`secretion by uroguanylin
`
`⫺
`
`NAM SOO JOO,1 ROSLYN M. LONDON,1,2 HYUN DJU KIM,1
`LEONARD R. FORTE,1,2 AND LANE L. CLARKE3,4
`1Department of Pharmacology, School of Medicine, 3Department of Veterinary Biomedical Sciences,
`College of Veterinary Medicine, and the 4Dalton Cardiovascular Research Center, University
`of Missouri and 2The Truman Veterans Affairs Medical Center, Columbia, Missouri 65212
`
`Joo, Nam Soo, Roslyn M. London, Hyun Dju Kim,
`Leonard R. Forte, and Lane L. Clarke. Regulation of
`⫺ secretion by uroguanylin. Am. J.
`intestinal Cl⫺ and HCO3
`Physiol. 274 (Gastrointest. Liver Physiol. 37): G633–G644,
`1998.—Uroguanylin is an intestinal peptide hormone that
`may regulate epithelial ion transport by activating a receptor
`guanylyl cyclase on the luminal surface of the intestine. In
`this study, we examined the action of uroguanylin on anion
`transport in different segments of freshly excised mouse
`intestine, using voltage-clamped Ussing chambers. Urogua-
`nylin induced larger increases in short-circuit current (Isc) in
`proximal duodenum and cecum compared with jejunum,
`ileum, and distal colon. The acidification of the lumen of the
`proximal duodenum (pH 5.0–5.5) enhanced the stimulatory
`action of uroguanylin. In physiological Ringer solution, a
`significant fraction of the Isc stimulated by uroguanylin was
`⫺ in the
`insensitive to bumetanide and dependent on HCO3
`bathing medium. Experiments using pH-stat titration re-
`⫺
`vealed that uroguanylin stimulates serosal-to-luminal HCO3
`⫺
`HCO3
`) together with a larger increase in Isc. Both
`secretion (J s=l
`⫺
`HCO3
`and Isc were significantly augmented when luminal pH
`J s=l
`was reduced to pH 5.15. Uroguanylin also stimulated the
`⫺
`HCO3
`and Isc across the cecum, but luminal acidity caused a
`J s=l
`generalized decrease in the bioelectric responsiveness to
`agonist stimulation. In cystic fibrosis transmembrane conduc-
`tance regulator (CFTR) knockout mice, the duodenal Isc
`response to uroguanylin was markedly reduced, but not
`eliminated, despite having a similar density of functional
`receptors. It was concluded that uroguanylin is most effective
`in acidic regions of the small intestine, where it stimulates
`⫺ and Cl⫺ secretion primarily via a CFTR-depen-
`both HCO3
`dent mechanism.
`cyclic GMP; bicarbonate transport; chloride transport; cystic
`fibrosis; cystic fibrosis transmembrane conductance regula-
`tor; guanylyl cyclase; mouse intestine; proximal duodenum
`
`
`
`G634
`
`PHYSIOLOGICAL FUNCTIONS OF UROGUANYLIN
`
`short-circuit current (Isc), an index of anion secretion.
`In the most responsive intestinal segments (proximal
`duodenum and cecum), we examined the pH depen-
`dence of uroguanylin in stimulating transepithelial
`⫺ across the mucosa.
`secretion of Cl⫺ and HCO3
`MATERIALS AND METHODS
`
`Animals
`
`Female C57BL6 mice 8–10 wk old were housed in a
`standard animal care room with a 12:12-h light-dark cycle.
`Animals were allowed free access to food and water until the
`time of study. CFTR knockout and normal littermate mice
`(B6.129-Cftrtm/UNC; C57BL/6J-Cftrtm/UNC) were also main-
`tained on standard laboratory chow, but the water contained
`an osmotic laxative (polyethylene glycol) to reduce intestinal
`malfunction in the cftr(⫺/⫺) mice (5). All experiments involv-
`ing the animals were approved by the University Institu-
`tional Animal Care and Use Committee.
