| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,
* Department of Medicine, Division of Gastroenterology, University of California, San Diego, La Jolla, California, USA;
Division of Digestive Diseases, University of Cincinnati, Cincinnati, Ohio, USA; and
Division of Gastroenterology, Hepatology and Nutrition, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA
2Correspondence: University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0063, USA. E-mail: h2dong{at}ucsd.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: cystic fibrosis • guanylyl cyclase C knockout mice • STa-binding receptor
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. It has been well demonstrated that STa binds to membrane-bound guanylyl cyclase C (GC-C), which stimulates activity of the cystic fibrosis transmembrane conductance regulator (CFTR) and inhibition of sodium/hydrogen exchangers (NHEs) by a cGMP-dependent protein kinase (PKGII)-dependent pathway, thus resulting in inhibition of Na+ absorption and stimulation of Cl–, HCO3–, and water secretion (2)
.
In recent years, it has been shown, in both cystic fibrosis (CF) patients (3)
and CFTR knockout (KO) mice (4)
, that STa also possesses the ability to stimulate duodenal bicarbonate secretion (DBS) independent of CFTR. DBS is not only a critical component in maintaining the physiological environment of the intestine, but its lack also plays a role in the intestinal pathology associated with CF. While lung dysfunction remains the primary cause of mortality of CF patients, gastrointestinal complications account for a significant proportion of morbidity in pediatric and adult CF patients (5
, 6)
. Specifically, fasting and postprandial intraluminal duodenal pH in CF patients are 1–2 pH units lower than normal (7)
, which can lead to malabsorption through poor micellar solubilization by bile acids (8)
and pancreatic damage through increased expression of inflammatory genes (9)
.
GC-C is a member of the family of particulate guanylyl cyclases and has a distinct ligand binding site not shared by any of the other known members (10
, 11)
. Variants of GC-C have been cloned (12
, 13)
, including a full-length cDNA from opossum kidney (14)
that binds STa. The predicted protein sequence of the ligand-binding domain of this opossum cDNA is 55–58% identical to GC-C from other mammals, while the catalytic domains share 92–95% identity. Although there is an abundance of evidence supporting GC-C as the primary receptor mediating STa-stimulated intestinal secretion, there is also strong evidence suggesting the existence of alternate receptors based on studies of receptor cross-linking and kinetic analysis of STa binding. Several studies have demonstrated at least two distinct STa-binding sites in the intestine, of low and high affinity (15
16
17
18)
. While some studies have likewise shown that GC-C exhibits high- and low-affinity binding states (19
, 20)
, others have attributed these additional sites to non-GC-C STa-binding receptors. In addition, Hakki et al. (16)
observed an STa-binding protein in rat intestinal membranes that was not associated with GC-C activity. We have previously identified a specific STa binding species in the rat intestinal cell line IEC-6 (which does not express GC-C) that is not coupled to guanylyl cyclase activity (21)
. More recently, a report described the discovery in human breast cancer cell lines of a novel STa binding species (22)
present in high abundance and differing from GC-C in ligand specificity. Our prior study showing that STa stimulates CFTR-independent DBS through a mechanism distinct from cGMP signaling (4)
underscores the importance of determining whether non-GC receptors for STa exist in the intestine.
To address the important question as to whether non-GC-C receptors play a role in STa-stimulated DBS in CFTR KO mice, we therefore examined the biochemical and functional existence of STa-binding receptors in the GC-C KO mouse. While significant 125I-STa-binding occurred in the proximal small intestine of both GC-C KO and wild-type (WT) mice, this binding involved two distinct receptors. Functionally, STa, uroguanylin, and guanylin all stimulated significant increases in DBS in GC-C KO mice. However, uroguanylin- and guanylin-stimulated DBS were dependent on CFTR, while STa stimulated CFTR-independent DBS in GC-C KO mice. Therefore, we have shown that a non-GC-C receptor exists in the proximal intestine and functionally stimulates DBS through a mechanism independent of CFTR. Further insight into the identity of this receptor and its signaling mechanisms may provide novel therapies to rectify DBS dysfunction in CF patients, thereby resulting in a decrease in the gastrointestinal complications associated with CF.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 24)
. WT mice were generated by mating mice that were homozygous for both the CFTR [CFTR (+/+)] and GC-C [GC-C (+/+)] genes. As homozygous CFTR KO [CFTR (–/–)] mice often die prematurely because of intestinal obstruction and perforation (25)
, an electrolyte solution containing polyethylene glycol 3350 (GoLYTELY; Braintree Laboratories, Braintree, MA, USA) was administered in drinking water to all mice ad libitum. This solution has been shown to increase survival significantly in CF mice, without altering the histomorphological integrity of the intestine (26)
. This solution was also made available to GC-C KO and WT mice to maintain experimental consistency. Genotypes were determined by PCR using gene-specific primers, as described previously (4
, 23)
.
