Mucosal potassium efflux mediated via Kcnn4 channels provides the driving force for electrogenic anion secretion in colon

Abstract

Intermediate conductance K⁺ (Kcnn4) channels are present in both mucosal and serosal membranes of colon. However, only serosal Kcnn4 channels have been shown to be essential for agonist-induced (cAMP and Ca²⁺) anion secretion. The present study sought to determine whether mucosal Kcnn4 channels also play a role in colonic anion secretion. Mucosal-to-serosal and serosal-to-mucosal unidirectional ⁸⁶Rb (K⁺ surrogate) fluxes as well as short-circuit current (I_sc; a measure of anion secretion) were measured under voltage-clamp conditions in distal colon from rats fed either a standard or K⁺-free diet. 5,6-Dichloro-1-ethyl-1,3-dihydro-2H-benzimidazole-2-one (DC-EBIO) was used to activate Kcnn4 channels. Mucosal DC-EBIO both induced K⁺ secretion and enhanced anion secretion in normal rat distal colon. The DC-EBIO-induced K⁺ secretion was completely blocked by nonspecific (Ba²⁺) and Kcnn4-specific (TRAM-34) inhibitors, but was not blocked by the large-conductance K⁺ (iberiotoxin), small-conductance K⁺ (apamin), or KCNQ1 (chromanol 293B) specific blockers. Ba²⁺ and TRAM-34 also inhibited DC-EBIO-enhanced anion secretion. The DC-EBIO-enhanced anion secretion was completely inhibited by the nonspecific anion channel blocker 5-nitro-2-(3-phenylpropyl-amino)benzoic acid, whereas it was only partially inhibited by CFTR [CFTR_inh-172, glibenclamide]- and CaCC (niflumic acid)-specific Cl⁻ channel blockers. In contrast, mucosal DC-EBIO-enhanced K⁺ and anion secretion was not present in distal colon of dietary K-depleted rats, indicating absence of mucosal Kcnn4 channels. These observations indicate that mucosal Kcnn4 channels are capable of driving agonist-induced anion secretion mediated via CFTR and CaCC and likely contribute to stool K⁺ losses that accompany diarrheal illnesses.

fluid secretion results from active anion (Cl⁻ and HCO₃⁻) secretion mediated by ion channels that reside within epithelial cell membranes (3). A component of active anion secretion contributes to virtually every diarrheal illness of inflammatory, neuroendocrine, and/or infectious etiology (23). Continuous active anion secretion requires a mechanism for maintaining cell hyperpolarization (3). In human, as well as in rat colon, agonist-induced (cAMP and Ca²⁺) anion secretion appears to be energized by intermediate conductance Ca²⁺-activated K⁺ channels (8, 21, 26). Molecular studies have identified the intermediate-conductance K⁺ channels as Kcnn4 channels (also known as rSK4, rIK1, MIK1, SMIK1, and KCa3.1) (5, 16, 20, 32, 34). Electrophysiological studies have characterized serosal Kcnn4 channels, whereas ion flux studies have provided evidence for mucosal Kcnn4 channels in rat colon (16, 34). Immunological studies have localized Kcnn4-like proteins to both mucosal and serosal membrane domains in intestine and colon (12, 15, 16). In recent studies, we have cloned three distinct Kcnn4 transcripts (Kcnn4a, Kcnn4b, and Kcnn4c) that exhibit tissue- and membrane-specific expression patterns in the colon (2). The Kcnn4a transcripts have been shown to be expressed in colonic smooth muscle, whereas Kcnn4b and Kcnn4c proteins have been localized to the basolateral and mucosal membranes of colonic epithelial cells, respectively (2).

