¹), ACh (10⁻⁴ mol·L⁻¹), and cyclopiazonic acid (10⁻⁵ mol·L⁻¹) on the cytosolic Ca²⁺ concentration measured as fura‐2 ratio signal (emission at excitation wave length 340 nm per emission at excitation wave length 380 nm) in isolated colonic crypts loaded with fura‐2 (as indicated by the schematic drawing). (a) Paradoxical fall in the fura‐2 ratio signal observed in 26 out of 52 cells. (b) Averaged response for 17 out of 52 cells, which responded with a significant increase in the fura‐2 ratio. Please note that the apparent biphasic increase in the fura‐2 signal is because most cells responded with a monophasic early rise in the fura‐2 signal and seven cells with a monophasic delayed response. Values are means (symbols) ± SEM (grey area). Nine cells did not respond with a fall or a rise of the fura‐2 signal (see Section 3 for threshold definition)

4.4. Stimulation of the Na⁺‐K ⁺‐ATPase involves a Ca²⁺‐dependent PKC

The data provided by the fura‐2 experiments, where a rapid rise of the fura‐2 signal was only detected in a minority (10 out of 52 cells) of cells, argue against the hypothesis that an activation of the Na⁺‐K⁺‐pump may be caused by an increase in the cytosolic Ca²⁺ concentration. Also, the delayed increase in the fura‐2 signal observed in some cells (seven out of 52 cells) would not fit to a Ca²⁺‐dependent activation of the pump, as there was only a short delay between the administration of nicotine and the onset of the pump current as registered by the prompt fall in the SBFI signal (Figure 1a) or the prompt increase in I_sc (Figure 3a). Consequently, the strong Ca²⁺ dependency of the stimulation of the pump current by nicotine in the Ussing chamber experiments (Table 2) suggests that a sufficient Ca²⁺ concentration in the cytosol might be necessary for the stimulation of the pump despite the fact that activation of the nicotinic receptors does not induce a consistent increase (but rather a decrease) in the cytosolic Ca²⁺ concentration (Figure 4a). To test this hypothesis, crypts were pretreated with the membrane–permeable Ca²⁺ chelator BAPTA‐AM (10⁻⁴ mol·L⁻¹ for 100 min), which is taken up by the cells as lipophilic acetmethoxyesther and converted in the cytosol into free BAPTA by intracellular esterases. Indeed, when the Ca²⁺ concentration was lowered by this manoeuvre, the fall in the SBFI signal induced by nicotine was strongly inhibited (Figure 5). Only 10 out of 80 cells loaded with BAPTA showed a measurable change in the SBFI signal, and also, the amplitude of the nicotine response in these remaining responding cells was significantly reduced in comparison to parallelly performed control crypts measured in the absence of BAPTA (Table 1).

Figure 5.

Effect of nicotine (2·10⁻⁴ mol·L⁻¹) on cytosolic Na⁺ concentration measured as SBFI ratio signal (emission at excitation wave length 340 nm per emission at excitation wave length 380 nm) in isolated colonic crypts loaded with SBFI as indicated by the schematic inset in the absence (a) and in the presence (b) of the calcium chelator BAPTA (preincubation with 10⁻⁴ mol·L⁻¹ BAPTA‐AM). At the end of the experiment, nystatin (100 μg·ml⁻¹) was administered in order to permeabilize the membrane which leads to a massive influx of extracellular Na⁺ and thereby an increase in the SBFI ratio signal. Values are means (symbols) ± SEM (grey area), n = 12 cells. For statistics, see Table 1