Intestinal membrane preparation
Brush-border membrane preparations were isolated from scraped mucosa of proximal small intestine from pools of 2–4 mice based on the method of Schmitz et al. (27)
, as previously modified for rat intestine (28)
. All steps were performed at 4°C with freshly added protease inhibitors. Fresh or frozen mucosal scrapings were diluted 30:1 in homogenization solution (0.050 M mannitol, 0.002 M Tris, pH 7.1, containing 0.1 M phenylmethylsulfonyl fluoride, 0.02 mg/ml leupeptin, and 0.001 mg/ml aprotinin) and homogenized (Ultra-Turrax, IKA Works, Staufen, Germany) for 1 min. For brush-border membrane preparations, CaCl2 was added to the homogenate to a final concentration of 0.01 M, mixed for 15 min, and centrifuged at 2000 g for 10 min. The pellet was discarded and the remaining supernatant centrifuged at 19,000 g for 20 min. The resulting pellet was then washed in 0.1 M NaCl, 0.05 M Tris, pH 7.4, resuspended, and stored as above. These steps resulted in a 6- to 9-fold enrichment in sucrase activity (29)
. Protein determinations were made using the Bio-Rad Protein Assay (Bio-Rad Laboratories, Hercules, CA, USA) with bovine serum albumin as the standard.
125I-STa binding assays
Purified STa (30)
was radioiodinated (Na125I, >350 mCi/ml, PerkinElmer Life and Analytical Sciences, Wellesley, MA, USA) by a lactoperoxidase method and 4-Tyr-125I-STa recovered by high-performance liquid chromatography, as described previously (21
, 31)
. Binding experiments were performed in a total volume of 0.5 ml by incubating membrane protein with 100,000 cpm 125I-STa for 60 min at 37°C in 0.1 M sodium acetate buffer, pH 4.8 containing 0.15% bovine serum albumin in the absence or presence of increasing amounts of unlabeled STa (0–1 µM). A rapid filtration assay was used as described previously (32)
to separate membrane-bound from free 125I-STa. Specific binding was determined by subtracting the amount of 125I-STa bound in the presence of excess unlabeled STa (1 µM) from total binding. All points were determined in triplicate. The affinity constants of binding (Ka) were calculated by Scatchard analysis of competitive inhibition of binding data for each genotype using the least-squares-fit computer program Ligand, as described by Munson and Rodbard (33)
(http://abs.cit.nih.gov/main/ligand.html).
Cross-linking of 125I-STa to membrane proteins
The cross-linker 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) was obtained from Pierce Biotechnology (Rockford, IL, USA). This agent is a carboxylic acid-amino group coupler and cross-linking was performed as described (28
, 34)
with the following modifications. Briefly, 125I-STa (500,000 cpm) was incubated with 100-µg membranes in a total volume of 500 µl for 60 min at 37°C in 0.1 M sodium acetate buffer, pH 4.8 containing 0.15% bovine serum albumin in the absence or presence of 2 µM of unlabeled STa. The reaction was centrifuged at 12,000 g for 10 min at 4°C, and the resulting pellet was rinsed with cold 0.05 M sodium acetate to remove uncoupled tracer. After resuspending the pellet in 100 µl of 0.1 M 2-[N-morpholino]ethane sulfonic acid, 0.5 M NaCl, pH 5.0, EDC was added to a final concentration of 2 mM, and the sample was incubated for 15 min at room temperature. Cross-linking was terminated by the addition of 10 µl 1 M Tris, pH 7.5. The cross-linked products were collected by centrifugation, resuspended in 40 µl sodium dodecyl sulfate (SDS) sample buffer under reducing conditions, and analyzed by SDS-PAGE. Gels were dried and exposed to BioMax MS film plus MS intensifying screen (Eastman Kodak, Rochester, NY, USA) for 2–4 days before developing film. Cross-linking experiments were performed 3 times, using different membrane preparations each time, with similar results.