Whereas Kcnn4 channels reside on both mucosal and serosal membranes, only serosal Kcnn4 channels have been hypothesized to hyperpolarize cells and provide the driving force for agonist-induced anion secretion (18, 34). This conclusion is based on observations that clotrimazole (CLT; an antifungal antimicrobial compound) inhibited intracellular free Ca²⁺-induced anion secretion (26, 34). In these studies, intracellular Ca²⁺ levels were enhanced either by carbachol (a cholinergic agonist) or by thapsigargin (an inhibitor of the Ca²⁺-ATPase of the endoplasmic reticulum). However, on the basis of these data, it remains unclear whether mucosal or serosal Kcnn4 channels provide the driving force for anion secretion because increasing intracellular free Ca²⁺ concentration would have activated, whereas lipophilic CLT would have inhibited, Kcnn4 channels within both domains. Thus the present study was initiated to identify whether the driving force for anion secretion in rat distal colon is mediated by activation of serosal Kcnn4 channels alone and/or by concurrent activation of mucosal Kcnn4 channels. In this study, membrane domain-specific Kcnn4 channels were activated by either mucosal or serosal addition of the Kcnn4 channel opener DC-EBIO (5,6-dichloro-1-ethyl-1,3-dihydro-2H-benzimidazole-2-one), which has been shown to activate mucosal cystic fibrosis transmembrane conductance (CFTR) Cl⁻ channels and serosal Kcnn4 channels in the human colonic tumor cell line T84 (29). Unidirectional ⁸⁶Rb⁺ fluxes (a K⁺ surrogate) and short-circuit currents (I_sc; a measurement of anion secretion) were measured in epithelial sheets from rat distal colon under voltage-clamp conditions. Addition of mucosal DC-EBIO enhanced both K⁺ and anion secretion in normal rat distal colon. The mucosal DC-EBIO-induced K⁺ and anion secretion was completely inhibited by the nonspecific K⁺ channel blocker (Ba²⁺), as well as the Kcnn4 channel-specific blocker (TRAM-34). Mucosal application of the nonspecific anion channel blocker NPPB [5-nitro-2-(3-phenylpropyl-amino)benzoic acid] completely inhibited DC-EBIO-enhanced anion secretion but only partially inhibited the DC-EBIO-enhanced K⁺ secretion in normal rat distal colon. DC-EBIO-induced K⁺ and anion secretion were present in the distal colon of the dietary K-depleted rats. These results demonstrate that mucosal Kcnn4 channels provide a component of the driving force for Cl⁻ secretion in secretory diarrhea and likely contribute to stool K⁺ loss.

METHODS

Animals.

Nonfasting normal male Sprague-Dawley rats (200–250 g) were maintained on either standard or a potassium-deficient rat chow (MP Biochemicals, Solon, OH). All animals had access to tap water ad libitum. All the experimental protocols used in this study were approved by the West Virginia University Institutional Animal Care and Use Committee.

Ussing chamber studies.

I_sc and ⁸⁶Rb⁺ (K⁺ surrogate; PerkinElmer, Billerica, MA) fluxes were measured in distal colon mounted under voltage-clamp conditions, as described previously (17, 33). In brief, colon excised from euthanized rats was flushed with ice-cold saline. Mucosal layers were gently separated from serosal muscular layers opened along the mesenteric border. Two distal (1 cm proximal to rectum) segments obtained from each animal were mounted in Lucite chambers (with an opening of 1.12 cm²), and both sides were bathed with equal volume of Ringer solution (in mM: 115 NaCl, 25 NaHCO₃, 2.4 K₂HPO₄, 0.4 KH₂PO₄, 1.2 CaCl₂, 1.2 MgCl₂, and 10 glucose; pH 7.4). The bathing solutions were maintained at 37°C and gassed continuously with 5% CO₂-95% O₂. I_sc and conductance were recorded every 20 s using an automated multichannel voltage/current clamp instrument (Physiological Instruments, San Diego, CA). For flux studies, 1 μCi ⁸⁶RbCl was added to either serosal or mucosal bath solution. After a 45-min equilibration period, serosal-to-mucosal (s-m) and mucosal-to-serosal (m-s) ⁸⁶Rb⁺ fluxes were measured on different tissues under voltage-clamp conditions. Net fluxes were calculated from the difference between m-s and s-m fluxes across tissue pairs that were matched on the basis of differences in basal conductance of <10%. Positive and negative values represent active absorption and active secretion, respectively. In all experiments, three 15-min flux periods were combined to obtain a single flux value for each tissue pair. The I_sc and ⁸⁶Rb fluxes values are expressed as microequivalents per hour per square centimeter, whereas conductance is expressed as millisiemens per square centimeter.

In additional experiments, following basal measurements, conductance, I_sc and ⁸⁶Rb⁺ fluxes were also measured in the presence of 100 μM mucosal and serosal DC-EBIO. Inhibitors were added to mucosal and/or serosal bath solution at the end of the DC-EBIO period: 3 mM Ba²⁺, 50 μM CLT, 50 μM TRAM-34, 100 nM iberiotoxin (IbTX), 10 nM apamin (APA), 30 μM chromanol 293B, 100 μM NPPB, 100 μM niflumic acid, or 300 μM glibenclamide was added to the mucosal and/or serosal bath solution. Na⁺ concentration was adjusted to maintain isosmolarity when Ba²⁺ was used.

Statistics.

Results are presented as means ± SE. Statistical analyses were performed by a paired t-test and Bonferroni's one-way ANOVA post hoc test. P < 0.05 was considered to be statistically significant.

