HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 289/3/C564    most recent 00561.2004v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (9) Citing Articles via Google Scholar Google Scholar Articles by Liao, T. Articles by Halm, D. R. Search for Related Content PubMed PubMed Citation Articles by Liao, T. Articles by Halm, D. R.

MEMBRANE TRANSPORTERS, ION CHANNELS, AND PUMPS

K⁺ channel K_VLQT1 located in the basolateral membrane of distal colonic epithelium is not essential for activating Cl^– secretion

¹Department of Neuroscience, Cell Biology and Physiology, Wright State University, Dayton, Ohio; and ²Division of Molecular Medicine, Harbor-UCLA Medical Center, David Geffen School of Medicine, University of California, Los Angeles, Torrance, California

Submitted 19 November 2004 ; accepted in final form 12 April 2005

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
The cellular mechanism for Cl^– and K⁺ secretion in the colonic epithelium requires K⁺ channels in the basolateral and apical membranes. Colonic mucosa from guinea pig and rat were fixed, sectioned, and then probed with antibodies to the K⁺ channel proteins K_VLQT1 (Kcnq1) and minK-related peptide 2 (MiRP2, Kcne3). Immunofluorescence labeling for Kcnq1 was most prominent in the lateral membrane of crypt cells in rat colon. The guinea pig distal colon had distinct lateral membrane immunoreactivity for Kcnq1 in crypt and surface cells. In addition, Kcne3, an auxiliary subunit for Kcnq1, was detected in the lateral membrane of crypt and surface cells in guinea pig distal colon. Transepithelial short-circuit current (I_sc) and transepithelial conductance (G_t) were measured for colonic mucosa during secretory activation by epinephrine (EPI), prostaglandin E₂ (PGE₂), and carbachol (CCh). HMR1556 (10 µM), an inhibitor of Kcnq1 channels (Gerlach U, Brendel J, Lang HJ, Paulus EF, Weidmann K, Brüggemann A, Busch A, Suessbrich H, Bleich M, and Greger R. J Med Chem 44: 3831–3837, 2001), partially (50%) inhibited Cl^– secretory I_sc and G_t activated by PGE₂ and CCh in rat colon with an IC₅₀ of 55 nM, but in guinea pig distal colon Cl^– secretory I_sc and G_t were unaltered. EPI-activated K⁺-secretory I_sc and G_t also were essentially unaltered by HMR1556 in both rat and guinea pig colon. Although immunofluorescence labeling with a Kcnq1 antibody supported the basolateral membrane presence in colonic epithelium of the guinea pig as well as the rat, the Kcnq1 K⁺ channel is not an essential component for producing Cl^– secretion. Other K⁺ channels present in the basolateral membrane presumably must also contribute directly to the K⁺ conductance necessary for K⁺ exit during activation of Cl^– secretion in the colonic mucosa.

HMR1556; K⁺ secretion; epinephrine; prostaglandin E₂; cholinergic

The colonic epithelium of mammals is composed of a relatively flat surface epithelium invaginated by numerous crypts of Lieberkühn (12). Within this epithelium, columnar cells and goblet cells are the predominant cell types, with a minor population of enteroendocrine cells. Goblet cells are distinguished from columnar cells by a large, dense cluster of apical mucous granules (59). Other cell types also are present in the colonic mucosa, including myoepithelial cells forming the pericryptal sheath, capillaries, and nerve fibers, as well as various types of leukocytes (8, 9, 48, 51). Previous studies support the concept that crypt columnar cells are major contributors to transepithelial ion secretion and mucus release (28, 29, 31, 32).

The two types of K⁺ channels proposed as the major components of basolateral membrane K⁺ conductance that support colonic Cl^– secretion, Kcnn4 and Kcnq1, have been observed in cells of isolated colonic crypts (24, 63). The intermediate conductance inwardly rectifying K⁺ channel, Kcnn4, could be the Ca²⁺-activated basolateral membrane K⁺ conductance supporting cholinergic stimulation (34, 63). In addition, a Ca²⁺-dependent inward rectifier that may be Kcnn4 has been observed in human colonic crypts (56). Association of Kcnq1 with minK-related peptide 2 (MiRP2, Kcne3) produces a cAMP-activated basolateral membrane K⁺ conductance (24, 63) that could support Cl^–-secretory activation by secretagogues such as vasoactive intestinal peptide or prostaglandin E₂ (PGE₂) (11, 27, 50), and both of these channel proteins have been localized to lateral membranes of mouse colonic crypts (14, 58). The involvement of Kcnq1/Kcne3 (K_VLQT1/MiRP2) K⁺ channels in colonic Cl^– secretion is supported further by the inhibition of cAMP-dependent secretion and channel activity by the chromanol 293B (24, 43, 58, 63). In contrast, even though 293B inhibited cAMP-dependent K⁺ currents in pancreatic acinar cells consistent with Kcnq1/Kcne1 (37, 62), 293B did not inhibit fluid secretion (38). However, the importance of cAMP as an intracellular signal that modulates secretory activity still makes the cellular location of Kcnq1 a useful indicator of possible secretory function.