Measurement of HCO3– secretion in vivo
In vivo experiments were performed using a well-validated technique, as described previously (35)
. Animals were maintained with free access to food and water for up to 1 h before the experiment, when food was removed. Anesthesia was induced by intraperitoneal administration of 10 µl/g body weight of hypnorm/midazolam cocktail. The abdomen was opened by two small lateral incisions, and the proximal 5–10 mm of duodenum, from the pylorus to just cephalad to the entry of the common bile duct, was isolated so as not to compromise vascular supply. A small polyethylene tube (PE-50) with a distal flange was advanced to the duodenal bulb via the stomach, and a ligature was secured around the pylorus. Distal to the pylorus and just proximal to the pancreaticobiliary duct, a small incision was made and PE-50 flanged tubing was secured by another ligature to allow for drainage. The isolated duodenal segment was gently flushed and then perfused (Harvard Infusion Pump, Harvard Apparatus, South Natick, MA, USA) at a rate of 0.17 ml/min with 154 mM NaCl (37°C). Effluents from the isolated segment were visually free of bile and blood throughout all experiments. Animals were fitted with a miniature O2 mask, maintained at 37°C, and hydrated with normal saline (50–100 µl/h subcutaneous). Anesthesia was maintained using the hypnorm/midazolam cocktail at 20% of the initial dose every 30–45 min as indicated by respiratory rate and toe pinch reflex. Respiratory and heart rates were determined every 10 min. Animals could be sustained for more than 3 h under these experimental conditions.
After an initial 20-min washout and recovery period, basal HCO3– secretion was measured for 30 min. Subsequently, STa (10–7 M), uroguanylin (10–6 M), or guanylin (10–6 M) was perfused intrasegmentally for 30 min. To measure their inhibitory effects, amiloride (10–3 M), DIDS (10–4 M), or glibenclamide (3x10–4 M), or DIDS plus glibenclamide were perfused for 30 min following the basal period, after which STa, uroguanylin, or guanylin plus DIDS alone, glibenclamide alone, or DIDS and glibenclamide were perfused for an additional 30 min. The segment was then gently flushed to remove residual stimulatory agents, and HCO3– secretion was measured for an additional 45 min. After each experiment, the length of the duodenal test segment was measured in situ to the nearest 0.5 mm.
Sample volumes were measured by weight to the nearest 0.01 mg. The amount of HCO3– in the effluents was quantitated by a validated micro back-titration method (Radiometer, Copenhagen, Denmark). Briefly, 100 µl of 50 mM HCl was added to the sample with 2 ml of distilled H2O. Samples were then gassed with N2, prewashed in Ba(OH)2 to remove all CO2, and back-titrated with 2.5 mM NaOH to an endpoint of pH 7.0 using an automated pH meter/titration unit (Radiometer, Copenhagen, Denmark). As reported previously (35)
, a series of in vitro standards was performed using the corresponding agonist perfusate, which showed excellent correlation between the amounts of HCO3– added and those recovered. To verify that the titratable HCO3– secretion measured by this pH-dependent technique was due to true HCO3– secretion and not altered H+ transport through inhibition of Na+/H+ exchange, we also measured HCO3– secretion via a CO2 gas-sensing electrode (model 950200; Thermo Orion, Beverly, MA, USA) as previously reported (36)
. Briefly, 0.2 ml of 1 M citrate buffer (pH 4.5) was added to each sample solution (2 ml) to convert free HCO3– to CO2, followed by measurement of electrode potential (mV) with the CO2 electrode. HCO3– output was calculated according to a calibration curve using freshly prepared 0.1, 1, and 10 mM NaHCO3 solution as standards, which generate 0.1, 1, and 10 mM CO2, respectively.
In addition, to verify that our observations were caused by stimulation or inhibition rather than a procedural artifact from changing perfusate solutions, we performed similar experiments in both WT and GC-C KO mice with perfusion of vehicular control. As expected, vehicle caused no significant stimulatory or inhibitory effect on duodenal HCO3– secretion (P>0.05, data not shown).
HCO3– secretion was determined in 15-min periods and expressed as micromoles per centimeter per hour and presented as HCO3– output over time or net peak HCO3– output (peak output minus average basal).