RESULTS

Initial studies were designed to identify whether the Kcnn4 channel opener DC-EBIO would enhance I_sc (as a measure of anion secretion) and ⁸⁶Rb (K⁺ surrogate) fluxes in normal rat distal colon. In these studies, the effect of DC-EBIO added simultaneously to both mucosal and serosal bathing solutions was examined. As shown in Fig. 1A, simultaneous addition of DC-EBIO to mucosal and serosal bathing solutions enhanced I_sc threefold (basal vs. DC-EBIO: 2.1 ± 0.4 vs. 7.0 ± 0.7 μeq·h⁻¹·cm⁻²; P < 0.001). The DC-EBIO activation of mucosal Kcnn4 channels also significantly enhanced the transepithelial tissue conductance from 8.3 ± 0.3 to 11.7 ± 0.6 mS/cm² (P < 0.002). The DC-EBIO-enhanced I_sc and conductance were stable for up to 60 s (Figs. 1A). In parallel, we also examined ⁸⁶Rb fluxes in the same tissues. As shown in Fig. 1B, similar to earlier demonstration by Sweiry and Binder (30), under basal conditions, net ⁸⁶Rb absorption occurs in normal rat distal colon. Mucosal and serosal addition of DC-EBIO did not alter the net ⁸⁶Rb flux (Fig. 1B). However, DC-EBIO significantly enhanced m-s and s-m fluxes (Fig. 1B). These observations indicate that 1) DC-EBIO-enhanced I_sc occurred as a result of anion secretion since net ⁸⁶Rb flux was not altered; and 2) DC-EBIO-activated K⁺ efflux/exit occurs in both mucosal and serosal membranes of normal rat distal colon.

Fig. 1.

Effect of mucosal (m) and serosal (s) presence of 5,6-dichloro-1-ethyl-1,3-dihydro-2H-benzimidazole-2-one (DC-EBIO) on short-circuit current (I_sc) and ⁸⁶Rb fluxes in normal rat distal colon. Both time courses of I_sc (A) and ⁸⁶Rb fluxes (B) were measured as described in methods. A: simultaneous addition of DC-EBIO (100 μM) to both mucosal and serosal medium enhanced I_sc. The DC-EBIO-enhanced I_sc was stable for up to 60 min. B: mucosal to serosal (m-s) and serosal to mucosal (s-m) ⁸⁶Rb fluxes in the absence (open bars), and in the presence of both mucosal and serosal DC-EBIO (solid bars). DC-EBIO significantly enhanced both m-s and s-m but did not alter the net ⁸⁶Rb fluxes. The basal and post-DC-EBIO tissue conductance were 8.2 ± 0.3 and 11.7 ± 0.6 mS/cm² (P < 0.002), respectively. Results presented represent means ± SE from 5 tissue pairs for ⁸⁶Rb fluxes. *P < 0.05, compared with basal fluxes; NS, not significant.

Although the benzimidazolone derivative DC-EBIO has been shown to activate serosal Kcnn4 channels and mucosal Cl⁻ channels in T84 cells (a human colon cancer cell line), the benzimidazolone derivative NS004 has been shown to activate large-conductance K⁺ (BK) and CFTR Cl⁻ channels (14, 22). Therefore, to establish that DC-EBIO-enhanced I_sc occurred as a result of activation of CFTR and that the driving force for the anion secretion is provided by DC-EBIO-activated K⁺ channels, the effect of anion and K⁺ channel blockers (added to both mucosal and serosal bath) were examined on DC-EBIO-enhanced I_sc (Fig. 2). The DC-EBIO-enhanced I_sc was almost completely (93%) inhibited by the nonspecific anion channel blocker NPPB, whereas the CFTR Cl⁻ channel blockers glibenclamide (50%) and CFTR_inh-172 (54%), as well as the CaCC (Ca²⁺-activated Cl⁻ channel) blocker niflumic acid (37%) only partially inhibited the DC-EBIO-enhanced I_sc (Fig. 2A). Simultaneous addition of glibenclamide and niflumic acid exhibited additive (86%) inhibition of the DC-EBIO-enhanced I_sc (Fig. 2A). Similar to anion channel blockers, K⁺ channel blockers also exhibited varying effects on inhibition of DC-EBIO-enhanced I_sc (Fig. 2B). The nonspecific K⁺ channel blocker Ba²⁺ inhibited DC-EBIO-enhanced I_sc by 97%, whereas the Kcnn4 channel blockers CLT and TRAM-34 inhibited DC-EBIO-enhanced I_sc by 98 and 94%, respectively (Fig. 2B). The BK channel blocker IbTX, the small-conductance K⁺ (SK) channel blocker APA, and the KCNQ1 channel blocker chromanol 293B did not inhibit DC-EBIO-enhanced I_sc (Fig. 2B). Both anion channel and K⁺ channel blockers also significantly inhibited DC-EBIO-enhanced conductance (Fig. 2, legend). These observations establish that DC-EBIO-enhanced I_sc occurred as a result of anion secretion mediated by mucosal CFTR and CaCC anion channels. These results also indicate that Kcnn4 channels provided the driving force for DC-EBIO-enhanced anion secretion.

Fig. 2.

