Flupirtine enhances NHE-3-mediated Na⁺ absorption in rat colon via an ENS-dependent mechanism

Abstract

Recent studies in our lab have shown that the K_V7 channel activator, flupirtine, inhibits colonic epithelial Cl⁻ secretion through effects on submucosal neurons of the enteric nervous system (ENS). We hypothesized that flupirtine would also stimulate Na⁺ absorption as a result of reduced secretory ENS input to the epithelium. To test this hypothesis, unidirectional ²²Na⁺ fluxes were measured under voltage-clamped conditions. Pharmacological approaches using an Ussing-style recording chamber combined with immunofluorescence microscopy techniques were used to determine the effect of flupirtine on active Na⁺ transport in the rat colon. Flupirtine stimulated electroneutral Na⁺ absorption in partially seromuscular-stripped colonic tissues, while simultaneously inhibiting short-circuit current (I_SC; i.e., Cl⁻ secretion). Both of these effects were attenuated by pretreatment with the ENS inhibitor, tetrodotoxin. The Na⁺/H⁺ exchanger isoform 3 (NHE-3)-selective inhibitor, S3226, significantly inhibited flupirtine-stimulated Na⁺ absorption, whereas the NHE-2-selective inhibitor HOE-694 did not. NHE-3 localization near the apical membranes of surface epithelial cells was also more apparent in flupirtine-treated colon versus control. Flupirtine did not alter epithelial Na⁺ channel (ENaC)-mediated Na⁺ absorption in distal colonic tissues obtained from hyperaldosteronaemic rats and had no effect in the normal ileum but did stimulate Na⁺ absorption in the proximal colon. Finally, the parallel effects of flupirtine on I_SC (Cl⁻ secretion) and Na⁺ absorption were significantly correlated with each other. Together, these data indicate that flupirtine stimulates NHE-3-dependent Na⁺ absorption, likely as a result of reduced stimulatory input to the colonic epithelium by submucosal ENS neurons.

NEW & NOTEWORTHY We present a novel mechanism regarding regulation of epithelial ion transport by enteric neurons. Activation of neuronal K_V7 K⁺ channels markedly stimulates Na⁺ absorption and inhibits Cl⁻ secretion across the colonic epithelium. This may be useful in developing new treatments for diarrheal disorders, such as irritable bowel syndrome with diarrhea (IBS-D).

INTRODUCTION

The epithelium of the colon plays a critical role in maintaining overall electrolyte and fluid homeostasis (1–4). Although the total volume of absorbed water is typically less than in the small intestine, the colon possesses a hugely dynamic range in absorptive capacity and readily adapts its electrolyte transport activity to fit the needs of the organism (5). Having tremendous influence over this activity is the enteric nervous system (ENS), one of the most important and powerful regulators of epithelial ion transport (6–9). Na⁺, Cl⁻, and water are actively absorbed under basal conditions, but the release of prosecretory neurotransmitters from ENS neurons can induce a robust efflux of Cl⁻ and HCO₃⁻ from the crypt cells (10, 11), driving water into the gut lumen through osmosis. At the same time, ENS-driven secretion is augmented by the concurrent inhibition of basal absorptive processes, such as Na⁺/H⁺ and Cl⁻/HCO₃⁻ exchange in surface epithelial cells (12, 13).

Recent work in our laboratory has revealed that pharmacological activation of voltage-gated K_V7 (KCNQ) K⁺ channels in the ENS inhibits neurogenic Cl⁻ secretion significantly within the colon (14). This is likely a result of suppressed activity of secretomotor neurons, as K_V7 channels mediate a hyperpolarizing K⁺ current that reduces excitability and actional potential firing (15–17). The importance of these channels has been highlighted within many central and peripheral circuits (18–20), and K_V7 channel activators have been used clinically to treat chronic pain and epilepsy—both of which are related to neuronal hyperexcitability (16, 21, 22). However, the role of K_V7 channels in the ENS with respect to epithelial transport remains incompletely understood.

Many common gastrointestinal (GI) disorders, such as irritable bowel syndrome with diarrhea (IBS-D), are related a hyperexcitable ENS (6, 11, 23, 24). IBS-D, although not usually life-threating, presents a massive economic burden to the health care systems in the United States and abroad (25). Many current treatment options that target ENS activity are effective in relieving symptoms but are commonly associated with adverse side effects (26–28). Thus, development of improved therapeutic strategies for IBS-D management is of great clinical importance. Gaining a deeper understanding of epithelial transport modulation by K_V7 channels in the ENS may provide new avenues for treatment, as the use of K_V7 activators has not yet been explored in this context.

In light of the recent evidence concerning the effects on neurogenic Cl⁻ secretion (14), we predicted that activation of K_V7 channels in the ENS would also enhance Na⁺ absorption simultaneously, as a result of diminished stimulatory ENS input. Na⁺ absorption—in particular, the activity of the predominant electroneutral Na⁺/H⁺ exchanger isoform 3 (NHE-3)—is known to be active at rest and inhibited by various ENS-derived neurotransmitters (29). We therefore tested the hypothesis that the K_V7 channel activator, flupirtine, stimulates active Na⁺ absorption in the colon. Unidirectional ²²Na⁺ fluxes and short-circuit current (I_SC) were measured under voltage-clamped conditions in an Ussing chamber system. Our findings suggest that active Cl⁻ secretion is inhibited, whereas NHE-3-mediated Na⁺ absorption is enhanced by activation of K_V7 channels within the ENS.

METHODS

Animals

Nonfasting male Sprague-Dawley rats (150–200 g) were housed 2 per cage in climate-controlled rooms under 12-h light/dark cycles and given free access to standard chow and water. One group of rats was given a substitute Na⁺-depleted diet for 7 consecutive days to induce secondary hyperaldosteronism, as has been described previously (30, 31). Rats given Na⁺-depleted diet were also allowed free access to water. Dietary Na⁺ depletion was used to study the effect of ENS K_V7 activation specifically on electrogenic versus electroneutral epithelial Na⁺ absorption, as electroneutral (primarily NHE-3-mediated) absorption is downregulated, whereas electrogenic (epithelial Na⁺ channel; ENaC-mediated) absorption is induced in hyperaldosteronaemic rats (32). On the day of experiments, all rats were anesthetized with 5% isoflurane balanced with oxygen and maintained under a deep plane of anesthesia with 2%–3% isoflurane delivered via nose cone. The colon was removed by cutting away mesenteric attachments and excising the tissue from caecum to rectum. Colons were immediately flushed with ice-cold modified Krebs solution containing the following (in mM): 140 Na⁺, 5.2 K⁺, 1.2 Ca²⁺, 1.2 Mg²⁺, 119.8 Cl⁻, 25 HCO₃⁻, 2.4 HPO₄²⁻, 0.4 H₂PO₄⁻, and 10 glucose. Krebs solution was bubbled with 5% CO₂ balanced with oxygen and the pH was adjusted to 7.41. After removal of the colon, rats were euthanized by cutting the diaphragm and the colon was used for downstream Ussing chamber, immunofluorescence, or Western blotting experiments. For some experiments, the ileum was also removed and flushed in the same manner. All experimental procedures were approved by West Virginia University Animal Care and Use Committee.