Measurement of HCO3– secretion in vitro
Stripped duodenal mucosae from WT and GC-C KO mice were mounted between two Lucite half chambers with an exposed area of 0.1 cm2 and placed in Ussing chambers. Experiments were performed under continuous short-circuited conditions (Voltage-Current Clamp, VCC 600; Physiological Instruments, San Diego, CA, USA), as described previously (37)
. The duodenal tissue from each animal was divided and examined in 3 chambers. Both mucosal and serosal solutions contained the following (in mM): 140 Na+, 5.4 K+, 1.2 Ca2+, 1.2 Mg2+, and 120 Cl–. The serosal bath contained (in mM): 25 HCO3–, 2.4 HPO42–, and 10 glucose. The luminal solution contained (in mM): 25 gluconate, and 10 mannitol. The osmolalities of all solutions were 
284 mOsm/kg.
After a 30-min measurement of basal parameters, STa (10–7 M), uroguanylin (10–6 M), or guanylin (10–6 M) was added to the mucosal side of tissues in Ussing chambers for 75 min. Measurements were recorded at 5-min intervals, and mean basal and peak values for consecutive 10-min periods were averaged. The rate of luminal HCO3– secretion is expressed as micromoles per square centimeters per hour. The short-circuit current (Isc) was measured in microamperes and converted into microequivalents per square centimeters per hour. Transepithelial conductance was determined by Ohm’s law and expressed as the delta millisiemens per square centimeter of tissue, which was calculated by subtracting peak values from average basal values.
Chemicals
STa, uroguanylin, guanylin, amiloride, DIDS, and glibenclamide were purchased from Sigma-Aldrich (St. Louis, MO, USA).
Statistical analysis
Results are expressed as means ± SE for a series of n experiments. Statistical analysis was performed using the Student’s t test for paired and unpaired data as appropriate. P values < 0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10% of the level seen with membranes from WT mice (23)
. We confirmed this finding in the mice used in the current study, which have been back-crossed to the C57BL/6J inbred strain for 5 generations, with the following similar results: WT, 6784 ± 848 cpm/20 µg vs. GC-C KO, 745 ± 88 cpm/20 µg (n=3). The level of binding present in the GC-C KO mouse intestine differs from WT as a function of pH (Fig. 1
A, B). While maximal binding for both genotypes was observed at pH 4.8 (the lowest pH studied), 50% of the binding with WT membranes remains at pH 6.5, while with membranes from GC-C KO intestine, 50% of the binding was lost at pH 5.5, and no apparent binding was seen at pH 6.5. Therefore, competitive displacement reactions were performed at pH 4.8 for both genotypes (Fig. 1C, D
). Brush-border membrane preparations (20 µg) were incubated with 125I-STa for 1 h at 37°C (equilibrium binding conditions), alone and in the presence of increasing concentrations of unlabeled STa. Analysis of these data using the least-squares fit computer program Ligand (33)
indicated that a single class of receptor sites was present in each genotype. The equilibrium binding constant of association (Ka) in WT mouse proximal small intestine was 4.8 x 109 L/mol; the Ka in GC-C KO proximal small intestine was 1.1 x 108 L/mol. Previously published Ka values have ranged from 8.7 x 108 L/mol for rat intestine (38)
to 2.0 x 109 L/mol seen for human intestine (32)
. Interestingly, the Ka for the 125I-STa binding species present in the intestinal cell line IEC-6, which does not express GC-C, is 2.0 x 108 L/mol (21)
. To further characterize the putative receptor/binding proteins present in GC-C KO intestine, we used the cross-linker EDC to covalently couple 125I-STa to intestinal membrane proteins. Autoradiograms of representative gels are shown in Fig. 2
. Cross-linking of STa to WT membrane proteins identified several species. As evidence of specificity, all were abolished in the presence of excess unlabeled STa, except for a 52-kDa protein. The specific cross-linked proteins identified were a predominant protein at 87 kDa, as well as a higher molecular weight protein of 120 kDa (more visible on longer exposure, not shown). In contrast, EDC cross-linking of the tracer to membrane proteins from GC-C KO mice identified a single species of 120 kDa, which was completely displaced in the presence of excess unlabeled STa. Nonspecific cross-linking to a 52-kDa protein was also apparent. For comparison, as previously published, Western blot analysis with an antibody directed against an extracellular peptide of GC-C identified 2 bands at 135 and 90 kDa in WT membranes (23)
. Both these bands were absent with membranes prepared from GC-C KO mice. Finally, EDC cross-linking with membranes prepared from IEC-6 cells, a cell type known to bind 125I-STa but not to express GC-C (21)
, resulted in a single specifically labeled protein at 120 kDa.