Secretagogues operating through cAMP-dependent mechanisms produce two distinct modes of electrogenic ion secretion in the colonic epithelium (27, 30, 41). The more familiar mode involves Cl^– secretion that is also accompanied by electrogenic K⁺ secretion. This flow of ions into the lumen creates fluid buildup in crypt lumens, producing fluid flow that sweeps along mucus and other material (28), such that the term flushing secretion best summarizes the action of these secretagogues. The second secretory mode produces electrogenic K⁺ secretion, but without large, sustained Cl^– secretion. Modulatory secretion is a useful term to conceptualize this secretory function because fluid flow is low, but ion composition would be altered. Rat and guinea pig distal colon both produce these modes of secretion (53, 64), with differences in rates that may serve the specific physiology of an omnivore and a herbivore, respectively (52, 54). Thus a comparison of these two modes of secretion in these species can be used to demonstrate the varied roles of basolateral membrane K⁺ channels. In particular, increased basolateral membrane K⁺ channel activity would aid Cl^– secretion by enhancing the electrochemical driving force for conductive apical Cl^– exit, whereas decreased activity could increase K⁺ secretion by limiting basolateral exit of K⁺ into the interstitial space (40). The focus of this study was to examine the epithelial location of Kcnq1/Kcne3 K⁺ channels in the colon and to use the chromanol derivative HMR1556 (22) to determine the involvement of this K⁺ channel type in secretory activation by physiological secretagogues.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
Guinea pigs (Hartley, male, 400- to 650-g body wt) and rats (Sprague-Dawley, male or female, 125- to 250-g body wt) were administered standard chow and water ad libitum. In accordance with protocols approved by the Wright State University Laboratory Animal Care and Use Committee, guinea pigs and rats were killed by decapitation or by administration of an intraperitoneal overdose (>80 mg/kg) of pentobarbital sodium before perfusion fixation. The colon was removed, cut open along the mesenteric line, and flushed with saline solution to remove fecal pellets. The mucosa was separated from underlying submucosa and external muscle layers using a glass slide to gently scrape along the length of the colonic sheet, thus producing an isolated mucosal preparation (3, 9, 15). Because the plane of dissection occurred at the base of the crypts, the muscularis mucosae also were removed. Tissue samples were taken from the distal colon of the guinea pig (54), at distances of 5–20 cm (late) and 40 cm (early) from the peritoneal border. Colonic tissue samples from rat (12 cm total length) were taken from the proximal portion that had palm leaf mucosal folds and from the distal 1–5 cm measured from the peritoneal border (17, 19, 42).

Tissue fixation. Colonic tissues were fixed either by perfusion of fixative or after isolation of the mucosa. For perfusion-fixation, animals were perfused transcardially (20) with a vascular rinse solution (4°C), followed by 4% paraformaldehyde in phosphate buffer (PB). The colon was removed, cut into annuli, and postfixed with 4% paraformaldehyde in PB for 1 h. Perfusion-fixation did not produce satisfactory structural preservation in guinea pig colonic mucosa. Fixation also was accomplished by pinning isolated mucosal sheets in a Sylgard-coated dish for immersion in fixation solutions. Each mucosal specimen was fixed in PB containing 1% paraformaldehyde and 0.125% glutaraldehyde (15 min at room temperature). The mucosal specimens were fixed further in PB with 4% paraformaldehyde (20 min at room temperature). Chemicals used for the preparation of solutions were obtained from Sigma Chemical (St. Louis, MO). Vascular rinse solution contained (in mM) 161 Na⁺, 3.4 K⁺, 140 Cl^–, 6.0 HCO₃^–, 1.9 H₂PO₄^–, and 8.1 HPO₄^2–. PB contained (in mM) 181 Na⁺, 19 H₂PO₄^–, and 81 HPO₄^2–. Phosphate-buffered saline (PBS) contained (in mM) 168 Na⁺, 2.7 K⁺, 153 Cl^–, 1.9 H₂PO₄^–, and 8.1 HPO₄^2–. Tris-buffered saline (TBS) contained (in mM) 137 Na⁺, 155 Cl^–, and 20 Tris.

Immunolocalization. Mucosal tissues were prepared for immunofluorescence (2, 20) by dehydration in PB (4°C) with sucrose (15% wt/vol) and then frozen with optimal cutting temperature compound. Sections were cut (6 µm) on a cryostat and thaw mounted on gelatin-coated slides. Sections were permeabilized with PBST (PBS with 0.1% Triton X-100; 30 min), blocked in PBST with normal horse serum (10%, 1 h, room temperature), and then incubated (4°C) overnight with primary antibody in PBST. The following antibodies for K⁺ channel and auxiliary subunits were obtained from commercial suppliers (Chemicon International, Temecula, CA; Santa Cruz Biotechnology, Santa Cruz, CA; Jackson ImmunoResearch, West Grove PA): polyclonal anti-K_VLQT1 (6.5 ng/µl; Chemicon, COOH-terminal residues of human Kcnq1), two polyclonal anti-Kcne3 [4.0 ng/µl, Santa Cruz Biotechnology, internal domain (L-20) and NH₂-terminal residues (N-18) of human Kcne3], and polyclonal anti-metabotropic glutamate receptor (1.0 ng/µl, Chemicon; residues 1180–1191 of rat mGluR1-, Grm1). After being washed three times in PBS, sections were incubated in the dark with the appropriate secondary antibodies (Jackson ImmunoResearch) and donkey-anti-rabbit or donkey-anti-goat IgG antibody conjugated to fluorescein isothiocyanate (FITC; 15 ng/µl for 2 h at room temperature). Sections were washed and mounted in Vectashield (Vector Laboratories, Burlingame, CA). Absorption controls were preformed by preincubation of primary antibody with the antigenic peptide in PBS (60–90 min at room temperature) before addition to sections. Fluorescence was visualized using an Olympus BX60 epifluorescence microscope.

Detection of ion channel proteins also was accomplished using immunoblot analysis. Isolated colonic mucosa was disrupted by performing sonication in a buffered solution containing protease inhibitors (on ice). The isolation solution contained (in mM) 178 Na⁺, 1.5 Mg²⁺, 153 Cl^–, 50 HEPES, 10 EDTA, 10% glycerol, 1% Triton X-100, and 1.0 4-(2-aminoethyl)benzenesulfonyl fluoride, as well as (in µM) 1.54 aprotinin, 23.5 leupeptin, and 14.6 pepstatin A. Samples were centrifuged at 6,000 g (for 10 min at 4°C) followed by centrifugation of the resulting supernatant at 100,000 g (for 60 min at 4°C) to obtain a membrane sample; protein content was determined using the Bradford method (7). Proteins were electrophoresed by performing SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. These membranes were blocked with 10% nonfat dry milk in TBST (TBS with 0.1% Tween 20), followed by incubation with specific primary antibody and then with horseradish peroxidase-conjugated secondary antibody. Membranes were developed (90 s) with LumiGLO (Cell Signaling Technology, Beverly, MA) before film was exposed to detect the product.