`
`Tissue Preparation
`
`Before each experiment, the mice were fasted for a mini-
`mum of 1 h and only water was provided. The mice were killed
`by a brief exposure to a 100% CO2 gas atmosphere (to induce
`narcosis), followed immediately by a surgical pneumothorax.
`A midline abdominal incision was used to excise the gallblad-
`der and the following intestinal segments: proximal duode-
`num (a portion from 2 mm distal to the pylorus to the
`sphincter of Oddi), midjejunum, ileum, cecum (a portion 1–2
`cm proximal to the cecal apex), and distal colon. The excised
`segments were opened along the mesenteric border in ice-
`cold, oxygenated Krebs-Ringer-bicarbonate (KRB) solution
`and pinned mucosal-side down on a pliable silicone surface.
`The intestinal sections were stripped of their outer muscle
`layers by shallow dissection with a scalpel and fine forceps.
`
`Bioelectric Measurements
`Each intestinal sheet (⬃1 cm in length) and the microdis-
`sected gallbladder (with a support of nylon gauze) were
`mounted in standard Ussing chambers with an exposed
`surface area of 0.25 cm2 for intestinal preparations (or 0.126
`cm2 for ileum and colon) and 0.049 cm2 for the gallbladder as
`previously described (5). The tissue sheets were indepen-
`dently bathed on the serosal and mucosal surfaces with 4 ml
`of KRB solution containing (in mM) 115 NaCl, 4 K2HPO4, 0.4
`KH2PO4, 25 NaHCO3, 1.2 MgCl2, and 1.2 CaCl2, pH 7.4. To
`facilitate pH adjustment of the medium in some experiments,
`a phosphate-free Ringer solution of the following composition
`was used (in mM): 115 NaCl, 5 KCl, 25 NaHCO3, 1.2 MgSO4,
`and 1.2 calcium gluconate, pH 7.4. In ion substitution experi-
`⫺ and Cl⫺ were replaced with TES and gluconate,
`ments, HCO3
`respectively. Glucose (10 mM) was included in the serosal
`solution (both baths in large intestinal preparations), and 10
`mM mannitol was substituted for glucose in the mucosal bath
`to prevent Na⫹-coupled glucose current stimulation. To mini-
`mize tissue exposure to endogenously generated prostaglan-
`dins during tissue preparation and mounting, indomethacin
`(1.0 µM) was present in both baths throughout the experi-
`ment. KRB solutions were gassed with 95% O2-5% CO2,
`⫺-free bathing solutions were gassed with 100%
`whereas HCO3
`O2. The solutions were circulated via a gas-lift recirculation
`system and were maintained at 37°C by water-jacketed
`reservoirs. In all experiments, TTX (0.1 µM) was added to the
`serosal bath at least 20 min before each experiment to
`prevent any intrinsic neural influence on ion transport regu-
`lation of the intestine (47).
`
`Transmural Isc (µA/cm2 tissue surface area) was measured
`with the use of an automatic voltage clamp device (VCC-600;
`Physiologic Instruments, San Diego, CA) that compensates
`for electrode offset and the fluid resistance between the
`potential-measuring electrode bridges. Transepithelial poten-
`tial difference (in mV) was measured via a pair of calomel
`half-cells connected to the serosal and mucosal baths by 4%
`agar-Ringer (wt/vol) bridges. Isc was applied across the tissue
`via a pair of Ag/AgCl electrodes that were kept in contact with
`the serosal and mucosal baths through 4% agar-Ringer
`bridges. All experiments were carried out under short-
`circuited conditions. Total tissue conductance (Gt, mS/cm2
`tissue surface area) was calculated by applying Ohm’s law to
`the current deflection resulting from a 5-mV bipolar pulse
`across the tissue every 5 min during the course of the
`experiment. In all cases, the serosal side served as ground
`and the Isc was conventionally referred to as negative when
`current flowed from the lumen to the serosa.
`After the tissues achieved a stable Isc (⬃20 min post-TTX),
`a 20-min period was required to adjust the luminal bath pH.