Effect of anion channel and K⁺ channel blockers on mucosal and serosal DC-EBIO-enhanced I_sc in normal rat distal colon. Basal I_sc was measured as described in methods. A: DC-EBIO-enhanced I_sc was measured in the absence (Control) and in the presence of 100 μM 5-nitro-2-(3-phenylpropyl-amino)benzoic acid (NPPB), 300 μM glibenclamide (Glib), 100 μM niflumic acid (Niflu), 20 μM CFTR_inh-172 (Inh-172), and Glib + Niflu. B: DC-EBIO-enhanced I_sc was measured in the absence (Control) and after the addition of 3 mM Ba²⁺, 50 μM clotramizole (CLT), 50 μM TRAM-34, 100 nM iberiotoxin (IbTX), 10 nM apamin (APA), and 30 μM chromanol 293B (293B). Absolute DC-EBIO-enhanced I_sc presented was calculated by subtracting basal I_sc obtained in absence of DC-EBIO from that in the presence of DC-EBIO. The control and post-anion channel blocker tissue conductances were 12.1 ± 0.8 (control), 8.2 ± 0.6 (P < 0.001; NPPB), 11.7 ± 1.1 (NS; Glib), 11.2 ± 0.9 (NS; CFTR_inh-172), 10.9 ± 0.8 (NS; niflumic acid), and 8.7 ± 0.6 (P < 0.05; Glib/Niflu). The control and post-K⁺ channel blocker tissue conductances were 11.1 ± 0.7 (Control), 7.8 ± 0.6 (P < 0.001; Ba²⁺), 7.6 ± 0.6 (P < 0.001; CLT), 7.6 ± 0.8 (P < 0.001; TRAM-34), 10.4 ± 1.3 (NS; IbTX), 10.9 ± 0.9 (NS; APA), and 10.9 ± 1.0 (NS; chromanol 293B). DC-EBIO and inhibitors were added to both mucosal and serosal baths. Results presented represent means ± SE of 10 tissues. *P < 0.001, compared with respective control.

Although DC-EBIO significantly enhanced m-s and s-m fluxes, the net ⁸⁶Rb flux was not altered in normal distal colon (Fig. 1B). It is possible that the DC-EBIO-induced ⁸⁶Rb secretion mediated by mucosal Kcnn4 channels is masked by ⁸⁶Rb absorption via mucosal H-K-ATPase. Thus ⁸⁶Rb fluxes were measured after the addition of mucosal orthovanadate (VO₄; H-K-ATPase inhibitor). As shown in Fig. 3, under basal conditions, mucosal VO₄ completely inhibited the net ⁸⁶Rb absorption by reducing m-s fluxes, whereas, in contrast, it did not inhibit s-m fluxes. In the presence of mucosal VO₄, serosal and mucosal addition of DC-EBIO-induced ⁸⁶Rb secretion (Fig. 3). This DC-EBIO-induced ⁸⁶Rb secretion occurred as a result of enhanced s-m fluxes (Fig. 3). Thus to characterize mucosal Kcnn4 channels, all further studies were performed in the presence 1 mM mucosal VO₄.

Fig. 3.

Effect of DC-EBIO on ⁸⁶Rb fluxes in the presence of orthovanadate (VO₄) in normal rat distal colon. Basal ⁸⁶Rb fluxes (open bars) were measured as described in methods. ⁸⁶Rb flux was also measured after the addition of VO₄ (shaded bars) and VO₄ + DC-EBIO (solid bars); 1 mM VO₄ was present only in the mucosal bath, whereas DC-EBIO was present in both mucosal and serosal baths. The basal, post-VO₄, and post-VO₄ plus DC-EBIO tissue conductances were 7.2 ± 0.6, 7.3 ± 0.7 (NS), and 10.1 ± 0.6 (P < 0.004), respectively. Results presented represent means ± SE of 4 tissue pairs. *P < 0.001, compared with respective basal values.