In a final series of experiments, in SBFI‐loaded crypts, those inhibitors were tested which had a significant effect on the nicotine‐induced pump current in the Ussing chamber experiments (Table 2). The percentage of responding cells (which showed a change in the SBFI signal larger than the defined threshold; see Section 3) was counted, and the amplitude of this change in the responding cells was quantified. In contrast to the chelation of cytosolic Ca²⁺ with BAPTA, simple removal of Ca²⁺ from the extracellular buffer did not affect the nicotine‐induced decrease in the cytosolic Na⁺ concentration. Also, PP2 (10⁻⁵ mol·L⁻¹), staurosporine (10⁻⁶ mol·L⁻¹), and YM‐254890 (10⁻⁷ mol·L⁻¹) were ineffective (Table 1). The effect of trifluoperazine could not be tested due to the strong background fluorescence of this drug. In contrast, preincubation with the PKC blocker GF 109203X (10⁻⁵ mol·L⁻¹) abolished the fall in the cytosolic Na⁺ concentration induced by nicotine (2·10⁻⁴ mol·L⁻¹) and even reversed the response into a consistent increase in the SBFI signal (Figure 6). Interestingly, this paradoxical increase was observed in all cells tested (80 out of 80 cells), that is, showed a much lower variability than the response to nicotine alone, which was never observed in all cells tested (see Table 1) suggesting that under control conditions, a (unknown) counterregulation might cover the fall in the cytosolic Na⁺ concentration which would follow the pump activation by nicotinic receptors.

Figure 6.

Effect of nicotine (2·10⁻⁴ mol·L⁻¹) in the (a) absence (only solvent present) or (b) presence of GF 109203X (10⁻⁵ mol·L⁻¹) on the cytosolic Na⁺ concentration measured as SBFI ratio signal (emission at excitation wave length 340 nm per emission at excitation wave length 380 nm) in isolated colonic crypts loaded with SBFI as indicated by the schematic inset. At the end of the experiment, nystatin (100 μg·ml⁻¹) was administered in order to permeabilize the membrane which leads to a massive influx of extracellular Na⁺ and thereby to an increase in the SBFI ratio signal. Values are means (symbols) ± SEM (grey area), n = 12 cells. For statistics, see Table 1

5. DISCUSSION

ACh regulates physiological functions via muscarinic and nicotinic receptors in organs innervated by the parasympathetic and the enteric nervous system. While muscarinic receptors are known to transmit cholinergic signals in tissues innervated by the autonomic nervous system, nicotinic receptors have been thought to be expressed only in electrically excitable tissues. This textbook view was challenged by the discovery of nicotinic receptor subunits in different epithelial cells (see Wessler & Kirkpatrick, 2008) including the colonic epithelium (Bader & Diener, 2015). Cl⁻ secretion across the intestinal epithelium is stimulated through cholinergic signals. They are transmitted via G‐protein coupled muscarinic receptors leading to a stimulation of Ca²⁺‐dependent basolateral K⁺ and apical Cl⁻ channels (Barrett & Keely, 2000) and via nicotinic receptors, which stimulate the basolateral Na⁺‐K⁺‐pump (Bader et al., 2017). The results of the present study suggest that this effect does not involve a “classical” ionotropic mode of action of epithelial nAChR but point to a metabotropic control of the activitiy of the Na⁺‐K⁺‐pump by these receptors.

The Na⁺‐K⁺‐ATPase is composed of a combination of different isoforms of the α‐subunit, containing the catalytic transport activity, and β‐subunits functioning as chaperons. In most epithelia, including rat colonic epithelium, that is, the tissue where the present experiments were performed, the dominant form of the Na⁺‐K⁺‐ATPase is a α₁β₁‐heterodimer (Escoubet, Coureau, Bonvalet, & Farman, 1997). Optionally, a further regulatory protein, FXYD, with seven isoforms known up to now can be part of this ATPase but does not seem to be essential for pump activity (Geering, 2005). Physiologically, this ion pump generates the driving force for many epithelial transport processes by establishing a Na⁺ concentration gradient, which can be used by secondarily active Na⁺‐dependent transporters (such as the basolateral Na⁺‐K⁺‐2Cl⁻ cotransporter) and generates a K⁺ gradient, which builds up the membrane potential when K⁺ leaves the cell across K⁺ channels in the membrane. Beside its central contribution to epithelial transport, other functions of this ATPase are known. For example, trans‐interactions of the α₁‐subunits of the Na⁺‐K⁺‐pump of neighbouring epithelial cells in the area of the desmosomes participate in the establishment of epithelial cell contacts (Vagin, Dada, Tokhtaeva, & Sachs, 2012), and also, signalling functions of this ATPase have been described (see, e.g., Jansson et al., 2015).