Transepithelial current measurement. Isolated mucosal sheets were used for measurement of transepithelial current and conductance (30, 53). Four mucosal sheets from each animal were mounted in Ussing chambers (0.64-cm² aperture) and supported on the serosal face by Nuclepore filters (10 µm thick, 5-µm pore diameter; Whatman, Clifton NJ). Bathing solutions (10 ml) were circulated by gas lift through water-jacketed reservoirs (38°C). Standard Ringer solution contained (in mM) 145 Na⁺, 5.0 K⁺, 2.0 Ca²⁺, 1.2 Mg²⁺, 125 Cl^–, 25 HCO₃^–, 4.0 H_(3–X)PO₄^X–, and 10 D-glucose. Solutions were continually gassed with 95% O₂-5% CO₂, which maintained the solution at pH 7.4. Chambers were connected to automatic voltage clamps (Physiologic Instruments, San Diego, CA) that permitted compensation for solution resistance and continuous measurement of short-circuit current (I_sc). Transepithelial electrical potential difference was measured using paired calomel electrodes connected to the chambers by Ringer-agar bridges. Current was passed across the tissue through two Ag-AgCl electrodes connected by Ringer-agar bridges. I_sc was referred to as positive for flow across the epithelium from the mucosal to the serosal side. Transepithelial conductance (G_t) was calculated on the basis of currents produced by bipolar square voltage pulses imposed across the mucosa (±5 mV, 3-s duration, 1-min intervals).

PGE₂, indomethacin, and NS398 were obtained from Cayman Chemical (Ann Arbor, MI), and epinephrine (EPI) was purchased from Elkins-Sinn (Cherry Hill, NJ). K⁺ channel blockers HMR1556 {(3R,4S)-(+)-N-[3-hydroxy-2,2-dimethyl-6-(4,4,4-trifluorobutoxy)chroman-4-yl]-N-methylethanesulfonamide} and 293B [trans-6-cyano-4-(N-ethylsulfonyl-N-methylamino)-3-hydroxy-2,2-dimethylchromane] were provided by Dr. Uwe Gerlach (Aventis Pharma Deutschland, Frankfurt-am-Main, Germany). All other chemicals were obtained from Sigma Chemical. Drugs were added in small volumes from concentrated stock solutions. PGE₂ was prepared in an ethanol stock solution that added 0.03% ethanol at 3 µM PGE₂. Stock solutions of HMR1556 (10 mM) and 293B (100 mM) were made with DMSO. Additions of 1% ethanol or DMSO alone did not alter transepithelial measures of K⁺ or Cl^– secretion (30).

Inhibitor-sensitive components of I_sc and G_t were calculated using the paired responses of adjacent mucosal tissues. Stripchart recordings of I_sc were digitized at 10-s intervals to examine secretory onset. Concentration responses of I_sc to inhibitors were fit to Henri-Michaelis-Menten binding curves using a nonlinear least-squares procedure (30). Results are reported as means ± SE. Statistical comparisons were performed using a two-tailed Student’s t-test for paired responses, with statistically significant differences accepted at P < 0.05.

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
Localization of K⁺ channel subunits Kcnq1 and Kcne3. Immunoreactivity for the K⁺ channel protein K_VLQT1, Kcnq1, was detected in a location consistent with the plasma membrane of colonic epithelial cells (Figs. 1 and 2) in accordance with previous reports in which immunolocalization (14, 62) and patch-clamp recording of channel activity were used (24, 58). Similar to mouse colon (14, 62), the rat colon (Fig. 1A) had prominent labeling in the lateral membrane of crypt epithelial cells, but surface epithelial cells did not exhibit clear labeling in the perfusion-fixed specimens. Rat mucosa fixed after isolation had lower background in surface epithelial cells than in perfusion-fixed specimens, such that faint but distinct labeling of lateral membranes was apparent in surface cells of rat proximal and distal colon (Fig. 1, E–G). The luminal margins of either crypt or surface epithelial cells were not labeled (Fig. 1), supporting an absence of Kcnq1 from the apical membranes of these epithelial cells. Similar results were obtained for the proximal colon (Fig. 1, D and E) and distal colon (Fig. 1, C, F, and G).

View larger version (95K): [in this window] [in a new window]   Fig. 1. Localization of K+ channel KVLQT1 (Kcnq1) immunoreactivity in rat colonic mucosa. Kcnq1 was detected using immunofluorescence in rat colonic mucosa, fixed either by perfusion or after isolation (anti-KVLQT1; Chemicon). A: crypt (C) and surface (S) epithelia are shown from a perfusion-fixed specimen of distal colon. Lateral membranes of crypt cells labeled distinctly. B: in perfusion-fixed distal colon with secondary antibody alone, nonspecific labeling and autofluorescence of epithelial cells was low, but mucosal leukocytes (arrow) showed nonspecific labeling. C: crypt from a perfusion-fixed distal colon specimen with a longitudinal profile of the lumen (L) showed a lack of apical membrane labeling, together with lateral membrane labeling. Some goblet cells (G) were apparent by the rounded cellular profiles with dark, round apical poles. Basal cell poles suggested labeling for only some cells. D: longitudinal profile of a crypt from a proximal colon specimen fixed as an isolated mucosa showed lateral membrane labeling and a lack of apical membrane labeling. Some goblet cells (G) were apparent by the dark round apical poles. Surface epithelium in proximal (E) and distal (F and G) colon specimens fixed as isolated mucosa showed faint but distinct labeling of lateral membranes without any indication of apical (a) or basal (b) membrane labeling. Scale bars, 25 µm for A and B; 10 µm for C and D; and 5 µm for E–G.

View larger version (116K): [in this window] [in a new window]   Fig. 2. Localization of Kcnq1 immunoreactivity in guinea pig distal colonic mucosa. Kcnq1 was detected using immunofluorescence in guinea pig distal colonic mucosa fixed after isolation (anti-KVLQT1; Chemicon). A: lateral membranes of crypt and surface cells showed distinct labeling. When the secondary antibody alone was applied, nonspecific labeling of epithelial cells was low (data not shown). B: crypt with a longitudinal profile of the lumen (L) showed the lack of apical membrane labeling (arrow), together with lateral membrane labeling. Some goblet cells (G) were apparent because of the rounded cellular profiles with dark apical poles. Basal membrane labeling was not distinctly apparent for columnar cells. Epithelial profiles of surface epithelium (C and D) showed labeling extending along the length of the lateral margin of columnar cells without any labeling in the apical membrane (a) and only weak labeling if any indications of labeling were observed in the basal membrane (b) region. Scale bars, 10 µm for A and B; 5 µm for C and D.