The effect on I_sc of sequential addition of mucosal-followed-by-serosal and serosal-followed-by-mucosal DC-EBIO was examined to identify whether mucosal and/or serosal Kcnn4 channels provide the driving force for anion secretion. Addition of mucosal DC-EBIO alone significantly enhanced I_sc (Fig. 4, A and C). The mucosal DC-EBIO-enhanced I_sc was further enhanced by the subsequent addition of serosal DC-EBIO [i.e., DC-EBIO (m/s)] (Fig. 4, A and C). An almost similar rate of enhanced I_sc by mucosal DC-EBIO alone (ΔI_sc 2.3 μeq·h⁻¹·cm⁻²) and serosal followed by mucosal DC-EBIO [DC-EBIO (m) vs. DC-EBIO (m/s): ΔI_sc 1.9 μeq·h⁻¹·cm⁻²] suggests that the activation of both mucosal and serosal Kcnn4 channels contributes equally to optimal anion secretion (Fig. 4, A and C). In reverse order of DC-EBIO addition (serosal followed by mucosal), serosal DC-EBIO alone substantially enhanced I_sc (ΔI_sc 2.6 μeq·h⁻¹·cm⁻²) (Fig. 4, B and D). Subsequent mucosal addition of DC-EBIO in the continued presence of serosal DC-EBIO minimally (ΔI_sc 0.8 μeq·h⁻¹·cm⁻²) but significantly enhanced I_sc (Fig. 4, B and D). The low rate of I_sc enhanced by mucosal-followed-by-serosal DC-EBIO (ΔI_sc 0.8 μeq·h⁻¹·cm⁻²) compared with serosal-followed-by-mucosal DC-EBIO (ΔI_sc 1.9 μeq·h⁻¹·cm⁻²) suggests that serosal membranes are more permeable to DC-EBIO, thus partially activating mucosal Kcnn4 channels. Despite significantly enhancing I_sc, mucosal DC-EBIO did not alter the conductance [basal vs. mucosal DC-EBIO: 7.2 ± 0.6 vs. 7.3 ± 0.7 mS/cm²; not significant (NS)]. However, subsequent addition of serosal-followed-by-mucosal DC-EBIO significantly enhanced the conductance [basal vs. DC-EBIO (m/s): 7.2 ± 0.6 vs. 10.1 ± 0.6 mS/cm²; P < 0.004]. In contrast, serosal addition of DC-EBIO alone significantly enhanced the I_sc (basal vs. serosal DC-EBIO: 6.5 ± 0.3 vs. 9.6 ± 0.4 μeq·h⁻¹·cm⁻²; P < 0.002). Subsequent mucosal addition of DC-EBIO did not alter the conductance [DC-EBIO (s) vs. DC-EBIO (s/m): 9.6 ± 0.4 vs. 9.8 ± 0.5 mS/cm² (NS), respectively]. These observations suggest that both mucosal and serosal Kcnn4 channels may, in part, provide a driving force for agonist induced anion secretion.

Fig. 4.

Effect of sequential addition of mucosal-to-serosal and serosal-to-mucosal DC-EBIO on I_sc in normal rat distal colon. Time course of I_sc was measured as described in methods. A: time course of mucosal [DC-EBIO (m)] and subsequent serosal [DC-EBIO (s)] addition of DC-EBIO in the continued presence of mucosal DC-EBIO [DC-EBIO (m/s)]. B: basal and DC-EBIO-enhanced rate of I_sc were derived from A. The basal and post-DC-EBIO (m) and post-DC-EBIO (m/s) tissue conductances were 7.2 ± 0.3, 8.1 ± 0.2 (P < 0.02) and 10.0 ± 0.6 (P < 0.002), respectively. C: time course of serosal [DC-EBIO (s)] and subsequent mucosal [DC-EBIO (m)] addition of DC-EBIO in the continued presence of serosal DC-EBIO [DC-EBIO (s/m)]. D: basal and DC-EBIO-enhanced rate of I_sc were derived from B. The basal and post-DC-EBIO (s) and post-DC-EBIO (s/m) tissue conductances were 6.5 ± 0.3, 9.6 ± 0.4 (P < 0.002), and 9.8 ± 0.5 (NS), respectively. Results presented represent means ± SE of 10 tissues from 5 different rats. *P < 0.001 compared with basal; **P < 0.001 compared with DC-EBIO (m); ^£P < 0.05 compared with DC-EBIO (s).

It is also possible that DC-EBIO might permeate through mucosal membranes to activate serosal Kcnn4 channels, thus underestimating the contribution of mucosal Kcnn4 channels to anion secretion (I_sc). Therefore, the effect of serosal Ba²⁺ was examined on mucosal DC-EBIO-enhanced I_sc and ⁸⁶Rb secretion. Serosal Ba²⁺ did not inhibit mucosal DC-EBIO-enhanced I_sc (Fig. 5A). In contrast, serosal Ba²⁺ significantly enhanced ⁸⁶Rb secretion in normal rat distal colon (Fig. 5B). It is to be noted here that in theory blocking the K⁺ exits via serosal K⁺ channels would enhance the K⁺ secretion by 3–4 μeq·h⁻¹·cm⁻², since both Na-K-pump and Na-K-2Cl cotransporter actively bring K⁺ into cells. However, serosal Ba²⁺ only enhanced the K⁺ secretion by only 0.2 μeq·h⁻¹·cm⁻² in rat distal colon (Fig. 5A). It is likely that the rest of the K⁺ might exit through KCl cotransport localized on the serosal membranes of rat distal colon (27). These observations suggest that, in the absence of functional serosal Kcnn4 channels, enhanced K⁺ secretion via mucosal Kcnn4 channels compensates to achieve an optimal driving force for anion secretion.

Fig. 5.