The present study confirms that nicotine via stimulation of nAChR induces a pump current across the basolateral membrane in apically permeabilized epithelia (Figure 3a) as shown previously (Bader et al., 2017), concomitant with a decrease in the cytosolic Na⁺ concentration in isolated colonic crypts loaded with the Na⁺‐sensitive dye SBFI (Figure 1). Both the stimulation of the pump current (Bader et al., 2017) and the fall in the cytosolic Na⁺ concentration (Figure 1) were inhibited (or slowed significantly) by scilliroside, an effective inhibitor of the α₁ subunit‐containing Na⁺‐K⁺‐pump in rats, which is quite insensitive in this species to the classical inhibitor, ouabain, in comparison to pump forms containing α₂ or α₃ subunits (Blanco & Mercer, 1998).

Different pathways have been described as being involved in the regulation of Na⁺‐K⁺‐pump in epithelia, which might therefore be candidates mediating the stimulation of Na⁺‐K⁺‐ATPase by nAChR in the colon. For example, the stimulation of sugar absorption in porcine jejunum by insulin‐like growth factor 1 (IGF‐1) is caused by a PI3K‐mediated stimulation of the Na⁺‐K⁺‐pump (Alexander & Carey, 2001). A similar action is induced by the thyroid hormone, triiodothyronine, which (in a non‐genomic manner) stimulates the Na⁺‐K⁺‐pump surface expression in tracheal epithelium via Src and PI3K (Bhargava et al., 2007). Depending on the subunit composition, phosphorylation by cAMP‐dependent PKA can either activate or inhibit pump activity (Blanco & Mercer, 1998), and also, PKC is able to phosphorylate the Na⁺‐K⁺‐ATPase (Poulsen et al., 2010).

In general, nicotinic receptors are considered as prototypic ionotropic receptors. The binding of ACh between neighbouring subunits (α₁ to α₇ and α₉ on one side and α₁₀, β₃, β₄, γ, δ, or ε at the opposing side) leads to a conformational change of the pentameric channel, allowing the influx of Na⁺ and Ca²⁺ into the cell (Albuquerque et al., 2009). However, despite the unequivocal ionotropic mode of action of nAChR in most cell types, evidence for metabotropic signalling pathways has been gathered from several tissues (Grando, 2014). One of the first examples reported is the hyperpolarization evoked by stimulation of nicotinic receptors at rat septal neurons, which is mediated by a G‐protein (Sorenson & Gallagher, 1996). Also, the neuroprotective effect of nicotine involves a PI3K and the kinase Src (Kihara et al., 2001). Another example is the nAChR‐induced inhibition of IL‐1β release in human monocytes via a metabotropic mode of action (Richter et al., 2016). Therefore, in the colonic epithelium, where nAChR exerts its effects via activation of the Na⁺‐K⁺‐pump, ionotropic as well as metabotropic mechanisms have to be considered.

Based on the known regulating factors, which are able to stimulate the pump, and the signalling pathways activated by ionotropic or metabotropic nicotinic receptors, several mechanisms could link epithelial nAChR activation to stimulation of Na⁺‐K⁺‐pump activity. The simplest model, connecting the known modes of regulation of Na⁺‐K⁺‐ATPase activity and the modes of function of nAChR, might be that a Na⁺ influx via the ligand‐gated, non‐selective cation channel formed by nAChR stimulates the pump due to an increased substrate concentration, that is, due to an elevated cytosolic Na⁺ concentration. Also consistent with an ionotropic mode of function would be an influx of Ca²⁺ via the ligand‐gated channel stimulating Ca²⁺‐dependent enzymes involved in the regulation of the ATPase. However, this was obviously not the case. Influx of cations via a ligand‐gated cation channel should lead to a membrane depolarization. However, exposure to nicotine hyperpolarized the membrane of epithelial cells in isolated colonic crypts (Figure 2), which is fully consistent with the electrogenic mode of action of the Na⁺‐K⁺‐ATPase extruding 3Na⁺ from the cell in exchange for only 2K⁺. Also, the cytosolic Na⁺ concentration showed a decrease instead of an increase as one should expect, if nicotine opened a non‐selective cation channel (Figure 1a). The change in the cytosolic Ca²⁺ concentration was more complex, as different cellular patterns could be observed. The largest proportion of the cells responded rapidly with a decrease in the cytosolic Ca²⁺ concentration (Figure 4a) and, only in a minority of cells, was an increase observed within minutes (Figure 4b). Both patterns, however, are inconsistent with the rapid influx of extracellular Ca²⁺ via a ligand‐gated cation channel. Consequently, these observations make an ionotropic function of the colonic epithelial nAChR unlikely, although it will be interesting, in further experiments, to study a possible interaction of nicotinic receptors with intracellular Ca²⁺ stores or Ca²⁺ efflux transporters, which might explain the paradoxical fall in the cytosolic Ca²⁺ concentration observed in the majority of cells tested (Figure 4a).