The use of the secondary antibody alone eliminated all membrane labeling of epithelial cells observed in rat (Fig. 1B) and guinea pig colon (data not shown), indicating that the Kcnq1 antibody was necessary for the observed labeling. The antigenic peptide was not available for preabsorption of the Kcnq1 antibody as a further control for nonspecific reactions. An antibody against the metabotropic glutamate receptor was used as an additional control for nonspecific labeling of membrane proteins, but no mucosal labeling was detected with this antibody (data not shown). In perfusion-fixed specimens, the surface epithelium had a diffuse, low-level labeling (Fig. 1A), predominantly in the cytoplasm (darker nuclei), which also was evident in the absence of the primary antibody (Fig. 1B). The bright cells in the interstitium were probably leukocytes as reported previously, with visibility likely due to autofluorescence of granule contents and nonspecific binding of the secondary antibody (8, 48).

Immunoreactivity for the K⁺ channel regulatory protein MiRP2, Kcne3, was detected in the lateral membrane of guinea pig colonic crypt epithelia (Fig. 3A), consistent with previous reports for mouse colon using immunolocalization and patch-clamp recording of channel activity (14, 58). Lateral membranes of surface cells also labeled for Kcne3 (Fig. 3, C and D). All of the membrane labeling was eliminated by preabsorption of the primary antibody with the antigenic peptide (Fig. 3B). The labeling in crypt and surface epithelial cells occurred with a beaded appearance, suggesting that Kcne3 clustered at sites along the lateral membrane. Neither apical nor basal membranes in surface cells showed detectable labeling.

View larger version (124K): [in this window] [in a new window]   Fig. 3. Localization of minK-related peptide 2 (MiRP2, Kcne3) immunoreactivity in guinea pig distal colonic mucosa. Kcne3 was detected using immunofluorescence in guinea pig distal colonic mucosa fixed after isolation (anti-Kcne3-L20; Santa Cruz Biotechnology). A: lateral membranes of crypt cells showed distinct labeling. B: when the primary antibody was preabsorbed with the antigenic peptide, nonspecific labeling of crypt and surface epithelial cells was low. Epithelial profiles of surface epithelium (C and D) showed labeling extending along the length of the lateral margin of columnar cells without any labeling in the apical membrane (a) and only weak if any indications of labeling in the basal membrane (b) region. Labeling of mucosa by a distinct antibody for Kcne3 (anti-Kcne3-N18) had a similar appearance (data not shown). Scale bars, 10 µm for A and B; 5 µm for C and D.

View larger version (58K): [in this window] [in a new window]   Fig. 4. K+ channel immunoblots. Protein isolated from colonic mucosa of guinea pig and rat were immunoblotted with antibodies against the K+ channel proteins (A) Kcnq1 and (B) Kcne3 (anti-Kcne3-L20). Arrowheads indicate bands of the size expected for these proteins (Kcnq1, 74 kDa; Kcne3, 14 kDa). Preabsorption of the anti-Kcne3-L20 antibody with antigenic peptide diminished the distinct band at 15 kDa for Kcne3, whereas the other, fainter bands represent nonspecific interactions of the secondary antibody (0.004 ng/µl donkey anti-goat IgG antibody; Jackson ImmunoResearch).

Distinct secretory states were produced after attaining the basal condition by adding specific secretagogues (30, 53). Sustained electrogenic K⁺ secretion of the modulatory type was stimulated by addition of either EPI (5 µM) or PGE₂ at low concentration (5 nM). Addition of PGE₂ at high concentration (3 µM) stimulated flushing-type secretion consisting of sustained Cl^– secretion together with K⁺ secretion. Adding carbachol (CCh; 10 µM) cumulatively with PGE₂ (3 µM) produced a further large synergistic increase in Cl^– secretion with a transient component lasting 10–20 min. Each of these distinct secretory modes was examined for sensitivity to Kcnq1 inhibitors, including 1) modulatory-type K⁺ secretion, 2) flushing-type Cl^– and K⁺ secretion, and 3) synergistic Cl^– secretion.

Inhibition of secretory modes. The chromanol 293B has been shown to inhibit the Kcnq1 K⁺ channel as well as Cl^– secretion (21, 43, 45, 62, 63). A higher-affinity inhibition of Kcnq1 is obtained with the 293B derivative HMR1556 (22); from different cell types, the IC₅₀ for HMR1556 ranges from 7 to 170 nM (5, 23, 37, 39, 61). In rat colonic mucosa (Fig. 5), HMR1556 (10 µM) inhibited a portion of the flushing-type Cl^–-secretory I_sc and G_t stimulated by PGE₂ but did not alter the modulatory-type K⁺-secretory I_sc and G_t stimulated by EPI. Inhibition was rapid for both I_sc and G_t. The HMR1556-sensitive component of the PGE₂ response reached a maximum within 10 min and remained stable with only a slight decline over 15 min (Fig. 5C). The HMR1556-sensitive and HMR1556-resistant components of the PGE₂-secretory response were similar in size.

View larger version (25K): [in this window] [in a new window]   Fig. 5. HMR1556 sensitivity of modulatory and flushing secretion in rat colon. Rat distal colonic mucosae were stimulated cumulatively by epinephrine (EPI; 5 µM) and prostaglandin E2 (PGE2; 3 µM), from the standard basal condition. The basal condition was produced by three successive bath replacements, indomethacin (2 µM) and NS398 (2 µM) in both bathing solutions and amiloride (100 µM) in the mucosal bathing solution. Short-circuit current (Isc; A) and transepithelial conductance (Gt; B) are shown. Gt changes shown (Gt) had the prestimulation value subtracted (, 9.5 mS/cm2; , 11.2 mS/cm2). HMR1556 (10 µM) was added to the serosal bath (asterisk) for an adjacent pair of mucosae during stimulation by either EPI () or PGE2 (). Differences within the pair for Isc and Gt (C and D) revealed the HMR1556-sensitive components (shaded region) of EPI and PGE2 responses. The dashed line connects periods of identical treatment conditions for the tissue pair. The rate of change for Gt in the secretory state was different between the mucosae in this pair, but an abrupt change occurred with HMR1556 addition during PGE2 stimulation.