Previous work in our laboratory regarding the effects of K_V7 activation in the ENS has been conducted using mice. In the present study, rats were chosen as the primary model system for several reasons: 1) to establish that the effect(s) of K_V7 activation on colonic epithelial transport are not unique to the mouse, 2) to allow for more substantiated interpretation of the findings regarding the modulation of Na⁺ transport, as much of the previous work in field of colonic Na⁺ transport has been conducted in rats (23, 33–36), and 3) to minimize the number of required animals, as the larger surface area of the rat colon permits the use of multiple tissue segments per animal for unidirectional flux experiments. In addition, only male rats were used in this study because many aspects of colonic ion transport are markedly different in females versus males, like owing to modulation by sex steroids (37–39). Future studies will be aimed at exploring any sex differences in the response to flupirtine or other ENS-modulating drugs.

Ussing Chamber Experiments

After flushing, the colon was opened longitudinally along the mesenteric border and placed mucosal side down in a dissecting dish, submerged in ice-cold bubbled Krebs solution. The serosa, longitudinal, and circular smooth muscle layers were gently stripped away first by making a lateral incision at the junction between proximal and distal segments (i.e., at the distal end of the proximal striations). Muscle and serosal layers were then peeled back using a glass slide and fine forceps to obtain a mucosa/muscularis mucosae/submucosa preparation. Tissues prepared in this way contain functional submucosal neurons but lack the myenteric plexus altogether. Colonic segments were then mounted on 1.1-cm² sliders with a circular aperture for use in an Ussing-style recording chamber (Physiologic Instruments, San Diego, CA). Mounted tissues were bathed bilaterally in 5 mL of cold Krebs buffer, bubbled with 5% CO₂ balanced with oxygen and gradually warmed to 37°C. For experiments using proximal colon, the tissue was prepared the same way as described here using segments of proximal colon characterized by striation patterns. For experiments using ileum, tissues were partially stripped and mounted on 0.3 cm² oval aperture sliders. This difference in slider aperture was corrected for in the electrical recordings and Na⁺ flux calculations described below.

Chambers were equipped with a pair voltage-sensing AgCl pellet electrodes and current-injecting AgCl wire electrodes connected to the chamber bath via agar-salt bridge (3.5% agar in 3 M KCl) embedded in the electrode tips. After mounting, tissues were allowed ∼15 min to equilibrate and be brought up to temperature under open-circuit conditions, during which time transepithelial potential (V_TE) was monitored. Tissues were then voltage-clamped to 0 mV using a multichannel voltage clamp/amplifier (VCC MC8; Physiologic Instruments) and short-circuit current (I_SC) was continuously monitored by automated computer software (Acquire & Analyze; Physiologic Instruments). The 5 mV, 200 ms bipolar pulses were applied at 30-s intervals to monitor transepithelial conductance (G_TE). By convention, positive I_SC values reflect net anion (Cl⁻/HCO₃⁻) secretion and/or cation (Na⁺) absorption; negative I_SC values reflect net anion absorption or cation (K⁺) secretion.

Unidirectional ²²Na⁺ fluxes were performed by first assigning tissue pairs with G_TE values within 15% of each other. The 2 µCi of ²²Na⁺ was then added to either the serosal (basolateral) or mucosal (apical) chamber bath and allowed 25 min to reach a steady-state rate of flux. A total of 250 µL samples were taken in duplicate from the opposite (“cold”) chamber bath at the beginning and end of each 30-min flux period and replaced with 500 µL of fresh buffer. Drugs were added as 1,000× stock solutions, administered in 5 µL volumes to the 5-mL chamber bath immediately following sample collection. Drug concentration and placement details are given in the figure legends. At the end of each experiment, 50 µL samples were taken from the “hot” chamber bath (where ²²Na⁺ tracer was added) to obtain a standard value used in the calculation of Na⁺ flux rate. Samples were then mixed with liquid scintillation cocktail (Sigma Aldrich, Cat. No. 03999-5 L) and counted using a Tri-carb 4910TR liquid scintillation counter (Perkin Elmer, Waltham, MA). Unidirectional ²²Na⁺ flux rates were calculated from the measured counts per minute (cpm) values using the following equation:

$${\text{Na}}^{\text{+}}\text{flux }\left(\text{J}\right){\text{ (mEq/cm}}^{\text{2}}\text{×h)}=\left(5\times B-\left[A\times 0.9\right]\right)/\left(\left[S/140\right]\times 1.1\times 0.5\right)$$

“A” and “B” represent cpm/mL values measured from the beginning and end of a flux period, respectively. 5 is the total volume (in mL) of the chamber bath and 0.9 is dilution factor accounting for removal of 500 µL of bath volume upon sample collection. “S” represents the cpm/µEq (in 1 mL) measured from the standard obtained from the “hot” bath of the chamber. Value 1.1 corresponds to the slider aperture (in cm²) and 0.5 is the time (in hours) for each flux period (40). Net Na⁺ fluxes were calculated by subtracting the serosal-to-mucosal (S-M) from the mucosal-to-serosal (M-S) fluxes from paired tissues for a given flux period.

Immunofluorescence Microscopy

To identify the presence of flupirtine-sensitive K_V7 channels in submucosal neurons, segments of mid-distal colon were dissected under a microscope to obtain a submucosal preparation, as described previously (14). First, the colon was opened longitudinally and pinned mucosal side down on a dissecting stage. A shallow lateral incision was made near the end of the striations of the proximal colon to transect the longitudinal and circular smooth muscle layers, which were then peeled away in a distal direction until completely removed. The tissue was then inverted and pinned mucosal side up. Another lateral incision was made near the rectum, just deep enough to isolate the mucosa/muscularis mucosae from the underlying submucosal layer. The mucosal/muscularis mucosae was removed by carefully creating a plane of separation using ultrafine forceps and then gently rubbing in a proximal direction with the index finger. The remaining submucosal layer was immediately fixed on 4% paraformaldehyde (PFA) dissolved in phosphate-buffered saline (PBS) overnight at 4°C.

The following day, the tissue was permeabilized with PBS + 1% Triton-X + 5% goat serum for 1 h at room temperature and blocked for an additional 30 min in PBS + 0.01% Triton-X (PBST) + 5% goat serum. The tissue was then incubated in PBST + 5% goat serum containing the primary antibodies: mouse antimicrotubule-associated protein 2 (MAP-2; neuronal marker; Millipore, Cat. No. MAB3418; 1:200 dilution) (41) and rabbit anti-KCNQ2 (flupirtine-sensitive K_V7 channel subtype; Thermo, Cat. No. PA1-929; 1:100 dilution) (20) overnight at 4°C. The next day, the tissue was washed five times for 5 min each in PBST before incubation in PBST + 5% goat serum containing AlexaFluor 488 goat anti-rabbit (Thermo, Cat. No. A11008; 1:2,000 dilution) and AlexaFluor 568 goat-anti-mouse (Thermo, Cat. No. A11031; 1:2,000 dilution) for 30 min at room temperature, after which it was again washed five times for 5 min each in PBST. The tissue was then rinsed briefly in water and placed flat on a charged glass microscope slide. SlowFade Diamond mountant with DAPI (Thermo, Cat. No. 36971) was used to mount a coverslip on the slide, which was then sealed with clear fingernail polish. Images were captured using a Zeiss 710 confocal microscope under ×20 magnification (Carl Zeiss, Oberkochen, Germany).