Effect of serosal Ba²⁺ on mucosal DC-EBIO-enhanced I_sc and net ⁸⁶Rb fluxes in normal rat distal colon. Both basal I_sc and ⁸⁶Rb fluxes were measured as described in methods (open bars). I_sc and net ⁸⁶Rb fluxes were also measured in presence of mucosal DC-EBIO alone (shaded bars) and in the presence of mucosal DC-EBIO and 3 mM serosal Ba²⁺ (solid bars). Mucosal medium contained 1 mM VO₄. The basal, post-VO₄, and post-DC-EBIO (m)/Ba²⁺ (s) tissue conductances were 7.3 ± 0.5, 7.8 ± 0.6 (NS), and 8.1 ± 1.0 (NS), respectively. Results presented represent means ± SE of 6 tissue pairs. *P < 0.001 compared with respective basal values; ^£P < 0.05 compared with DC-EBIO (m).

To establish the relationship between mucosal anion channels and Kcnn4 channels, we examined the effects of anion channel blockers and K⁺ channel blockers on mucosal DC-EBIO-enhanced I_sc and ⁸⁶Rb secretion (Figs. 6 and 7). Mucosal NPPB, glibenclamide, and niflumic acid inhibited DC-EBIO-enhanced I_sc by 94, 38, and 27%, respectively (Fig. 6A). Simultaneous presence of glibenclamide and niflumic acid inhibited DC-EBIO enhance I_sc by 85% (Fig. 6A). Mucosal NPPB, glibenclamide, and niflumic acid also significantly inhibited the DC-EBIO-induced ⁸⁶Rb secretion (Fig. 6B). Simultaneous addition of both glibenclamide and niflumic acid inhibited the DC-EBIO-induced ⁸⁶Rb secretion similar to that of NPPB (Fig. 6B). Nonspecific K⁺ channel blocker Ba²⁺ and Kcnn4 channel-specific blockers CLT and TRAM-34 inhibited both the mucosal DC-EBIO-enhanced I_sc and DC-EBIO-induced ⁸⁶Rb secretion (Fig. 7, A and 7B). The observations that mucosal anion channel blockers partially inhibited the DC-EBIO-enhanced ⁸⁶Rb secretion indicate that part of the driving force for K⁺ secretion is provided by anion secretion and that the mucosal Kcnn4 channels and anion channels are coupled to regulate anion secretion.

Fig. 6.

Effect of anion channel blockers on mucosal DC-EBIO-enhanced I_sc and net ⁸⁶Rb fluxes in normal rat distal colon. Both basal I_sc (A) and net ⁸⁶Rb flux (B) were measured as described in methods. I_sc and net ⁸⁶Rb fluxes were also measured in presence of mucosal DC-EBIO (control) and mucosal DC-EBIO + mucosal inhibitors. The inhibitors used were 100 μM NPPB, 300 μM glibenclamide, 100 μM CFTR_inh-172, 100 μM niflumic acid, Glib/Niflu, and Glib/Inh-172/Niflu. The anion channel blockers did not significantly alter the control conductance (7.9 ± 0.7). Absolute DC-EBIO-enhanced I_sc and net ⁸⁶Rb fluxes presented were calculated by subtracting I_sc and ⁸⁶Rb fluxes obtained in the absence of DC-EBIO from that in presence of DC-EBIO, respectively. Mucosal medium always contained 1 mM VO₄, whereas serosal medium contained 3 mM Ba²⁺. Results presented represent means ± SE of 6 tissue pairs. *P < 0.001 compared with respective control; ^£P < 0.05 compared with respective control.

Fig. 7.

Effect of mucosal K⁺ channel blockers on mucosal DC-EBIO-enhanced I_sc and net ⁸⁶Rb fluxes in normal rat distal colon. Both basal I_sc (A) and net ⁸⁶Rb fluxes (B) were measured as described in methods. I_sc and net ⁸⁶Rb fluxes were also measured in presence of mucosal DC-EBIO (control) and mucosal DC-EBIO + inhibitors. The inhibitors used were 3 mM Ba²⁺, 50 μM clotrimazole, 50 μM TRAM-34, 100 nM iberiotoxin, and 10 nM apamin. Absolute DC-EBIO-enhanced rate of I_sc and net ⁸⁶Rb secretion presented were calculated by subtracting basal rate of I_sc and net ⁸⁶Rb fluxes obtained in the absence of DC-EBIO from that in the presence of DC-EBIO, respectively. Mucosal medium always contained 1 mM VO₄, whereas serosal medium contained 3 mM Ba²⁺. The K⁺ channel blockers did not significantly alter the control conductance (8.2 ± 0.6) (data not shown). Results presented represent means ± SE of 6 tissue pairs. *P < 0.001 compared with control.