View larger version (32K): [in this window] [in a new window]   Fig. 6. HMR1556 sensitivity of modulatory and flushing secretion in guinea pig colon. Guinea pig mucosae from late distal colon were stimulated cumulatively by EPI (5 µM) and PGE2 (3 µM) from the standard basal condition as shown in Fig. 5. Isc (A) and Gt (B) are shown. Gt changes shown (Gt) had the prestimulation value subtracted (, 12.7 mS/cm2; , 11.2 mS/cm2). HMR1556 (10 µM) was added to the serosal bath (asterisk) for an adjacent pair of mucosae during stimulation by either EPI () or PGE2 (). Abrupt changes with HMR1556 were not apparent, and paired HMR1556 responses during EPI (Isc = 4.9 ± 0.8 µA/cm2, Gt = 0.20 ± 0.24 mS/cm2; n = 7) or PGE2 (Isc = 0.9 ± 4.0 µA/cm2, Gt = –0.18 ± 0.17 mS/cm2; n = 7) were not significantly different from zero (P < 0.05), except for the small (3%) increase in Isc with EPI.

The secretory responses in the rat were dependent on the position along the colon (Fig. 7), with a larger EPI-stimulated, modulatory K⁺ secretion at more distal sites and a larger PGE₂-stimulated, flushing Cl^– secretion at more proximal sites. The I_sc and G_t in basal and PGE₂ conditions were similar to earlier results found using mucosal preparations of rat distal colon (3, 15). In addition, G_t tended to be larger at proximal sites, similar to findings in earlier studies (33, 57). The PGE₂-stimulated Cl^–-secretory response was calculated as the difference between the I_sc after PGE₂ addition and the prior EPI-stimulated I_sc to include the full range of secretory capacity (Fig. 7B). The HMR1556-sensitive and HMR1556-resistant components of this PGE₂-stimulated Cl^–-secretory response in rat colon were not different in size at any position examined (Fig. 7, B and C). The IC₅₀ was 55 ± 11 nM for this HMR1556 inhibition of the PGE₂-stimulated Cl^–-secretory response (Fig. 8). For the guinea pig distal colon, the EPI-stimulated, modulatory I_sc, and PGE₂-stimulated flushing I_sc were approximately –100 µA/cm² and +120 µA/cm², respectively, in both early and late portions, and the lack of inhibition by HMR1556 also was similar along the length of the distal colon.

View larger version (33K): [in this window] [in a new window]   Fig. 7. HMR1556 sensitivity along the proximal-distal axis in rat colon. Rat colonic mucosae were stimulated as in Fig. 5, with HMR1556 added after PGE2-stimulation. Isc and Gt values (means ± SE; N) are shown (A and C) for positions measured orad from the peritoneal border (, basal; , amiloride; , EPI; , PGE2; , HMR1556); the error bars for position include the chamber aperture (0.9 cm) and an estimate of uncertainty for the position of the mucosal specimen. Paired differences between successive conditions also are shown (B and C), with the HMR1556-resistant component included (). Asterisks (B) mark Isc values for amiloride and EPI additions that are significantly different from zero (P < 0.05); all other Isc are significantly different from zero. The amiloride-sensitive Isc value at 1.0 cm was significantly different from the value at 2.5 cm (P < 0.05), and the EPI-stimulated Isc values at the three distal positions were significantly different from each other (P < 0.05). The PGE2-stimulated Isc value at 5.5 cm () was significantly different from the value at 2.5 cm (P < 0.05). Gt values (C) in basal, amiloride, and EPI conditions were statistically identical (P < 0.05), so only the amiloride values are shown; all Gt shown were significantly different from zero (P < 0.05). The three recognized morphologic regions along the rat colon are proximal, major flexural, and distal, with the proximal region having distinct palm leaf mucosal folds (42). The transition from proximal to major flexure occurs at 7.5 cm, and the transition from major flexure to distal occurs at 4.5 cm. The end of the colon and the beginning of the rectum are defined anatomically (42) to occur at the peritoneal border (0.0 cm), whereas functionally this transition to rectal colon occurs 1 cm orad from the peritoneal border (17, 19). The rat colon also has been divided into ascending and descending portions (18) at 7 cm.

View larger version (16K): [in this window] [in a new window]   Fig. 8. Concentration-dependent inhibition of flushing secretion by HMR1556. Rat distal colonic mucosae were stimulated as in Fig. 5, followed by cumulative additions of HMR1556 to the PGE2-stimulated state. The HMR1556-resistant proportion of Isc is shown at four concentrations (n = 5); the resistant proportion was calculated as (IHMR – IEPI)/(IPGE2 – IEPI). This concentration dependence was not distinctly different along the proximal to distal axis, so results from all positions were included. The IC50 was 55 ± 11 nM.

View larger version (24K): [in this window] [in a new window]   Fig. 9. HMR1556 sensitivity of cholinergic response in rat colon. Rat distal colonic mucosae were stimulated as in Fig. 5, followed by carbachol (CCh; 10 µM) addition (A and B). Gt changes (Gt) shown had the prestimulation value subtracted (, 7.3 mS/cm2; , 9.1 mS/cm2). HMR1556 (10 µM) was added to the serosal bath (asterisk) for an adjacent pair of mucosae during stimulation by either PGE2 () or CCh (). Differences within the pair for Isc and Gt (C and D) revealed the HMR1556-sensitive components (shaded region) of PGE2 and CCh responses. The dashed line connects periods of identical treatment conditions for the tissue pair.

View larger version (19K): [in this window] [in a new window]   Fig. 10. Cholinergic stimulation of electrogenic K+ secretion in guinea pig colon. Guinea pig mucosae from late distal colon were stimulated with CCh (10 µM), from the standard basal condition as shown in Fig. 5. Isc (A) and Gt (B) from a representative mucosa are shown. Gt changes (Gt) shown had the prestimulation value subtracted (8.2 mS/cm2). Bumetanide (100 µM) was added to the serosal bath. Average CCh-stimulated Isc was –47.8 ± 3.9 µA/cm2, and mean Gt was 3.20 ± 0.23 mS/cm2 (n = 15).