|
|
STa-stimulated duodenal HCO3– secretion in GC-C KO mice
After determining that a non-GC-C receptor that binds STa is present in the proximal intestine of mice, we sought to determine whether it could contribute to the ability of STa to stimulate DBS. To examine this, we utilized the GC-C KO mouse model to study STa-stimulated DBS in vivo (Fig. 3
A) and in vitro (Fig. 3B
). As expected, we found that STa stimulated significant increases in DBS over baseline in WT mice both in vivo (P<0.01) and in vitro (P<0.001). Likewise, STa-stimulated a significant increase in Isc in WT mice in vitro (P<0.01). When we examined the effect of STa on DBS in GC-C KO mice, we found that STa reproducibly elicited significant increases in DBS over baseline both in vivo (P<0.01) and in vitro (P<0.05) (Fig. 3)
. In contrast to WT mice, STa did not elicit a significant increase in Isc in GC-C KO mice (P>0.05) (Fig. 3B
). When we examined the transepithelial conductance, we found that whereas in WT mice, STa failed to stimulate a significant increase in conductance (–0.77±0.77 
mS/cm2, P>0.05), it stimulated a significant increase in conductance in GC-C KO mice (9.55±1.84 mS/cm2, P<0.01).
|
Uroguanylin- and guanylin-stimulated duodenal HCO3– secretion
To determine whether the endogenous homologues of STa, uroguanylin and guanylin, could also stimulate GC-C-independent DBS, we performed similar in vivo and in vitro experiments with these agonists. As seen for STa, we found that uroguanylin stimulated significant increases in DBS over baseline in both WT and GC-C KO in vivo (P<0.01, P<0.005, respectively) (Fig. 4
A) and in vitro (P<0.001, P<0.01, respectively) (Fig. 4B
). Similarly, uroguanylin stimulated a significant increase in Isc in vitro in WT (P<0.05), but not GC-C KO mice (P>0.05) (Fig. 4B
). In addition, we found that in WT mice uroguanylin failed to cause an increase in transepithelial conductance (–3.57±3.68 mS/cm2, P>0.05), whereas in GC-C KO mice, uroguanylin stimulated a significant increase in conductance (8.19±2.45 mS/cm2, P<0.05), similar to STa. When we examined guanylin-stimulated DBS in GC-C KO mice, we found that guanylin stimulated significant increases in DBS in both WT and GC-C KO mice in vivo (P<0.005, P<0.01, respectively) (Fig. 5
A) and in vitro (P<0.01, P<0.001, respectively) (Fig. 5B
). However, in contrast to uroguanylin and STa, guanylin stimulated a significant increase in Isc in both WT (P<0.05), and GC-C KO mice (P<0.001) (Fig. 5B
). Also, in contrast to STa and uroguanylin, guanylin stimulated a significant increase in the transepithelial conductance in both WT and GC-C KO mice (6.18±2.38 mS/cm2, P<0.05; 5.77±1.77 mS/cm2, P<0.05, respectively).
|
|
Because of the transient nature of the in vivo response to STa, uroguanylin, and guanylin, we verified that our observations were caused by stimulation rather than a procedural artifact from changing perfusate solutions, by performing similar experiments in both WT and GC-C KO mice with perfusion of vehicular control. As expected, vehicle caused no significant stimulatory or inhibitory effect on duodenal HCO3– secretion (data not shown), indicating that our observed responses are real biological phenomenon.
Involvement of Na+/H+ exchange
Subsequently, we performed experiments to verify that our observed changes in STa-, uroguanylin-, and guanylin-stimulated HCO3– secretion were, in fact, due to HCO3– transport and not due to alterations in H+ transport by NHE. To examine a potential effect of H+ secretion, we repeated experiments with STa, uroguanylin, and guanylin in GC-C KO mice and measured HCO3– secretion via a CO2 sensing electrode, which measures HCO3– directly rather than utilizing pH. Using this technique, we found that STa, uroguanylin, and guanylin stimulated significant increases in DBS in a similar manner as seen above (data not shown), thus indicating that H+ secretion was not involved in our observations. In addition, to determine whether inhibition of H+ secretion is involved in GC-C-independent HCO3– secretion in GC-C KO mice, we examined the effect of amiloride, a selective NHE inhibitor on STa-, uroguanylin-, and guanylin-stimulated HCO3– secretion. We found that inhibition of NHE by amiloride (10–3 M) resulted in no significant alterations in STa-, uroguanylin-, or guanylin-stimulated HCO3– secretion (STa: 2.30±0.19 vs. 2.51±0.15; uroguanlyin: 3.07±0.52 vs. 1.89±0.43; guanylin: 2.52±0.53 vs. 3.38±1.39; µmol·cm–1·h–1, P>0.05 vs. GC-C KO alone, n
3), again ruling out any contribution of H+ transport to STa-, uroguanylin-, or guanylin-stimulated HCO3– secretion.