In contrast, in the Ussing chamber experiments, the nicotine‐induced pump current was inhibited by blockers of several signalling pathways usually coupled to metabotropic receptors in the membrane. PP2, a Src kinase inhibitor (Zachos et al., 2013), reduced the nicotine‐induced pump current in the Ussing chamber experiments (Table 2). However, this inhibition could not be confirmed when measuring the fall in the cytosolic Na⁺ concentration in isolated colonic crypts loaded with SBFI (Table 1), suggesting that the inhibition of the nicotine‐induced I_sc might represent a non‐specific action. Blockade of PI3K with wortmannin or LY 294002 was ineffective, making a contribution of this kinase unlikely. Removal of extracellular Ca²⁺ strongly inhibited the nicotine‐induced pump current in the nystatin‐permeabilized epithelia (Table 2). At first glance, this seems to contradict the observation that removal of extracellular Ca²⁺ did not inhibit the fall in the SBFI signal in the experiments with isolated crypts (Table 1). However, in the imaging experiments the effect of nicotine is tested in non‐permeabilized epithelia, that is, in the absence of nystatin. Nystatin, as used in the Ussing chamber experiments, equilibrates intracellular and extracellular Ca²⁺ concentration due to the pores formed by this ionophore in the membrane. In other words, simple removal of extracellular Ca²⁺ as performed in the experiments at isolated crypts (Table 1) will have a much smaller effect on the cytosolic Ca²⁺ concentration compared to Ussing chamber experiments where Ca²⁺‐free buffer was combined with nystatin (Table 2). The involvement of cytosolic Ca²⁺ in the nicotine‐induced fall of the cytosolic Na⁺ concentration could, nevertheless, be demonstrated, when the crypts were loaded with the Ca²⁺ chelating agent, BAPTA (Figure 5; Table 1). Preincubation with BAPTA/AM strongly inhibited the fall in the SBFI ratio signal evoked by nicotine. These results indicate that nicotine does not induce a relevant increase in the cytosolic Ca²⁺ concentration (Figure 4) via Ca²⁺ influx through channel opening. However, a sufficient cytosolic Ca²⁺ concentration seems to be necessary as prerequisite for the agonist to exert its action. Calmodulin, a Ca²⁺‐binding protein which mediates many intracellular effects of the second messenger Ca²⁺, is obviously not involved in this action as shown by the inability of a typical calmodulin antagonist to inhibit the nicotine‐induced pump current (Table 2), although calmodulin has been shown to interact (probably in a non‐specific manner) via hydrophobic surfaces with subunits of the Na⁺‐K⁺‐ATPase in murine brain (Berggård et al., 2006). This conclusion is supported by the finding that KN‐62 did also not interfere with the nicotine‐induced I_sc. Beside its blocking action on, for example, human (but not rat) purinergic P2X7 receptors (Donnelly‐Roberts, Namovic, Han, & Jarvis, 2009), this drug is a potent blocker of Ca²⁺/calmodulin kinase II (see MacLeod & Hamilton, 1999), which often mediates intracellular effects of calmodulin.