View larger version (34K): [in this window] [in a new window]   Fig. 11. HMR1556 sensitivity of the cholinergic modulatory and synergistic responses in guinea pig colon. Guinea pig mucosae from late distal colon (A and B) were stimulated cumulatively by CCh (10 µM) and PGE2 (3 µM) from the standard basal condition as shown in Fig. 5. Gt changes (Gt) shown had the prestimulation value subtracted (, 8.6 mS/cm2; , 9.8 mS/cm2). HMR1556 (10 µM) was added to the serosal bath (asterisk) for an adjacent pair of mucosae during stimulation by either CCh () or PGE2 (). Subsequently, HMR1556 concentration was increased to 30 µM () for one mucosa of the pair (). Abrupt changes with HMR1556 were not apparent, and paired HMR1556 responses during CCh (Isc = –0.7 ± 3.2 µA/cm2, Gt = –0.14 ± 0.38 mS/cm2; n = 4) or CCh/PGE2 (Isc = 6.0 ± 9.3 µA/cm2, Gt = –0.61 ± 0.89 mS/cm2; n = 4) were not significantly different from zero (P < 0.05).

View larger version (36K): [in this window] [in a new window]   Fig. 12. HMR1556 sensitivity of flushing and synergistic activation in rat colon. Rat distal colonic mucosae were stimulated (as in Fig. 5) by EPI (5 µM), followed by PGE2 (3 µM) at time 0 (A and B). Average PGE2-stimulated Isc (A) and Gt (B) are shown (n = 5) for control stimulation (), with HMR1556 (10 µM) () and for the paired difference values between these conditions (). SE values were calculated from traces normalized to the maximal response for indication of time course variability; comparisons of maximal responses are in Fig. 7. The half-times for Isc activation (n = 5) were 40.4 ± 11.2 s for control, 6.3 ± 2.3 s for the HMR-resistant component, and 80.5 ± 15.6 s for the HMR-sensitive component. Paired mucosae (n = 6) also were stimulated (as shown in Fig. 9) with CCh (10 µM) added at time 0 (C and D), including control stimulation (), with HMR1556 (10 µM) () and the difference values between these conditions (). The starting condition for C and D was the same as the final condition shown in A and B. SE values were calculated without normalization. All of the HMR-sensitive Isc (C) at times >1.4 min were significantly different from the value before CCh stimulation. Asterisk in D indicates a time point at which HMR1556-sensitive Gt was significantly different from the value before CCh stimulation. The minimum HMR-sensitive Isc values at 2 min of activation were significantly different from zero, as were the maximal values at 8 min; the maximal HMR-sensitive Gt at 8 min of activation also was significantly different from zero.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION GRANTS REFERENCES

Voltage-sensitive K⁺ channels (K_V) are not obvious choices as components of epithelial transport mechanisms, because changes in membrane electrical potential differences are relatively modest compared with other cell types, such as neurons and muscle cells. However, in addition to voltage-dependent gating, other signaling cascades regulate many of the K_V channels. In particular, K_VLQT1 (Kcnq1) is activated by cAMP-dependent mechanisms (36, 55, 58, 65), possibly acting through protein kinase A, and has been shown by several means to contribute to the basolateral membrane K⁺ conductance necessary for electrogenic Cl^– secretion. The presence in the colon of the mRNA encoding Kcnq1 was shown using Northern blot analysis (14, 65), and in situ hybridization confirmed that colonic crypt epithelial cells contained this message (58). With the use of rat crypt epithelial cells, previous investigators obtained a full-length cDNA (669 amino acids) of Kcnq1 that was 90% identical to the amino acid sequence for human Kcnq1 (36). The involvement of Kcnq1 in Cl^– secretion was demonstrated using the chromanol 293B that inhibits both Kcnq1-dependent currents (6, 36, 58), as well as a current component in rat colonic crypt cells and transepithelial electrogenic Cl^– secretion in rat and mouse colon (36, 43, 45, 58, 62).

Immunofluorescence labeling of mouse colon for Kcnq1 supports a presence of this K⁺ channel in the lateral membrane of crypt epithelial cells (14, 62). Similarly, guinea pig and rat colon showed lateral membrane localization of Kcnq1 immunoreactivity (Figs. 1 and 2). Although nonspecific staining obscured detection of Kcnq1 in surface epithelial cells of perfusion-fixed rat colon, the low background staining in isolation-fixed guinea pig and rat mucosa allowed a clear determination that Kcnq1 also was localized to the lateral membrane of surface epithelial cells (Figs. 1 and 2). Further support for the presence of Kcnq1 in surface epithelial cells was indicated using RT-PCR to detect Kcnq1 mRNA in rat colonic surface cells (36). Similar to results in other studies of colonic mucosa (14, 36, 58), Kcnq1 was present together with MiRP2 (Kcne3) (Fig. 3), an auxiliary subunit that modifies gating kinetics and inhibitor sensitivity (6, 36, 49, 55, 58). Of course the mere localization of both Kcnq1 and Kcne3 to the same membrane using immunofluorescence does not prove a functional connection. However, 293B-sensitive currents have been measured in rat colonic crypt cells with properties similar to defined Kcnq1/Kcne3 currents (36, 63), such that these two K⁺ channel subunits likely combine to form a component of lateral membrane K⁺ conductance. Also, the low IC₅₀ for HMR1556 (Fig. 8) supports the involvement of Kcnq1/Kcne3 rather than Kcnq1 alone (39). The additional finding of both Kcnq1 and Kcne3 in lateral membranes of colonic surface epithelial cells (Figs. 2 and 3) suggests that these two subunits also combine at this location to produce the weakly voltage-dependent K⁺ channel characteristic of Kcnq1/Kcne3. Thus the presence of this K⁺ channel type does not distinguish surface cells from crypt cells, but perhaps instead a distinction between these cell types occurs at the level at which signaling pathways activate these channels to augment K⁺ flow.

Ion transport characteristics vary along the length of the colon with amiloride-sensitive Na⁺ absorption occurring predominantly at distal sites (18, 52, 54). A similar gradient of amiloride-sensitive I_sc was detected in this study (Fig. 7B). The response to physiological secretagogues also was examined and showed a gradient along the length of the rat colon. The I_sc stimulated by the flushing secretagogue PGE₂ was 40% higher at proximal positions compared with more distal positions. For stimulation by the modulatory secretagogue EPI, a sustained I_sc was apparent only at distal positions. These results are consistent with an earlier study in which the investigators used a mucosal-submucosal preparation (33), except that the basal I_sc was generally higher (by 20–50 µA/cm²) than in the present study (Fig. 7A), suggesting a difference in secretory status.