Mechanism of GC-C-independent duodenal HCO3– secretion stimulated by STa, uroguanylin, and guanylin
In order to relate these findings to our prior observation that STa can stimulate DBS independent of CFTR in CFTR KO mice (4)
, we investigated whether GC-C-independent STa-, uroguanylin-, and guanylin-stimulated DBS were mediated by CFTR or apical Cl–/HCO3– exchange in vivo. To determine whether CFTR was involved in the responses, glibenclamide, a commonly used CFTR inhibitor, was perfused prior to and during stimulation in both WT and GC-C KO mice. We found that pretreatment with DIDS alone or DIDS plus glibenclamide completely abolished STa-stimulated DBS in GC-C KO mice, while glibenclamide alone had no inhibitory effect on STa-stimulated DBS (Fig. 6
A). Similarly, in WT mice, pretreatment with glibenclamide alone did not significantly inhibit STa-stimulated DBS (10.27±1.49 vs. 6.56±2.07 µmol·cm–1·h–1, n=3, P>0.05). In contrast, when we examined the effects of glibenclamide and DIDS on uroguanylin- and guanylin-stimulated DBS (Fig. 6B, C
), we found that glibenclamide alone or glibenclamide plus DIDS significantly inhibited both uroguanylin- and guanylin-stimulated DBS, while DIDS had no inhibitory effect (Fig. 6B, C
). These findings agree with our findings that STa, but not uroguanylin or guanylin, stimulated significant increases in DBS in CFTR KO mice in vivo (4.40±0.66, 1.74±0.70, 0.88±0.75 µmol·cm–1·h–1, n4, P<0.05 vs. WT, respectively). Taken together, these data indicate that STa-stimulated DBS in GC-C KO mice is mediated by apical Cl–/HCO3– exchange, but not by CFTR, while uroguanylin- and guanylin-stimulated DBS are CFTR-dependent in GC-C KO mice.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. However, previous reports that STa can stimulate a significant increase in HCO3– secretion in the duodenum of CF patients (3)
and CFTR KO mice (4)
implied a potential avenue via which HCO3– secretory defects in CF could be circumvented. In this study, we sought to further understand the underlying mechanism by investigating whether STa could bind to and stimulate duodenal HCO3– secretion independent of the classical STa receptor, GC-C.
While our previous study in CFTR KO mice suggested indirectly that STa can stimulate DBS independent of GC-C (4)
, in this study, we demonstrate that STa can not only bind to non-GC-C receptors but also can functionally stimulate an increase in DBS independent of GC-C. Previously, it was shown that GC-C-deficient mice continue to exhibit a small (10%) but reproducible degree of specific binding of STa to jejunal and colonic membranes, despite lacking GC-C receptors (23)
. In our current study, we have expanded these findings to show that 125I-STa binding in GC-C KO mice is 11% of WT in the proximal small intestine. The 40-fold difference in 125I-STa binding affinity between WT and GC-C KO intestinal membranes suggests that the major pathway for STa in WT intestine is via binding to GC-C. Any effects of STa through the lower affinity receptor may therefore be masked, given the high levels of expression of GC-C in the intestine. It is also likely that this large discrepancy accounts for differences in the functional responses of WT and GC-C KO intestines to STa. Although we found that STa stimulated a significant increase in DBS in GC-C KO mice, this response was significantly attenuated both in its magnitude and duration from that in WT mice. This may also reflect intrinsic differences in signaling via STa stimulation of GC-C vs. the lower affinity receptor. The transient nature of the response in GC-C KO mice is intriguing and requires further exploration in order to understand the signaling mechanism employed in GC-C KO mice in response to STa.