In previous studies, we have shown downregulation of Kcnn4 specific mRNA abundance and Ca²⁺-activated K⁺ secretion as a mechanism for K⁺ conservation in dietary K-depleted rat proximal colon (16). Therefore, to establish that mucosal Kcnn4 channels both mediate K⁺ secretion and provide driving force for anion secretion, we also examined the effect mucosal DC-EBIO in dietary K-depleted rat distal colon. Under basal condition, net ⁸⁶Rb absorption in dietary K-depleted rat colon is significantly higher than that in normal rat colon (Fig. 8B). Mucosal VO₄ completely inhibited the net K⁺ absorption in both normal and dietary K-depleted rat distal colon (Fig. 8B). Similar to earlier observations, in the presence of mucosal VO₄ and serosal Ba²⁺, mucosal addition of DC-EBIO enhanced I_sc and induced ⁸⁶Rb secretion in normal colon. In contrast, neither enhanced I_sc nor ⁸⁶Rb secretion is present in K-depleted rat colon (Fig. 8, A and B). This observation establishes the conclusion that mucosal Kcnn4 channels provide driving force for anion secretion in normal rat distal colon.

Fig. 8.

Effect of mucosal DC-EBIO on I_sc and ⁸⁶Rb secretion in dietary K-depleted rat distal colon. Both basal I_sc (A) and net ⁸⁶Rb fluxes (B) were measured in normal and dietary K-depleted rat distal colon as described in methods. I_sc and net ⁸⁶Rb fluxes were also measured in the presence of mucosal VO₄ (1 mM) and mucosal VO₄ + DC-EBIO. Serosal medium always contained 3 mM Ba²⁺. The basal conductance for normal and dietary K-depleted animal was 7.6 ± 0.6 and 8.1 ± 0.6 (NS), respectively. The mucosal addition of VO₄ and DC-EBIO did not significantly alter the conductance either in normal and dietary K-depleted diet-fed rat colon (data not shown). Results presented represent means ± SE of 4 tissue pairs. *P < 0.001 compared with respective basal value; **P < 0.05 compared with respective basal value; ^£P < 0.001 compared with normal basal flux.

DISCUSSION

Although both inhibition of active Na⁺ absorption and stimulation of active anion (Cl⁻ and HCO₃⁻) secretion contribute to isotonic fluid loses, anion secretion mediated via activated anion channels (CFTR and CaCC) localized on mucosal membranes of intestine and colon is the major driving force for fluid secretion in diarrhea (3, 10). Intracellular levels of second messengers like cAMP and Ca²⁺ that activate anion channels are elevated during chronic inflammation in inflammatory bowel diseases such as ulcerative colitis and infectious diarrhea such as cholera (1, 28, 35). Agonist-induced anion secretion that depolarizes cells requires compensatory K⁺ efflux to maintain continuous anion secretion (3). Although large- (BK), intermediate- (Kcnn4), and small-conductance (SK) Ca²⁺-activated K⁺ channels and cAMP-activated K⁺ channels (KCNQ1) are expressed, only CLT-sensitive Kcnn4 channels have been shown to provide a driving force for agonist-induced anion secretion in both human and rat colon (13, 16, 25, 26, 34). Despite the functional demonstration and immunofluorescence localization of Kcnn4 channels on both mucosal and serosal membranes, only serosal Kcnn4 channels have been hypothesized to provide the driving force for agonist-induced Cl⁻ secretion (13, 34). Since this conclusion was arrived at on the basis of lipophilic CLT inhibition of Cl⁻ secretion, it is not known whether mucosal and/or serosal Kcnn4 channels drive agonist-induced anion secretion (26, 34). Thus the present study was designed to identify the role of mucosal Kcnn4 channels on anion secretion in rat distal colon. In this study, mucosal Kcnn4 channels were activated by use of DC-EBIO, whereas serosal K⁺ channels were blocked by use of the nonspecific K⁺ channel blocker Ba²⁺.