The modulatory mode of secretion, characterized by sustained electrogenic K⁺ secretion without sustained Cl^– secretion, may be considered the most fundamental secretory mode in the distal colon. This concept is supported by experiments in which colonic mucosa were stimulated after the secretory influences from nerves and endogenous production of paracrine factors were first reduced. Stimulation from a quiescent basal state by -adrenergic (27, 53, 64) (Figs. 5 and 6), prostanoid EP2 subtype (30, 53), and cholinergic agonists (10, 46) (Figs. 10 and 11) all result in modulatory K⁺ secretion. In addition, stimulation with a low concentration of forskolin, which activates adenylyl cyclase to produce cAMP, also leads to modulatory secretion (41), consistent with the action of -adrenergic and EP2 prostanoid receptors to increase cellular cAMP. The cholinergic activation of modulatory secretion from a quiescent state (Fig. 10) indicates that other second messengers also are capable of activating this secretory mode and indicates that a distinct secretory state exists compared with the cholinergically induced reduction in flushing-mode K⁺ secretion of rabbit distal colon (16).

Addition of the lipid-soluble 293B or HMR1556 could block either apical or basolateral K⁺ channels, but the localization of Kcnq1 to the basolateral membrane (Figs. 1 and 2) suggests that these inhibitors would act to enhance modulatory K⁺ secretion by diverting K⁺ exit to the apical membrane. Because neither HMR1556 nor 293B altered the modulatory response (Figs. 5, 6, and 11) similarly to results with 293B in human and cystic fibrosis mouse colon (45, 47), Kcnq1 appears to be largely inactive during the modulatory mode of secretion.

Flushing secretion is driven by electrogenic Cl^– secretion, together with an accompanying electrogenic K⁺ secretion, and is elicited by several types of secretagogues. In particular, PGE₂ is produced within the mucosa and at high concentration activates the flushing mode via receptors distinct from the EP prostanoid type (30). The regulatory mechanism stimulating flushing secretion likely involves cAMP because at high concentration forskolin produces large Cl^–-secretory I_sc (11, 50). In the guinea pig distal colon, increasing forskolin concentration reverses a negative I_sc to a positive I_sc, which is indicative of conversion from modulatory secretion to flushing secretion (41). The flushing-type Cl^–-secretory I_sc stimulated in colonic mucosa from human (46), mouse (45, 62), rabbit (43), and rat (36, 64) by either forskolin or inhibitors of phosphodiesterase was inhibited from 60 to 90% with the chromanol 293B. In rabbit distal colon, flushing secretion produced by the secretagogues PGE₂, adenosine, and vasoactive intestinal peptide was inhibited 70–80% by 293B (43).

A difficulty with assigning a quantitatively specific role for Kcnq1 on the basis of 293B inhibition is that 293B also inhibits the Cl^– channel CFTR with an IC₅₀ of 20–30 µM (4). Because CFTR is a component of the apical Cl^– conductance needed for Cl^– secretion (50), the potency of 293B may result from action at both secretory K⁺ and Cl^– conductance. The chromanol derivative HMR1556 inhibits Kcnq1 with 100-fold higher affinity than 293B (22, 23), such that any similar nonspecificity would not be encountered at concentrations sufficient to inhibit Kcnq1. In addition, the dose-response curve of Cl^–-secretory I_sc in rat colon (Fig. 8) did not include an inflection that would be consistent with such a high concentration inhibitory effect on CFTR. Thus the 50% inhibition of flushing-type secretion by HMR1556 (Figs. 5 and 7) in rat colon strongly supports a limited requirement for Kcnq1 K⁺ channels and the need for at least one other K⁺ channel type. The lack of inhibition by HMR1556 in guinea pig distal colon (Fig. 6) was not due to the insensitivity of guinea pig Kcnq1, because HMR1556 blocks Kcnq1 currents in guinea pig cardiomyocytes (5, 23). This failure further indicates that flushing secretion could occur without the involvement of Kcnq1 as part of the basolateral membrane K⁺ conductance.

The activation time course for flushing secretion in rat colon (Fig. 12, A and B) supports the concept that Kcnq1 was needed to produce the slower-onset, secondary phase of secretory capacity. In contrast, the flushing response in guinea pig colon (Fig. 6) had a rapid onset resembling the HMR-resistant component observed in rat colon. Because Kcnq1 was present in the epithelial cells of guinea pig colonic mucosa, the signaling elicited by PGE₂ in these cells apparently lacked regulatory pathways to produce this secondary phase of flushing secretion. Another major difference with rat colon is the higher relative rate of K⁺ secretion during PGE₂ stimulation in the guinea pig colon (53), such that apical membrane K⁺ channels may satisfy the requirements for additional K⁺ conductance needed to support high rates of Cl^– secretion.

The present results using the Kcnq1/Kcne3 inhibitor HMR1556 support the concept that distinct K⁺ channels are needed to produce the secretory modes activated by different types of secretagogues (24). However, the results also indicate that a single type of K⁺ channel would not be sufficient to produce Cl^– secretion via the ubiquitous flushing secretagogue PGE₂. Other studies have indicated that the possible involvement of Kcnq1 may not be apparent until additional K⁺ channel types are inhibited (44). However, those observations still suggest that Kcnq1 apparently is not the preferred K⁺ channel for activation by those secretagogues. This concept that the secretory cells have a reserve capacity for the activation of K⁺ channels able to support secretion is underscored by the large Cl^–-secretory currents (300–500 µA/cm²) that are possible without the need for Kcnq1 (Figs. 11 and 12). The presence of Kcnq1 in colonic epithelial cells could serve requirements for cell volume regulation (25) as well as secretory needs. The early oscillatory behavior of the HMR-resistant response (Fig. 12A) may represent a volume instability of these secretory cells in attempting to initiate the Kcnq1-dependent secondary phase of secretion. Thus the choices that epithelial cells make with regard to which type of K⁺ channel to activate during secretion may depend on advantages conveyed by the specific activation and kinetic details of each channel type.

	GRANTS

TOP ABSTRACT METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
This study was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-39007 and DK-65845 and the Wright State University Research Challenge Program.