In the current study, we employed several different approaches to show that the STa-binding receptor found in GC-C KO mice was distinct from GC-C. Thus, we examined its binding affinity and pH dependence, as well as identified binding proteins through EDC cross-linking. Each of these studies showed distinct characteristics of the receptors found in WT and GC-C KO mice. As seen in Fig. 1B
, binding of ST to the non-GC-C receptor occurs only at pH < 6. This is of physiological significance, particularly in the duodenum in which there is a marked drop in postprandial pH. In humans, the fasting duodenal pH is neutral (6.8), while there is a marked drop in pH in the first hour after a meal, with pH < 5. In patients with CF, postprandial pH is lower and persists for a longer period (7
, 40)
. We speculate that endogenous uroguanylin may bind in a similar fashion to the non-GC-C receptor under conditions of low pH, such as seen transiently after a meal, resulting in HCO3– secretion and aiding in the return to neutral pH.
While we showed differences in EDC cross-linking between WT and GC-C KO mice, the identity and relevance of these proteins are unknown at this time. EDC is termed a zero-length cross-linker and therefore would identify proteins in close proximity to the tracer. These proteins may directly participate in the binding of STa, may be a nonbinding part of the actual receptor-STa complex, or may be unrelated to the receptor. Full-length GC-C has a mass of 130–160 kDa, depending on glycosylation status. However, the 87-kDa protein identified in WT membranes is very similar in size to the major proteolytic fragment of GC-C identified previously by several different groups (11
, 23
, 41
, 42)
. It has been suggested that proteolytic cleavage of GC-C may occur in the brush-border membrane, as well as during membrane isolation (11
, 40)
. The appearance of this 87-kDa protein only in membranes from WT mice is consistent with this band as a component of the GC-C receptor complex. Furthermore, the appearance of the 120-kDa protein in membranes from both KO and WT intestine, as well as IEC-6 cells, is consistent with the existence of a separate, non-GC-C, STa-binding species. Of note, the unidentified STa binding receptor recently reported in breast cancer cell lines by Giblin and colleagues (22
, 40)
also migrated at the same size, 123 kDa, as the STa binding species characterized in our present research. In contrast, Western blot detection of OK-GC, a highly similar relative of GC-C cloned from opossum kidney, revealed protein bands at 140–160 kDa in opossum kidney and intestine (14
, 40)
.
In addition to studying STa-stimulated DBS, we also examined uroguanylin- and guanylin-stimulated DBS. Because of the methods employed, we were not able to examine uroguanylin- or guanylin-binding; however, our functional data support the hypothesis that in addition to STa, uroguanylin and guanylin also bind to a non-GC-C receptor in the intestine to stimulate GC-C-independent DBS. The observations showing that uroguanylin and guanylin can stimulate ion transport in GC-C KO intestine are consistent with a number of studies in the kidney showing that these peptides can stimulate both GC-C-dependent and GC-C-independent ion transport (12
, 43)
. Despite being able to stimulate GC-C-independent DBS, however, uroguanylin and guanylin appear to recruit different signaling components in GC-C KO mice, as uroguanylin produced a significant increase in transepithelial conductance with no change in Isc, while guanylin significantly stimulated an increase in both conductance and Isc. In addition, in WT mice, while both uroguanylin and guanylin caused significant increases in DBS, only guanylin, but not uroguanylin, caused an increase in transepithelial conductance. In a similar manner, Sindic et al. (43)
previously reported that uroguanylin and guanylin can elicit distinct cellular responses. In principal cells of mouse cortical collecting ducts, it was found that depending on the concentration used, guanylin depolarized cells, while uroguanylin could cause both hyperpolarizing and depolarizing effects. The cause of these differences is unknown, but some have suggested different affinities of these endogenous ligands as a potential explanation (44)
. In our current study, we have not attempted to dissect the different signaling mechanisms between uroguanylin and guanylin. However, we feel that examination of the relative affinity of the endogenous ligands, uroguanylin and guanylin, for this alternate receptor, and characterization of the resulting signaling pathways is an intriguing line of future research.
While our finding that STa, uroguanylin, and guanylin can all stimulate DBS via a receptor distinct from GC-C is in itself novel, our primary purpose in undertaking this study was to gain further insight into how STa can stimulate significant amounts of DBS in CFTR KO mice. To determine whether our findings in GC-C KO mice could be applied to our results in CFTR KO mice (current study) (4)
, we determined whether STa-stimulated DBS in GC-C KO mice occurred through CFTR or Cl–/HCO3– exchange. We found that DIDS, but not glibenclamide, inhibited STa-stimulated DBS in GC-C KO mice, indicating that similar to CFTR KO mice, STa-stimulated DBS in GC-C KO mice also occurs via Cl–/HCO3– exchange. These results, together with our previous finding that STa-stimulated DBS in CFTR KO mice does not occur via cGMP signaling (4)
, provide further support for our hypothesis that STa-stimulated DBS in CFTR KO mice occurs via a GC-C-independent mechanism. In contrast to STa, uroguanylin and guanylin-stimulated DBS were both inhibited by glibenclamide, not DIDS. These results are in accordance with data in this study and previous reports in CFTR KO mice that uroguanylin- and guanylin-stimulated ion transports are CFTR-dependent. Thus, it appears that while STa, uroguanylin, and guanylin can all stimulate GC-C-independent DBS, only STa has the ability to stimulate a pathway that is independent of CFTR.