For immunofluorescence experiments to visualize NHE-3 localization in the epithelium, intact (unopened) colons were stripped of longitudinal and circular smooth muscle and incubated in bubbled Krebs solution with or without flupirtine for 10 min at 37°C. The tissues were then immediately fixed in prewarmed 4% PFA in PBS for 15 min. After washing three times in cold PBS, tissues were flash frozen in isopentane cooled with liquid nitrogen and embedded in tissue freezing medium (Scigen, Cat. No. 4586, Gardena, CA). Ten-micrometer-sections were cut with a cryostat and mounted on charged glass slides. Sections were permeabilized in PBS + 0.1% Triton-X + 5% goat serum for 20 min at room temperature, washed briefly in PBST and blocked in PBST + 5% goat serum for an additional 30 min. Next, sections were incubated in PBST+ 5% goat serum containing NHE-3-specific primary antibody (Abcam, Cat. No. 95299; 1:100 dilution) (39) for 1 h at 37°C in a humidifying chamber. The NHE-3 antibody used in this study has been previously validated for immunofluorescence microscopy use in rat intestinal tissues (42), and specificity has been confirmed in experiments using Slc9a3^−/− (NHE-3) mice (43). The sections were subsequently washed five times for 5 min each and incubated in PBST + 5% goat serum containing AlexaFluor 488 goat anti-rabbit (Thermo, Cat. No. A11008; 1:2,000 dilution) and TRITC-conjugated phalloidin (filamentous actin labeling; Thermo, Cat. No. R415; 1:1,000 dilution) for 30 min at room temperature. Finally, sections were washed again five times for 5 min each and mounted using SlowFade Diamond mountant with DAPI (Thermo, Cat. No. 36971). Slides were sealed with clear fingernail polish and imaged using a Zeiss 710 confocal microscope under ×20 magnification (Carl Zeiss, Oberkochen, Germany).

Western Blotting

Colons from normal or dietary Na⁺-depleted rats were flushed with ice-cold Krebs solution to remove fecal contents. A clamp was then placed on the distal end of the colon, near the rectum, and the colons were filled with ice-cold epithelial isolation solution containing the following (in mM): 30 NaCl, 8 HEPES-Tris (pH 7.5), 5 ethylenediaminetetraacetic acid (EDTA), 1 phenylmethylsulfonyl fluoride (PMSF), and 0.5 dithiothreitol (DTT). After filling, the colons were secured on the proximal end with a small ligature and incubated in this same solution on ice for 30 min. Then the colons were emptied, opened longitudinally along the mesenteric border and scraped with a glass slide to remove the mucosa, which was flash frozen in liquid nitrogen and stored at −80°C until use.

Proteins were extracted from the tissues by homogenizing in ice-cold radio-immunoprecipitation assay (RIPA) lysis buffer and briefly sonicating to disrupt membranes. Samples were centrifuged for 20 min at 12,000 g, and the supernatants were collected and mixed with 4× Laemmli buffer. The samples were then heated to 90°C for 5 min and immediately chilled on ice before adding 2-mercaptoethanol (5% w/v). Proteins were resolved on 8% polyacrylamide gels and transferred to PVDF membranes, which were subsequently blocked in 3% bovine serum albumin (BSA) in tris-buffered saline plus 0.1% Tween-20 (TBST) for 1 h at room temperature. Membranes were then incubated overnight in TBST plus 3% BSA solution containing rabbit anti-NHE-3 (Abcam, Cat. No. 95299; 1:500 dilution) or horseradish peroxidase (HRP)-conjugated mouse anti-β-actin (SCBT, Cat. No. SC-47778; 1:1,000 dilution) primary antibody. The following day, membranes were washed in TBST five times for 5 min each. Membranes that were probed for NHE-3 were then incubated in HRP-conjugated goat anti-rabbit secondary antibody (Thermo, Cat. No. 31462) for 1 h at room temperature and subsequently washed five times again for 5 min each. Immune complexes were detected via chemiluminescence using West Dura extended duration substrate (Thermo, Cat. No. 34075) and imaged on a G-box. Protein band intensity was quantified using FIJI (Image J) software.

Drugs and Chemicals

Flupirtine, vasoactive intestinal polypeptide (VIP), and tetrodotoxin were purchased from Tocris (Minneapolis, MN). Veratridine was purchased from Alomone laboratories (Jerusalem, IL). Amiloride and S3226 were purchased from Sigma Aldrich (St. Louis, MO). HOE694 was a generous gift from Dr. Selvi Krishnan. All other chemicals and salts were purchased either from Sigma Aldrich or Fisher Scientific (Waltham, MA).

Data Analysis

I_SC and G_TE values were averaged from each tissue over the final 5 min for a given flux period when used for comparison between treatment groups or values obtained from consecutive flux periods unless otherwise noted. Duplicate cpm values were averaged for use in the calculation of unidirectional ²²Na⁺ flux for a given 30-min period. Evaluation of statistical significance was determined using GraphPad Prism 9 software, via unpaired Student’s t test for comparison between two treatment groups/flux periods, or via one-way analysis of variance (ANOVA) with Tukey’s post hoc when comparing more than two treatment groups/flux periods. Statistical test details are given in the figure legends. For correlation analysis of change in I_SC (ΔI_SC) versus change in M-S Na⁺ (ΔJ_M-S) flux, Pearson’s r coefficient was determined using product-moment correlation analysis in GraphPad Prism 9 software. Statistical significance was defined as P < 0.05 for all analyses.

RESULTS

K_V7 Channel Activator, Flupirtine, Attenuates Neurogenic Secretion in Rat Distal Colon

Previously studies in our laboratory have shown that flupirtine inhibits neurogenic Cl⁻ secretion (I_SC), while sparing the effects of epithelium-targeted secretagogues. Figure 1 demonstrates that a similar pattern is observed in the rat colon. Segments of mid-distal colon pretreated with flupirtine (FLU; 100 µM) exhibited markedly reduced basal I_SC compared with control (untreated vs. FLU: 18.6 ± 1.2 vs. 121.2 ± 11.6 µA/cm²; P < 0.01; Fig. 1, A and B). Flupirtine-treated tissues responded to serosal VIP [epithelial VPAC receptor ligand; untreated vs. FLU: ΔI_SC(VIP) = 84.1 ± 16.0 vs. 94.1 ± 15.8 µA/cm²; P = 0.67; Fig. 1D] but displayed a significantly diminished response to serosal veratridine [Na_V channel activator; untreated vs. FLU: ΔI_SC(VER) = 69.5 ± 11.5 vs. 24.9 ± 4.0 µA/cm²; P < 0.01; Fig. 1C]. Labeling of K_V7.2 in MAP-2⁺ cells of rat submucosal ganglia (Fig. 1E) confirmed the presence of flupirtine-sensitive channels in these tissues, consistent with those observed in mouse. These preliminary studies confirm that the effects of flupirtine are similar between mouse and rat with respect to neurogenic secretion and support the use of the rat colon as a valid model for studying the effects flupirtine on Na⁺ transport in the experiments to follow.

Figure 1.