	FOOTNOTES

 

Address for reprint requests and other correspondence: D. R. Halm, Dept. of Neuroscience, Cell Biology and Physiology, Wright State Univ., 3640 Colonel Glenn Hwy., Dayton, OH 45435 (e-mail: dan.halm{at}wright.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
1. Abbott GW and Goldstein SAN. K⁺ channel subunits encoded by the KCNE gene family: physiology and pathophysiology of the MinK-related peptides (MiRPs). Mol Interv 1: 95–107, 2001.[Abstract/Free Full Text]

2. Allan VJ (ed.). Protein Localization by Fluorescence Microscopy: A Practical Approach. Oxford, UK: Oxford University, 2000.

3. Andres H, Rock R, Bridges RJ, Rummel W, and Schreiner J. Submucosal plexus and electrolyte transport across rat colonic mucosa. J Physiol 364: 301–312, 1985.[Abstract/Free Full Text]

4. Bachmann A, Quast U, and Russ U. Chromanol 293B, a blocker of the slow delayed rectifier K⁺ current (I_Ks), inhibits the CFTR Cl^– current. Naunyn Schmiedebergs Arch Pharmacol 363: 590–596, 2001.[CrossRef][ISI][Medline]

5. Bosch RF, Schneck AC, Csillag S, Eigenberger B, Gerlach U, Brendel J, Lang HJ, Mewis C, Gögelein H, Seipel L, and Kühlkamp V. Effects of the chromanol HMR 1556 on potassium currents in atrial myocytes. Naunyn Schmiedebergs Arch Pharmacol 367: 281–288, 2003.[CrossRef][ISI][Medline]

6. Boucherot A, Shreiber R, and Kunzelmann K. Regulation and properties of KCNQ1 (K_VLQT1) and impact of the cystic fibrosis transmembrane conductance regulator. J Membr Biol 182: 39–47, 2001.[CrossRef][ISI][Medline]

7. Bradford M. A rapid and sensitive method for the quantification of microgram quantities of proteins utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254, 1976.[CrossRef][ISI][Medline]

8. Brandtzaeg P. Immunofluorescence studies of mucous membranes and exocrine glands. In: Immunofluorescence Technology: Selected Theoretical and Clinical Aspects, edited by Wick G, Traill KN, and Schauenstein K. New York: Elsevier, 1982, p. 167–217.

9. Bridges RJ, Rack M, Rummel W, and Schreiner J. Mucosal plexus and electrolyte transport across the rat colonic mucosa. J Physiol 376: 531–542, 1986.[Abstract/Free Full Text]

10. Carew MA and Thorn P. Carbachol-stimulated Cl^– secretion in mouse colon: evidence of a role for autocrine prostaglandin-E₂ release. Exp Physiol 85: 67–72, 2000.[Abstract]

11. Chang EB and Rao MC. Intestinal water and electrolyte transport: mechanisms of physiological and adaptive responses. Physiology of the Gastrointestinal Tract (3rd ed.), edited by Johnson LR. New York: Raven, 1994, vol. 2, p. 2027–2081.

12. Chang WWL and Leblond CP. Renewal of the epithelium in the descending colon of the mouse: I. Presence of three cell populations: vacuolated-columnar, mucous and argentaffin. Am J Anat 131: 73–99, 1971.[CrossRef][ISI][Medline]

13. Coetzee WA, Amarillo Y, Chiu J, Chow A, Lau D, McCormack T, Morena H, Nadal MS, Ozaita A, Pountney D, Saganich M, Vega-Saenz de Miera E, and Rudy B. Molecular diversity of K⁺ channels. Ann NY Acad Sci 868: 233–285, 1999.[Abstract/Free Full Text]

14. Dedek K and Waldegger S. Colocalization of KCNQ1/KCNE channel subunits in the mouse gastrointestinal tract. Pflügers Arch 442: 896–902, 2001.[CrossRef][ISI][Medline]

15. Diener M, Bridges RJ, Knobloch SF, and Rummel W. Neuronally mediated and direct effects of prostaglandins on ion transport in rat colon descendens. Naunyn Schmiedebergs Arch Pharmacol 337: 74–78, 1988.[ISI][Medline]

16. Freel RW, Vaziri ND, and Hatch M. Muscarinic down-regulation of cAMP-stimulated potassium ion secretion by rabbit distal colon. Pflügers Arch 440: 243–252, 2000.[ISI][Medline]

17. Fromm M and Hegel U. Net ion fluxes and zero flux limiting concentrations in rat upper colon and rectum during anaesthesia-induced aldosterone liberation. Pflügers Arch 408: 185–193, 1987.[CrossRef][ISI][Medline]

18. Fromm M, Hegel U, and Lüderitz S. Segmental heterogeneity of epithelial transport in rat large intestine. Pflügers Arch 378: 71–83, 1978.[CrossRef][ISI][Medline]

19. Fromm M, Schulzke JD, and Hegel U. Aldosterone low-dose, short-term action in adrenalectomized glucocorticoid-substituted rats: Na⁺, K⁺, Cl^–, HCO₃^–, osmolyte, and water transport in proximal and rectal colon. Pflügers Arch 416: 573–579, 1990.[CrossRef][ISI][Medline]

20. Geiman EJ, Zheng W, Fritschy JM, and Alvarez FJ. Glycine and GABA_A receptor subunits on Renshaw cells: relationship with presynaptic neurotransmitters and postsynaptic gephyrin clusters. J Comp Neurol 444: 275–289, 2002.[CrossRef][ISI][Medline]

21. Gerlach U. I_Ks channel blockers: potential antiarrhythmic agents. Drugs Future 26: 473–484, 2001.[CrossRef]

22. Gerlach U, Brendel J, Lang HJ, Paulus EF, Weidmann K, Brüggemann A, Busch A, Suessbrich H, Bleich M, and Greger R. Synthesis and activity of novel and selective I_Ks-channel blockers. J Med Chem 44: 3831–3837, 2001.[CrossRef][ISI][Medline]

23. Gögelein H, Brüggemann A, Gerlach U, Brendel J, and Busch AE. Inhibition of I_Ks channels by HMR1556. Naunyn Schmiedebergs Arch Pharmacol 362: 480–488, 2000.[CrossRef][ISI][Medline]

24. Greger R, Bleich M, Riedemann