This study demonstrates that activation of K⁺ exit across both mucosal and serosal membranes, in part, provides a driving force for anion secretion mediated by both mucosal CFTR and CaCC anion channels (Fig. 1, A and B). Inhibition of anion secretion by the Kcnn4 channel blocker TRAM-34 established that Kcnn4 channels provided the driving force for DC-EBIO-enhanced anion secretion in normal rat distal colon (Figs. 2B and 7A). Partial inhibition of anion secretion by the individual addition of a mucosal CFTR-inhibitor (glibenclamide) and a CaCC-inhibitor (niflumic acid) invoked complete inhibition, which indicated that DC-EBIO activated K⁺ channels enhanced anion secretion as a result of activation of both CFTR and CaCC anion channels (Figs. 2A and 6A). Both CFTR and CaCC channels have been shown to be expressed in mucosal membranes of colonic epithelial cells (11, 24). DC-EBIO, which opens Kcnn4 channels, would have also stimulated CFTR Cl⁻ channels (14), whereas CaCC channels might have been activated by indirect cellular mechanisms. DC-EBIO and other benzimidazolone derivatives such as NS004 and 1-EBIO have been shown to activate glibenclamide-sensitive Cl⁻ currents in mucosal membranes of T84 cells and rat colon, whereas NS004 has also been shown to activate CFTR expressed in in vitro (6, 7, 14, 22, 29). Although both CFTR and CaCC anion channels are expressed on mucosal membranes of T84 cells and rat distal colon, only CFTR has been shown to be activated by benzimidazolone derivatives (7). Thus DC-EBIO activation of CaCC anion channels present in rat distal colon might have occurred as a result of enhanced intracellular free Ca²⁺ levels. The hyperpolarized cells generated by DC-EBIO activated Kcnn4-mediated K⁺ exit might have enhanced intracellular free Ca²⁺ levels by mobilizing intracellular stores or activating extracellular Ca²⁺ influx via mucosal Ca²⁺ channels. Ca²⁺ channels have been suggested to localize in mucosal membranes of rat distal colon (9). Although CaCC anion channels are expressed, Devor et al. (7) did not identify DC-EBIO-stimulated mucosal Cl⁻ conductance mediated via CaCC in T84 cells and rat distal colon, since this study utilized serosal membrane permeabilized monolayers and dexamethasone-treated rat distal colon, respectively.

This study hypothesizes that K⁺ exit mediated via both mucosal and serosal Kcnn4 channels, in part, contributes to the driving force for anion secretion. However, in the absence of serosal K⁺ exit, the enhanced mucosal K⁺ efflux compensates to provide the driving force for optimal anion secretion. The conclusion that mucosal K⁺ efflux provides a driving force for anion secretion is supported by the following observations: 1) Simultaneous addition of both mucosal and serosal DC-EBIO, but not addition of either mucosal or serosal DC-EBIO alone, enhanced the maximal rate of anion secretion (Fig. 4). 2) Cumulative rate of anion secretion is enhanced by the addition of mucosal DC-EBIO alone (55%), and subsequent serosal DC-EBIO addition (45%) (i.e., in the continued presence of mucosal DC-EBIO) is essentially equivalent to the maximal rate of anion secretion seen with DC-EBIO in both mucosal and serosal bath (Fig. 4, A and C). 3) Inhibition of serosal K⁺ exit enhanced mucosal DC-EBIO-induced K⁺ efflux (secretion), but serosal K⁺ channel inhibition did not affect DC-EBIO-enhanced anion secretion (Fig. 5, A and B). 4) Absence of mucosal DC-EBIO-induced K⁺ secretion failed to enhance anion secretion in dietary K-depleted rat distal colon (Fig. 8, A and B). The absence of DC-EBIO-induced K⁺ secretion is consistent with our recent report that downregulation of mucosal Kcnn4 channels is a mechanism of K⁺ conservation in the dietary K-depleted rat colon (16). The absence of mucosal Kcnn4 channels resulted in the inability of mucosal DC-EBIO to activate anion secretion in dietary K-depleted rat distal colon. This observation supports the conclusion that mucosal K⁺ efflux, in part, provides the driving force for anion secretion in normal rat distal colon.

In theory, since agonist-induced anion secretion is electrogenic, anion channels that mediate anion secretion and hyperpolarizing K⁺-exit channels should be expressed on mucosal and serosal membranes for a cellular model of anion secretion, respectively (3, 4). In contrast, it might have been presumed, if both anion channels and K⁺ channels are localized on the same membrane (i.e., mucosal membrane), that the agonist-induced anion secretion would be electroneutral. However, it may not be the case because the rate of K⁺ efflux is 8- to 10-fold lower than that of anion secretion (Figs. 5–8) and provides the driving force for anion secretion. The rate of serosal K⁺ exit has been reported to be less than 4% of the rate of Cl⁻ secretion in anion-secreting T84 cells (19). Since negligible amounts of K⁺ exit are required to maintain cell hyperpolarization to drive agonist-induced anion secretion, it is physiologically feasible for mucosal K⁺ efflux to subserve this role. In addition to driving anion secretion, mucosal K⁺ efflux also accounts for the substantially increased level of K⁺ (i.e., 5- to 10-fold) that prevails in diarrheal stools (31).

In summary, to our knowledge this is the first study to report that K⁺ efflux mediated via mucosal Kcnn4 channels provides, in part, the driving force for agonist-induced anion secretion. The Kcnn4 channels provide the driving force for anion secretion that is mediated via both CFTR and CaCC Cl⁻ channels. We conclude from these data that mucosal Kcnn4 channels are capable of driving agonist-induced anion secretion mediated via CFTR and CaCC and likely contribute to stool K⁺ losses that accompany diarrheal illnesses.

GRANTS

This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases Grant (DK-018777).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).