Flupirtine inhibits neurogenic Cl⁻ secretion in rat distal colon. Short-circuit current (I_SC) recordings from control (open circles) and 100 µM flupirtine-treated (closed circles) rat distal colon in response to 10 µM veratridine (VER) (A) or 100 nM vasoactive intestinal polypeptide (VIP) (B). All drugs were added to the serosal chamber bath. C: group data showing the change in I_SC following veratridine administration in control and flupirtine-treated tissues. D: group data showing the change in I_SC following VIP administration in control and flupirtine-treated tissues. Lines and error bars represent mean ± SE. n = 5 tissues obtained from at least three separate animals for all groups; *P < 0.05, as determined by unpaired Student’s t test. E: immunofluorescent labeling of flupirtine-sensitive K_V7.2 channels (green) in MAP2⁺ cells (red) from a submucosal ganglion of rat distal colon. Nuclei are labeled with DAPI (blue). Images were captured at ×20 magnification. Scale bar = 20 µm. MAP2⁺, mouse antimicrotubule-associated protein 2.

Flupirtine Enhances Na⁺ Absorption in Rat Distal Colon in an ENS-Dependent Manner

²²Na⁺ flux experiments were performed next to determine whether epithelial Na⁺ transport was also altered by flupirtine. For initial experiments, colonic tissues were allowed to equilibrate for 15 min with or without flupirtine in the serosal bath, before a subsequent 30-min flux period as described above. Na⁺ fluxes as well as I_SC and G_TE data are listed in Table 1. Under these conditions, flupirtine enhanced net Na⁺ absorption by ∼75% (untreated vs. FLU: 6.6 ± 1.2 vs. 11.7 ± 1.2 µEq/cm²·h; P < 0.05) via a significant increase in the mucosal-to-serosal (M-S) flux (P < 0.05), as well as a slight decrease in serosal-to-mucosal (S-M), which was not statistically significant (P = 0.148). At the same time, basal I_SC and G_TE were both reduced in flupirtine-treated tissues (P < 0.01 for both).

Table 1.

Effect of flupirtine on unidirectional ²²Na⁺ fluxes in rat distal colon

	22Na+ Fluxes
Condition	n	JM-S, µEq/cm2·h	JS-M, µEq/cm2·h	JNet, µEq/cm2·h	n	ISC, µa/cm2	GTE, mS/cm2
Untreated	6	11.2 ± 0.6	4.2 ± 0.7	6.6 ± 1.2	12	73.7 ± 9.4	6.6 ± 0.4
Flupirtine	6	14.3 ± 0.9	3.0 ± 0.4	11.7 ± 1.2	12	19.1 ± 2.5	4.7 ± 0.4
P		0.0220*	0.1480	0.0117*		<0.0001*	<0.0031*

Unidirectional ²²Na⁺ fluxes were measured across seromuscular-stripped rat distal colon under voltage-clamped conditions in the presence and absence of 100 µM serosal flupirtine. Mucosal-to-serosal (M-S) and serosal-to-mucosal (S-M) fluxes were measured over a 30-min period. Net fluxes were calculated by subtracting S-M from M-S fluxes using tissues pairs matched by conductance, as described in methods. Flux values are given as mean ± SE from six tissue pairs. I_SC and G_TE values represent the averages recorded over the 30-min flux period and are given as mean ± SE from 12 tissues (6 tissues each for M-S and S-M fluxes). P values are given for each treatment group/parameter. G_TE, transepithelial conductance; I_SC, short-circuit current; J_M-S, mucosal-to-serosal flux; J_S-M, serosal-to-mucosal flux.

*Significance compared with control (untreated), as determined by unpaired Student’s t test.

Next, experiments were designed to determine whether the enhancement of Na⁺ absorption by flupirtine was mediated through effects on the ENS, as is the case for the inhibition of epithelial Cl⁻ secretion. To test this, Na⁺ fluxes were measured in sequential 30-min periods, before and after application of flupirtine, using tissues that were either untreated (control) or pretreated with 1 µM tetrodotoxin (Na_V channel blocker) to abolish ENS activity (12, 44). Flupirtine-sensitive I_SC was much greater in the absence versus in the presence of tetrodotoxin [control vs. TTX: ΔI_SC(FLU) = −92.8 ± 11.4 vs. –9.5 ± 2.6 µA/cm²; P < 0.01; Fig. 2C], as was flupirtine-sensitive G_TE [control vs. TTX: ΔG_TE(FLU) = −2.9 ± 0.5 vs. −1.0 ± 0.1 mS/cm²; P < 0.01; Fig. 2D). Na⁺ absorption was acutely enhanced by flupirtine in control tissues (basal vs. FLU: 2.4 ± 1.0 vs. 7.8 ± 0.8 µEq/cm²·h; P < 0.05) via an increase in M-S flux (P < 0.01) and a slight decrease in S-M flux (P < 0.05). However, pretreatment with tetrodotoxin negated this effect (basal vs. FLU: 10.4 ± 1.6 vs. 9.9 ± 0.4 µEq/cm²·h; P = 0.55). Table 2 provides the details of Na⁺ fluxes, I_SC, and G_TE data from these experiments.

Figure 2.

Flupirtine-induced I_SC and G_TE inhibition is attenuated by tetrodotoxin. A, B: I_SC and G_TE recordings from control (open circles) and 1 µM tetrodotoxin (TTX)-treated (closed circles) seromuscular-stripped rat distal colon in response to 100 µM flupirtine (FLU). All drugs were added to the serosal chamber bath. C: group data showing the change in I_SC following flupirtine administration in control and tetrodotoxin-treated tissues. D: group data showing the change in G_TE following flupirtine administration in control and tetrodotoxin-treated tissues. Lines and error bars represent mean ± SE. n = 12 for control and n = 10 for tetrodotoxin-treated tissues, obtained from six and five separate animals, respectively; **P < 0.01, as determined by unpaired Student’s t test. ²²Na⁺ fluxes were also performed during these experiments and are summarized in Table 2. G_TE, transepithelial conductance; I_SC, short-circuit current.

Table 2.

Effect of tetrodotoxin on flupirtine-stimulated ²²Na⁺ fluxes in rat distal colon

		22Na+ Fluxes
	Flux period	n	JM-S, µEq/cm2·h	JS-M, µEq/cm2·h	JNet, µEq/cm2·h	n	ISC, µa/cm2	GTE, mS/cm2
Control	Basal	6	6.1 ± 0.9	3.7 ± 0.2	2.4 ± 1.0	12	109.2 ± 11.3	7.4 ± 0.5
	Flupirtine	6	10.7 ± 0.7	2.9 ± 0.2	7.8 ± 0.8	12	15.8 ± 1.4	4.5 ± 0.3
	P		0.0022*	0.0365*	0.0017*		<0.0001*	<0.0001*
TTX-treated	Basal	5	14.0 ± 0.8	3.6 ± 0.3	10.4 ± 1.6	10	24.4 ± 3.0	5.7 ± 0.2
	Flupirtine	5	13.8 ± 0.6	3.9 ± 0.3	9.9 ± 0.4	10	14.9 ± 0.8	4.7 ± 0.2
	P		0.8437	0.5640	0.5541		<0.0001*	<0.0031*

Unidirectional ²²Na⁺ fluxes were measured across seromuscular-stripped rat distal colon under voltage-clamped conditions before (basal) and after the addition of 100 µM flupirtine (flupirtine) to the serosal chamber bath in the absence (control) and presence of tetrodotoxin (TTX-treated). Flux values are given mean ± SE from six tissue pairs from control and five tissue pairs from the tetrodotoxin-treated group. I_SC and G_TE values represent the averages recorded over final 5 min of each 30-min flux period and are given as mean ± SE from 12 control and 10 tetrodotoxin-treated tissues. P values are given for each treatment group/parameter. G_TE, transepithelial conductance; I_SC, short-circuit current; J_M-S, mucosal-to-serosal flux; J_S-M, serosal-to-mucosal flux.