A possible candidate for a Ca²⁺‐dependent enzyme mediating the activation of the Na⁺‐K⁺‐ATPase by nicotine is PKC. This protein family contains Ca²⁺‐dependent and Ca²⁺‐independent forms. GF 109203X, a subtype‐independent blocker of PKC, strongly inhibited the nicotine‐induced pump current measured in apically permeabilized epithelia in the Ussing chamber (Figure 3; Table 2). In the SBFI experiments, in the presence of this inhibitor, the usually observed fall in the cytosolic Na⁺ concentration was even reversed into an increase (Figure 6; Table 1). Furthermore, the observed variability in the change of the SBFI signal (e.g., only 40 out of 53 cells responded to nicotine in the control series performed in parallel to the GF 109203X experiments) could not be detected anymore, that is, all cells (80 out of 80 cells) showed a consistent nicotine‐induced increase in the SBFI ratio after preincubation with GF 109203X (Table 1). These experiments not only strongly suggest the involvement of a PKC in the regulation of epithelial Na⁺‐K⁺‐ATPase by nAChR but also further suggest that the usual variability in the response to nicotine at the level of the individual crypt cells does not represent a heterogenous expression of nAChR but that a counter‐regulatory (or parallel‐activated) Na⁺‐uptake mechanism covers the stimulation of the pump in a fraction of the cells. The nature of this counter‐regulatory mechanism(s), for instance the Na⁺‐H⁺ exchanger, Na⁺‐Ca²⁺‐exchanger or Na⁺‐K⁺‐2Cl⁻ cotransporter, is still to be established. The variability of the nicotine response on the cellular level (i.e., in the imaging experiments) in comparison to the constant ensemble average response of larger mucosal surfaces (in the Ussing chamber experiments) is probably related to the multiple cell types found in the intestinal epithelium. In colonic crypts, there are enterocytes, goblet cells, enteroendocrine cells, tuft cells, stem cells, and many other cell types (Crosnier, Stamataki, & Lewis, 2006), which might be differentially affected by nicotine or might express different receptors for nicotine. However, double labelling of cells with markers for the different cell types and nicotinic receptor subunits is handicapped by the lack of specific antibodies for nAChR (see Moser et al., 2007).

Evidence for stimulation of the Na⁺‐K⁺‐ATPase by nAChR in other tissues is rare. The only example we have found is rat skeletal muscle (Krivoi et al., 2006). However, in this tissue, the activation of the pump is mediated by a direct protein–protein interaction of the desensitized form of the nAChR with the α₂ subunit of the Na⁺‐K⁺‐ATPase (Heiny et al., 2010). Consequently, the results described here point to a new pathway previously unknown in the coupling between nAChR and Na⁺‐K⁺‐ATPase. Further experiments are needed to clarify this interaction, for example, to solve the question of which isoform(s) of PKC is involved in this process.

Pathophysiologically, nicotine and its effects mediated via nAChR may contribute to the initiation and/or promotion of different forms of cancer including colonic cancer (Schuller, 2009). For example, nicotinic actions via nAChR are known to influence cell proliferation, growth arrest, and apoptosis (Grando, 2014) including proliferation of intestinal stem cell (Takahashi, Shiraishi, & Murata, 2018). Consequently, epithelial nAChR may represent a new potential target for anti‐tumour therapies. However, before this can be achieved, the physiological relevance of the expression of epithelial nAChRs and their role for the Na⁺‐K⁺‐ATPase with its pivotal function within the intestinal epithelium must be further clarified.

AUTHOR CONTRIBUTIONS

L.L. and S.B. performed the experiments. M.D. drafted the manuscript. All authors planned the experiments and approved the final version.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

DECLARATION OF TRANSPARENCY AND SCIENTIFIC RIGOUR

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research as stated in the BJP guidelines for Design & Analysis, and Animal Experimentation, and as recommended by funding agencies, publishers and other organisations engaged with supporting research.

Lottig L, Bader S, Jimenez M, Diener M. Evidence for metabotropic function of epithelial nicotinic cholinergic receptors in rat colon. Br J Pharmacol. 2019;176:1328–1340. 10.1111/bph.14638