One caveat of our studies is that pharmacological inhibition of apical anion exchange cannot be entirely conclusive. In the duodenum, Slc26a6 (PAT-1), Slc26a3 (DRA), and Slc4a9 (AE4) are all expressed, with PAT-1 being the major anion exchanger responsible for DBS in the upper villous and DRA in the lower villous (AE4 has minimal contribution in either location) (45)
. While there have been conflicting reports regarding the DIDS-sensitivities of PAT-1 and DRA, there is evidence to support the notion that PAT-1 is DIDS sensitive (46)
and DRA is DIDS insensitive (47)
. On the basis of this premise, it is tempting to hypothesize that the complete abolition of STa-stimulated DBS in GC-C KO mice indicates that STa mediates GC-C-independent DBS through PAT-1. Similarly, while DIDS alone did not statistically significantly inhibit uroguanylin-stimulated DBS in GC-C KO mice, there was an inhibitory trend that suggests that PAT-1 may also play a role in uroguanylin-mediated DBS in GC-C KO mice. In contrast, DIDS had no inhibitory effect on guanylin-stimulated DBS, which may indicate that anion exchangers are not involved or that a DIDS-insensitive exchanger, such as DRA, is involved. Regardless, further work using PAT-1 and DRA KO mice is necessary to elucidate the exact roles of these anion exchangers in STa,- uroguanylin-, and guanylin-stimulated DBS.
Although often characterized primarily as a disease of chloride secretory dysfunction, defective HCO3– secretion accounts for a significant proportion of morbidity and mortality associated with CF. In this light, our finding that STa can stimulate HCO3– secretion independent of CFTR via a novel GC-C-independent mechanism is exciting. While previous studies have alluded to the existence of non-GC-C STa-binding receptors in the intestine, our study is apparently the first to show that non-GC-C receptors exist in the proximal intestine (duodenum) on both biochemical and functional grounds.
Furthermore, our observation that this novel pathway can mediate CFTR-independent HCO3– secretion holds promise as a potential mechanism to circumvent HCO3– secretory defects in CF. Because of the small, transient nature of the response in GC-C KO mice, therapeutic potential of this mechanism is not realistic. However, our studies in CFTR KO mice (current study) (4)
have shown that STa can stimulate prolonged increases in HCO3– secretion throughout the duration of stimulation, which are comparable to those in WT mice. At the present time, it is uncertain why these discrepancies between GC-C and CFTR KO mice occur. We have previously shown that STa-stimulated HCO3– secretion in CFTR KO mice occurs independently of cGMP (4)
, but we cannot rule out other potential roles of the GC-C receptor in this response. Although it is possible that these may involve different mechanisms, it may also be that long-term elimination of CFTR activity leads to an up-regulation of this alternate secretory pathway in CFTR KO mice. Further work is necessary to determine the identity of the receptor and signaling pathway involved in STa-stimulated DBS independent of GC-C and CFTR, which may lead to important insights into ways to alleviate complications in CF associated with deficient HCO3– secretion.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication November 5, 2006. Accepted for publication November 20, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  N. G. Moss, D. A. Riguera, R. M. Solinga, M. M. Kessler, D. P. Zimmer, W. J. Arendshorst, M. G. Currie, and M. F. Goy The Natriuretic Peptide Uroguanylin Elicits Physiologic Actions Through 2 Distinct Topoisomers Hypertension, May 1, 2009; 53(5): 867 - 876. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Lucas Is an increase in duodenal bicarbonate concentration after STa really enhanced bicarbonate ion secretion? FASEB J, February 1, 2009; 23(2): 294 - 295. [Full Text] [PDF]
|
|
|
|
|
|
Z. M. Sellers and H. Dong Response to: "Is an increase in duodenal bicarbonate concentration after STa really enhanced bicarbonate ion secretion?" FASEB J, February 1, 2009; 23(2): 295 - 296. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