*Significance when compared with the basal flux period, as determined by unpaired Student’s t test.

Figure 3A summarizes the net Na⁺ absorption measured before and after flupirtine treatment in the presence and absence of tetrodotoxin. When the net flux values were compared with each other via one-way ANOVA with Tukey’s post hoc, a significant increase was again observed only in control tissues (pre- vs. postflupirtine; P < 0.01), but not tetrodotoxin-treated tissues (pre- vs. postflupirtine; P > 0.97). In addition, net Na⁺ absorption was significantly greater in both tetrodotoxin- and tetrodotoxin/flupirtine-treated tissues compared with the basal flux for control (P < 0.01 for both). However, there was not a significant difference in Na⁺ flux between control tissues postflupirtine and tetrodotoxin-treated tissue either before or after flupirtine application (P = 0.11 and 0.23, respectively).

Figure 3.

Flupirtine enhances Na⁺ absorption via a tetrodotoxin-sensitive mechanism. A: net Na⁺ fluxes before (open circles/squares) and after (closed circles/squares) the addition of 100 µM serosal flupirtine in control and 1 µM tetrodotoxin (TTX)-treated seromuscular-stripped rat distal colon. Values in A are derived from Table 2. B: net Na⁺ fluxes in time-matched tissues that were untreated (Control, open triangles), treated with 100 µM serosal flupirtine (FLU; closed triangles), or treated with 1 µM serosal tetrodotoxin (TTX; open squares). Values in B are derived from Tables 1 and 2. Lines and error bars represent mean ± SE from six (control) or five (tetrodotoxin-treated) tissue pairs obtained from six and five separate animals, respectively. *P < 0.05, **P < 0.01, as determined by one-way ANOVA with Tukey’s post hoc.

In these experiments, ²²Na⁺ fluxes were measured in sequential 30-min periods, before and after administration of flupirtine. This differs from the experiments summarized in Table 1, where fluxes were time-matched and measured either in the presence or absence of flupirtine. The additional equilibration time at the beginning of those experiments may explain the relatively higher rates of net Na⁺ flux, as more time had passed before the addition of isotope. For additional comparison, time-matched Na⁺ fluxes were compiled from both sets of experiments (Tables 1 and 2) and are depicted in Fig. 3B. When compared via one-way ANOVA with Tukey’s post hoc, Na⁺ absorption was again significantly greater in flupirtine-treated tissues (P < 0.05), whereas no difference was observed between treatment with flupirtine and tetrodotoxin (P = 0.799). Although not statistically significant (P = 0.076), Na⁺ absorption was also greater in tetrodotoxin-treated colon compared with control.

Flupirtine Enhances Na⁺ Absorption That Is Primarily Mediated by NHE-3

Increased Na⁺ absorption without an accompanying increase in I_SC indicates that flupirtine enhanced Na⁺ flux through an electroneutral absorptive pathway. Na⁺/H⁺ exchanger isoform 3 (NHE-3) is the predominant electroneutral Na⁺ transporter within the intestinal mucosa (29) and is also known to be regulated by intracellular signaling events occurring in direct response to ENS stimulation. However, Na⁺/H⁺ exchanger isoform 2 (NHE-2) activity has also been reported to play an important role in electroneutral Na⁺ absorption under certain conditions as well (45–47). Studies were therefore initiated to evaluate the relative contributions of NHE-2 and NHE-3 transporters in flupirtine-stimulated Na⁺ absorption.

Na⁺ fluxes were measured in the presence of flupirtine alone (control) or in the presence of 50 µM HOE694 (NHE-2 selective inhibitor) (47), 10 µM S3226 (NHE-3 selective inhibitor) (48), or 1 mM amiloride (nonselective inhibitor of electroneutral/electrogenic Na⁺ absorption) (49). As illustrated in Fig. 4, mucosal HOE694 had no discernable effect (FLU vs. HOE694: 10.5 ± 0.4 vs. 10.0 ± 0.5 µEq/cm²·h; P = 0.98). However, net Na⁺ absorption was substantially inhibited by both S3226 (FLU vs. S3226: 10.5 ± 0.4 vs. 4.2 ± 0.3 µEq/cm²·h; P < 0.01) and amiloride (FLU vs. AMIL: 10.5 ± 0.4 vs. 1.9 ± 0.5 µEq/cm²·h; P < 0.01), owing to a large decrease in M-S flux (details given in Table 3). Amiloride inhibited Na⁺ absorption to a greater extent than S3226 amiloride (S3226 vs. AMIL: 4.2 ± 0.3 vs. 1.9 ± 0.5 µEq/cm²·h; P < 0.01), also a result of reduced M-S flux (see Table 3), indicating that a concentration of 10 µM S3226 may not be sufficient to maximally inhibit NHE-3 under these conditions. Nonetheless, S3226 abolished the majority (∼70%) of amiloride-sensitive Na⁺ absorption in the presence of flupirtine, suggesting critical involvement of NHE-3 activity in this process.

Figure 4.

Flupirtine-stimulated Na⁺ absorption is primarily mediated by Na⁺/H⁺ exchanger isoform 3 (NHE-3). Net Na⁺ fluxes in the presence of 100 µM serosal flupirtine alone (open circles), or flupirtine plus 50 µM mucosal HOE694 (NHE-2-selective inhibitor; closed circles), 10 µM mucosal S3226 (NHE-3-selective inhibitor; light gray squares), or 1 mM mucosal amiloride (AMIL, nonselective NHE/ENaC inhibitor; dark gray triangles). Lines and error bars represent mean ± SE from five tissue pairs obtained from five separate animals, for all groups. **P < 0.01, as determined by one-way ANOVA with Tukey’s post hoc.

Table 3.

Effect of HOE694, S3226, and amiloride on unidirectional ²²Na⁺ fluxes in flupirtine-treated rat distal colon

	22Na+ Fluxes
Condition	n	JM-S, µEq/cm2·h	JS-M, µEq/cm2·h	JNet, µEq/cm2·h	n	ISC, µa/cm2	GTE, mS/cm2
Flupirtine	5	14.1 ± 0.3	3.2 ± 0.2	10.5 ± 0.4	10	19.6 ± 2.0	5.3 ± 0.5
HOE694	5	13.0 ± 0.7	3.0 ± 0.2	10.0 ± 0.5	10	17.7 ± 1.8	5.6 ± 0.5
S3226	5	6.5 ± 0.4*,#	2.2 ± 0.2*,#	4.2 ± 0.3*,#	10	12.8 ± 0.8*	4.0 ± 0.1*,#
Amiloride	5	4.3 ± 0.5*,#,£	2.4 ± 0.1*	1.9 ± 0.5*,#,£	10	13.3 ± 0.9*	4.5 ± 0.1

Unidirectional ²²Na⁺ fluxes were measured across seromuscular-stripped rat distal colon under voltage-clamped conditions in the presence of 100 µM serosal flupirtine alone, or flupirtine plus 50 µM mucosal HOE694, 10 µM mucosal S3226, or 1 mM mucosal amiloride. Flux values are given as mean ± SE from five tissue pairs for all treatment groups. I_SC and G_TE values were averaged from 10 tissues (5 tissue pairs) over the 30-min flux period for each treatment group and are given as mean ± SE. G_TE, transepithelial conductance; HOE694, NHE-2 selective inhibitor; I_SC, short-circuit current; J_M-S, mucosal-to-serosal flux; J_S-M, serosal-to-mucosal flux; NHE, Na⁺/H⁺ exchanger isoform; S3226, NHE-3 selective inhibitor.

*P < 0.01 versus flupirtine alone, #P < 0.05 versus flupirtine plus HOE694, £P < 0.05 versus flupirtine plus S3226, as determined by one-way ANOVA with Tukey’s post hoc.

Extensive characterization of acute NHE-3 regulation in the intestine has revealed a primary mechanism involving rapid changes in membrane localization through trafficking between the apical membrane and an intracellular vesicular pool in response to prosecretory stimuli (50). If flupirtine stimulates NHE-3-dependent Na⁺ absorption by suppressing neurogenic input (Figs. 1–4), then NHE-3 trafficking is likely to play a role in this process. Qualitatively, seromuscular-stripped rat distal colon that was treated with flupirtine appeared to have more NHE-3 protein localized near the apical membrane (labeled with phalloidin) compared with control (Fig. 5). The NHE-3-specific labeling was more diffused beneath the apical surface in untreated control tissue, whereas it was appearing more focused around the apical membrane in flupirtine-treated tissue. Although quantitative analysis was not performed here, the apparent change in NHE-3 localization in response to flupirtine-mediated suppression of ENS activity is congruent with the results from our Ussing chamber experiments, as well as with the currently accepted model of acute NHE-3 regulation in colonic epithelia.

Figure 5.

Effect of flupirtine on surface localization of NHE-3 in rat distal colon. Seromuscular-stripped rat distal colons, incubated for 10 min in the absence (control, row A_i–A_iv) or presence of flupirtine (FLU, row B_i–B_iv), were labeled for NHE-3 (green) and filamentous actin (phalloidin, red). Nuclei are labeled with DAPI (blue). Zoom of control (C) and flupirtine-treated (D) colon are also shown below. Arrows indicate the presence of subapical NHE-3 proteins, which are more prominent in control versus flupirtine-treated rat distal colon. Images were captured at ×20 magnification. Scale bar = 25 µm. NHE-3, Na⁺/H⁺ exchanger isoform 3.

Flupirtine Does Not Enhance ENaC-Mediated Na⁺ Absorption in Hyperaldosteronaemic Rat Distal Colon

Both electroneutral and (NHE-2/3-mediated) and electrogenic (ENaC-mediated) Na⁺ absorption are present in the colon of most mammalian species, but there is considerable variability in the functional importance, as well as the anatomical distribution of these pathways (45). Along with this, second messenger systems responding to ENS-derived neurotransmitters differentially regulate NHE-3 and ENaC as well. For example, cAMP is known rapidly inhibit NHE-3 activity, whereas ENaC has been shown to be activated by elevated cAMP (51, 52) and its expression may be governed by Ca²⁺ events (53). Beyond this, there has been little work done to elucidate the regulation of ENaC by ENS neurotransmission, specifically, within the colon. To determine the effect of flupirtine specifically on ENaC-dependent absorption, Na⁺ fluxes were measured in distal colon of dietary Na⁺-depleted rats, which exhibit secondary hyperaldosteronism (30). One of the hallmark characteristics of this model is the induction of electrogenic (ENaC-mediated) Na⁺ absorption and the concurrent downregulation of NHE-3, as well as NHE-2 (45).

Basal I_SC is typically much higher (two- to fourfold) in the distal colon of hyperaldosteronaemic (a.k.a. “aldo”) rats compared with normal controls, which is attributable mostly to ENaC activity. In “aldo” distal colon, addition of flupirtine to the serosal bath caused a reduction in I_SC (basal vs. FLU: 297.8 ± 7.2 vs. 193.5 ± 11.8 µA/cm²; P < 0.01) and G_TE (basal vs. FLU: 11.8 ± 0.7 vs. 7.4 ± 0.3 mS/cm²; P < 0.01; Fig. 6, A and B), but this was qualitatively more gradual in onset compared with the I_SC inhibition observed in normal rats. Net Na⁺ flux was unaffected by flupirtine in “aldo” distal colon (basal vs. FLU: 11.4 ± 0.7 vs. 10.6 ± 0.8 µEq/cm²·h; P = 0.66; Fig. 6C), despite the concurrent reduction in I_SC and G_TE, suggesting a portion of the basal I_SC was not mediated by electrogenic Na⁺ transport (e.g., Cl⁻/ HCO₃⁻ secretion). However, subsequent application of 10 µM amiloride (ENaC-selective at this concentration) (54) dramatically inhibited I_SC (FLU vs. AMIL: 193.5 ± 11.8 vs. −27.9 ± 2.5 µA/cm²; P < 0.01) and G_TE (FLU vs. AMIL: 7.4 ± 0.3 vs. 5.0 ± 0.2 mS/cm²; P < 0.01; Fig. 6, A and B), as well as Na⁺ absorption (FLU vs. AMIL: 10.6 ± 0.8 vs. 2.0 ± 0.3 µEq/cm²·h; P < 0.01; Fig. 6C) due to a substantial reduction in M-S flux (P < 0.01) (Table 4).

Figure 6.

Flupirtine does not enhance ENaC-mediated Na⁺ absorption in “aldo” rat distal colon. I_SC (A) and G_TE (B) recordings from seromuscular-stripped hyperaldosteronaemic (“aldo”) rat distal colon during basal, 100 µM serosal flupirtine (FLU) and 10 µM apical amiloride (AMIL) flux periods. C: net ²²Na⁺ fluxes, measured during the sequential 30-min periods shown in A–B. Lines and error bars represent mean ± SE from five tissue pairs obtained from five separate animals. D: representative Western blot showing NHE-3 protein expression in control and “aldo” rat distal colonic epithelium. Protein ladder is shown in the extreme left lane, and corresponding molecular weights are given to the left of the blot images. E: densitometric quantification of two blots similar to the one shown in panel D (n = 4 separate animals, for both control and “aldo” groups). NHE-3 band intensity was normalized to actin as a loading control and represented as relative abundance in reference to control samples. Lines and error bars represent mean ± SE. ^**P < 0.01, as determined by one-way ANOVA with Tukey’s post hoc (C) or unpaired Student’s t test (E). ENaC, epithelial Na⁺ channel; G_TE, transepithelial conductance; I_SC, short-circuit current; NHE-3, Na⁺/H⁺ exchanger isoform 3.

Table 4.

Effect of flupirtine and amiloride on unidirectional ²²Na⁺ fluxes in “aldo” rat distal colon

	22Na+ Fluxes
Condition	n	JM-S, µEq/cm2·h	JS-M, µEq/cm2·h	JNet, µEq/cm2·h	n	Isc, µA/cm2	GTE, mS/cm2
Basal	5	13.9 ± 0.7	2.4 ± 0.2	11.4 ± 0.7	10	297.8 ± 7.2	11.8 ± 0.7
Flupirtine	5	12.3 ± 0.8	1.6 ± 0.2*	10.7 ± 0.8	10	195.5 ± 11.8*	7.4 ± 0.3*
Amiloride	5	3.7 ± 0.3*,#	1.7 ± 0.1*	2.0 ± 0.3*,#	10	−27.9 ± 2.5*,#	5.0 ± 0.2*,#

Unidirectional ²²Na⁺ fluxes were measured across seromuscular-stripped “aldo” rat distal colon under voltage-clamped conditions for three sequential flux periods: under basal conditions, in the presence of 100 µM serosal flupirtine, and flupirtine plus 10 µM mucosal amiloride. Flux values are given as mean ± SE from five tissue pairs for all treatment groups. I_SC and G_TE values were averaged from the 10 tissues (5 tissue pairs) over the final 5 min of each 30-min flux period for each treatment group and are also given as mean ± SE. G_TE, transepithelial conductance; I_SC, short-circuit current; J_M-S, mucosal-to-serosal flux; J_S-M, serosal-to-mucosal flux.

*P < 0.01 versus basal flux period, #P < 0.05 versus flupirtine, as determined by one-way ANOVA with Tukey’s post hoc.

The profound effects of low-dose amiloride on I_SC, G_TE, and Na⁺ flux indicate the prominent activity of ENaC in the distal colon of these animals. In addition, Western blot analysis using distal colonic mucosae from normal and “aldo” rats confirmed that NHE-3 expression was significantly downregulated (<25% of control), as shown in the representative blot image (Fig. 6D) and densitometric quantification of two similar blots (Fig. 6E). Based on these observations, activation of K_V7 channels in the ENS may only enhance electroneutral NHE-3-mediated absorption in rat colon, as ENaC-mediated absorption is already quite high at baseline, and not altered by treatment with flupirtine.

Effects of Flupirtine on Cl⁻ Secretion and Na⁺ Absorption Are Specific to the Colon

The results described thus far suggest that enhanced Na⁺ absorption in response to K_V7 activation was mediated only by electroneutral, NHE-3-mediated transport. Indeed, NHE-3 is the major Na⁺ absorptive mechanism along the length of the small and large intestine. However, the compositional make-up of submucosal ENS neuronal networks is widely varied between anatomical regions (8). Still undetermined is whether segmental differences exist regarding the response to flupirtine on either Cl⁻ secretion (I_SC) or Na⁺ absorption, as the specific distribution pattern of K_V7 channel subtypes has not been described in these networks to date. Thus, the effects of K_V7 activation on Cl⁻ secretion and Na⁺ absorption were assessed in both the proximal colon and the ileum to determine whether the response was region-specific.

Flupirtine inhibited I_SC (basal vs. FLU: 70.7 ± 14.2 vs. 25.8 ± 2.2 µA/cm²; P < 0.01) and G_TE (basal vs. FLU: 9.7 ± 0.5 vs. 7.6 ± 0.4 mS/cm²; P < 0.01; Fig. 7, A and C) whereas stimulated Na⁺ absorption (basal vs. FLU: 3.1 ± 0.8 vs. 5.7 ± 0.6 µEq/cm²·h; P < 0.05; Fig. 7E) in rat proximal colon, suggesting the presence of flupirtine-sensitive K_V7 channels in this region. Flupirtine-stimulated Na⁺ absorption was also inhibited by mucosal S3226 (FLU vs. S3226: 5.7 ± 0.6 vs. 2.8 ± 0.2 µEq/cm²·h; P < 0.01; not illustrated), suggesting that NHE-3 is the primary mediator in the proximal colon as well. In the ileum, however, flupirtine did not inhibit I_SC (basal vs. FLU: 96.2 ± 6.0 vs. 97.9 ± 6.4 µA/cm²; P = 0.84), slightly increased G_TE (basal vs. FLU: 23.9 ± 0.9 vs. 26.7 ± 0.8 mS/cm²; P < 0.05; Fig. 6, B and D), and did not alter Na⁺ absorption (basal vs. FLU: 4.9 ± 0.6 vs. 5.7 ± 0.7 µEq/cm²·h; P = 0.41; Fig. 6F). Together, these data indicate that the effects of K_V7 channel activation in ENS neurons may be specific to the colon, at least in rats.

Stimulation of Na⁺ Absorption and Inhibition of Cl⁻ Secretion Are Significantly Correlated in Flupirtine-Treated Rat Distal Colon

Our data suggest that activation of K_V7 channels in the ENS simultaneously inhibits electrogenic secretion and stimulates electroneutral absorption in the colon. To assess whether these two outcomes were correlated to each other, the change in M-S Na⁺ flux was plotted against the change in I_SC recorded from the same tissue. Tissues from all experiments in this project were used for analysis that incorporated normal distal or proximal colon, where fluxes were measured before and after flupirtine administration (n = 20). Simple linear regression and correlation analysis revealed a significant relationship (P < 0.0001; r² = 0.6051) between the magnitude of I_SC inhibition and the magnitude in M-S flux enhancement (Fig. 8). Such a strong correlation between these effects is presumably the result of varying degrees of basal ENS activity being inhibited by K_V7 activation, leading to a proportional (but opposite) response with respect to Cl⁻ secretion and Na⁺ absorption because both processes are governed by ENS input.

Figure 8.

Stimulation of Na⁺ absorption is significantly correlated to the inhibition of Cl⁻ secretion in rat colonic tissues treated with flupirtine. The absolute value of the change in I_SC (inhibition of Cl⁻ secretion) was plotted against change in M-S ²²Na⁺ flux (stimulation of Na⁺ absorption) from rat colonic tissues that were treated with flupirtine. Every tissue used in this study was included that was assigned as “M-S,” where unidirectional ²²Na⁺ fluxes were measured before and after addition of 100 µM flupirtine to the serosal chamber bath (n = 20). Simple linear regression analysis was performed using GraphPad Prism 9.0 software, and the fit is shown as a solid black line. P < 0.0001 indicates the line of best fit has a slope significantly greater than zero. I_SC, short-circuit current; M-S, mucosal-to-serosal.

DISCUSSION

Here we provide evidence that a K_V7 channel activator, flupirtine, both suppresses neurogenic Cl⁻ secretion and enhances NHE-3-dependent Na⁺ absorption simultaneously in the rat colon. In support of this claim are the following observations: 1) Flupirtine-sensitive K_V7.2 channels were detected by immunofluorescence in submucosal ganglia, and ENS-driven Cl⁻ secretion (via veratridine) was suppressed by flupirtine, whereas the response to epithelium stimulation directly (via VIP) was unaffected. 2) Flupirtine inhibited I_SC and G_TE and increased both M-S and net Na⁺ fluxes across rat colonic tissues, but this effect was not observed in the presence of ENS blockade via tetrodotoxin. 3) S3226 and amiloride, but not HOE694, significantly inhibited M-S and net Na⁺ fluxes in the presence of flupirtine, whereas NHE-3 localization near the apical membranes of surface epithelial cells was apparently enhanced in flupirtine-treated colon compared with control. 4) Flupirtine did not alter Na⁺ fluxes in “aldo” rat distal colon, which has significantly reduced NHE-3 expression/activity. 5) Flupirtine inhibited I_SC and G_TE and increased both M-S and net Na⁺ fluxes in the proximal colon, but not in the ileum, suggesting a colon-specific effect.

Under normal conditions, the primary function of the colon is to desiccate fecal material via active absorption of Na⁺, Cl⁻, and water. Efferent signals from the ENS can halt these absorptive processes and induce a secretory response to either normal (e.g., mechanical distension) (55, 56) or pathological stimuli (e.g., infection or stress) (57, 58). Often, epithelial secretion occurs in conjunction with increased motility of colonic smooth muscle to propagate material onward toward its eventual elimination (59). Secretomotor neurons of the ENS are thus frequently targeted in the treatment of diarrheal GI disorders, such as IBS-D. Alosetron (serotonin receptor antagonist) and loperamide (opioid receptor agonist) are two of the most commonly prescribed IBS-D treatments, both of which dampen ENS activity (60, 61). However, these drugs and others that modulate ENS activity are associated with adverse side effects, such as nausea, abdominal pain, constipation, and dependence (26). Novel pharmacotherapies that dampen ENS-driven symptoms may therefore potentially have great therapeutic potential. The results from this study demonstrate that K_V7 channel activators may be of use in this context, as both inhibition of Cl⁻ secretion and stimulation of Na⁺ absorption favor water absorption and would likely combat diarrhea. Furthermore, if the effects observed here are truly specific to the colon (see Fig. 7), potential off-target effects of interfering with ENS function in the small intestine (i.e., related to normal digestive function) may possibly be avoided as well. In vivo studies will be required for a more complete assessment of efficacy.

Figure 7.

Flupirtine inhibits Cl⁻ secretion and stimulates Na⁺ absorption in normal rat proximal colon, but not ileum. I_SC recordings from seromuscular-stripped rat proximal colon (A) and ileum (B) before and after the addition of 100 µM flupirtine (FLU) to the serosal chamber bath. G_TE recordings from seromuscular-stripped rat proximal colon (C) and ileum (D) before and after the addition of 100 µM FLU to the serosal chamber bath. Net ²²Na⁺ fluxes measured in seromuscular-stripped rat proximal colon (E) and ileum (F), corresponding to the sequential 30-min flux periods before (basal) and after addition of flupirtine (FLU) as illustrated in panels A–D. Lines and error bars represent mean ± SE from five tissue pairs obtained from five separate animals for proximal colon and three separate animals for ileum. *P < 0.05, as determined by unpaired Student’s t test. G_TE, transepithelial conductance; I_SC, short-circuit current.

The clinical use of K_V7 channel activators, including flupirtine, has historically been centered around the treatment of chronic pain and epilepsy (21). These drugs are effective analgesics and anticonvulsants because of their suppressive effects on neuronal excitability. Opening of K_V7 channels in the neuronal plasma membrane produces a hyperpolarizing K⁺ current that stabilizes membrane potential and thus dampens the response to incoming stimuli (62). However, the exact mechanism of action in terms of their analgesic effects is not completely clear, and it may involve interactions with γ-amino butyric acid (GABA) receptors, G protein-coupled inward rectifier K⁺ (GIRK) channels, or possibly voltage-dependent Mg²⁺ block of NMDA receptors (63–65). Whether direct or indirect, here we show for the first time that the suppressive effects of K_V7 channel activators can be seen in the context of the ENS and its modulation of multiple ion transport pathways in the colon. Somewhat related to this, Peiris et al. demonstrated that activation of K_V7 channels in sensory afferents with retigabine—a structural analog of flupirtine—blunted the nociceptive response to noxious stimuli in mouse and human colon (20). The potential benefits of targeting neuronal K_V7 channels in the treatment of GI disorders such as IBS-D may therefore be multifaceted. Generalized GI-related side effects (i.e., nausea, abdominal discomfort) are associated with flupirtine use in patients treated for chronic pain. However, to our knowledge, there has not been an analysis to determine whether patients with preexisting GI conditions, such as IBS-D, experienced these side effects differentially. A thorough review of the clinical data surrounding the use of flupirtine in patients would be useful in this respect.

One of the major findings from this study is that specifically NHE-3-mediated Na⁺ absorption is enhanced by K_V7 activation. Modulation of NHE-3 activity by secretagogues originating from ENS neurons has been thoroughly described (29). Cholinergic and VIP-ergic neurons constitute the two primary secretomotor classes within the submucosal plexus of the colon (8, 66). Both of the second messenger systems induced by the activation of their target receptors—Ca²⁺ and cAMP, respectively—inhibit NHE-3 activity by triggering its rapid internalization from the apical membranes of surface colonocytes. Our observations that NHE-3 surface localization is enhanced by treatment with flupirtine and that NHE-3-dependent Na⁺ absorption is increased by flupirtine to a degree that is correlated to the inhibition of Cl⁻ secretion, fits well into the existing model of Na⁺ and Cl⁻ transport modulation by ENS input. In spite of this, the experimental data presented here offer no definitive insight regarding the intracellular events underlying the observed effects on Na⁺ or Cl⁻ transport. Further studies will be required to assess the involvement of cholinergic or VIP-ergic (or other) pathways, as well as to resolve the details of NHE-3 trafficking and/or other possible means of regulation in response to flupirtine.

Flupirtine is generally considered to be nonselective for K_V7 channels. One member of this family, K_V7.1, has a well-defined role in the colonic epithelium itself (67). K_V7.1 is expressed in crypt cells and localizes to the basolateral membrane to support apical Cl⁻ secretion by maintaining the electrochemical driving force for sustained Cl⁻ efflux (68). In this study, as well as recent studies from our laboratory, there was no indication that flupirtine stimulates Cl⁻ secretion by opening basolateral K_V7.1 channels, despite the drug being “nonselective” for the family subtypes. Importantly, another nonselective K_V7 channel activator, retigabine, does not act on the K_V7.1 subtype in vitro (16). Although analysis of the effects of flupirtine, specifically, on K_V7.1 is lacking, there have been reports that the channel may also be insensitive to flupirtine (21). Indeed, flupirtine and retigabine are very closely related in both structure and function. We have also observed no effect from flupirtine on basal or stimulated Cl⁻ secretion in T84 cells, which express K_V7.1 channels. K_V7.1 does exhibit significant homology to other K_V7 family members (69). The reason for the apparent insensitivity to these nonselective activators is unknown but may be related to the association of K_V7.1 with an auxiliary minK (KCNE) subunit in the colonic epithelium (70, 71) or may be related to secondary effects on GABA or NMDA receptors, as mentioned above. Nevertheless, flupirtine clearly modulates epithelial transport in an ENS-dependent manner, as evidenced by the attenuating effect of pretreatment with tetrodotoxin on the response to flupirtine, as well as the similarity in I_SC, G_TE, and Na⁺ absorption properties among tissues treated with either of the two agents. The mechanistic details underlying K_V7 activation of neuronal versus epithelial subtypes, as well as possible “off-target” inhibitory effects within ENS neurons, remain to be elucidated but should be the subject of future studies.

Conclusions

In summary, these data show that the K_V7 channel activator, flupirtine, simultaneously inhibits Cl⁻ secretion and stimulates NHE-3-dependent Na⁺ absorption in the rat colon. Further, these effects are likely mediated by an ENS-dependent mechanism. Both of inhibition of Cl⁻ secretion and stimulation of Na⁺ absorption may have significant bearing on water absorption in the colon. Therefore, the use of K_V7-activating drugs for the treatment of ENS-associated diarrheal conditions—such as IBS-D—should be explored in the future.

GRANTS

This study was supported by the National Institute of Diabetes and Digestive and Kidney Diseases R01DK104791 and DK112085 grants to V.M.R.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.J.N. and V.M.R. conceived and designed research; A.J.N. performed experiments; A.J.N. and V.M.R. analyzed data; A.J.N. and V.M.R. interpreted results of experiments; A.J.N. prepared figures; A.J.N. drafted manuscript; A.J.N. and V.M.R. edited and revised manuscript; A.J.N. and V.M.R. approved final version of manuscript.