`The small intestine and gallbladder preparations were then
`sequentially exposed to a peptide (uroguanylin, 1.0 µM;
`guanylin, 1.0 µM; or STa, 0.02 µM) in the luminal bath for 30
`min and then to bumetanide (0.1 mM) in the serosal bath for
`10 min to inhibit the Na⫹-K⫹-2 Cl⫺ cotransporter. After pH
`adjustment, large intestinal preparations were first treated
`with amiloride (0.1 mM) in the luminal bath for 20 min to
`inhibit electrogenic Na⫹ absorption and were then treated
`with peptide addition followed by bumetanide. In studies in
`which the effects of acidic pH on the action of the peptides
`were examined, the proximal duodenum, jejunum, and cecum
`were used, and the pH of the luminal bath was decreased to
`pH 5.0–5.5 by addition of 1 N HCl. An equal amount of 1 M
`NaCl was added simultaneously to the serosal bath to pre-
`vent a transepithelial Cl⫺ diffusion potential. At the end of an
`experiment, glucose (10 mM) was added to the luminal bath
`of the small intestinal preparations and carbachol (CCh; 0.1
`mM) was added to the serosal bath of the large intestinal
`preparations as measures of tissue viability.
`
`Bioassay for cGMP Accumulation in Mouse Intestine
`
`Mucosal epithelium was prepared by scraping the intesti-
`nal segment from cftr(⫺/⫺) and cftr(⫹/⫹) mice and washing
`it gently once in 0.9% NaCl and twice in DMEM containing 20
`mM HEPES, pH 7.4. The mucosal suspension (⬃60 mg wet
`wt) was placed in 0.2 ml DMEM (pH 7.4) at 4°C. The tissue
`was incubated for 40 min at 37°C with either 1.0 µM
`uroguanylin, 1.0 µM guanylin, or vehicle that was added to
`the DMEM-HEPES with 1.0 mM 3-isobutyl-1-methylxan-
`thine. At the end of the 40-min period, perchloric acid was
`added to a final concentration of 3.3%, the cells were centri-
`fuged, and the resulting supernatants were neutralized with
`1 N KOH. The supernatants were used to measure cGMP
`concentration by radioimmunoassay as described previously
`(15).
`
`pH-Stat Titration
`
`Proximal duodenum or cecum was mounted in a standard
`Ussing chamber bathed with 156.2 mM NaCl in the luminal
`bath and KRB in the serosal bath. In cecal experiments, 1.2
`mM CaCl2 and MgCl2 were also added to the luminal bath.
`The luminal bath was gassed with 100% O2 and the serosal
`bath with 95% O2-5% CO2. To decrease the pH of the luminal
`bath, the luminal saline solution was gassed with 95% O2-5%
`CO2 and maintained at pH 5.15. The serosal-to-luminal flux
`⫺
`HCO3
`
`of HCO3⫺ (J s=l
`) was measured by continuously titrat-
`
`Downloaded from journals.physiology.org/journal/ajpgi (165.001.202.250) on February 23, 2022.
`
`
`
`PHYSIOLOGICAL FUNCTIONS OF UROGUANYLIN
`
`G635
`
`ing the luminal solution to pH 7.4 (or pH 5.15) with 5 mM
`HCl, using either a computer-aided titrimeter (model 455/
`465; Fisher) or manual addition of titrant. The volume of
`⫺ flux, taking into
`added acid was used to calculate the HCO3
`account the exposed surface area of tissue (0.25 cm2) and
`⫺
`HCO3
`time. Typically, J s=l
`stabilized within 30 min after the
`tissue was mounted and a basal flux period was initiated.
`After 30 min, uroguanylin (1.0 µM) was added to the luminal
`⫺
`stabilized (⬃20 min), a second
`HCO3
`bath, and when the J s=l
`30-min flux period was initiated.
`
`Receptor Autoradiography
`
`Dissected intestinal segments (proximal duodenum, midje-
`junum, ileum, cecum, and distal colon) from control and
`CFTR knockout mice were quickly frozen with dry ice and
`stored at ⫺80°C. The frozen tissue samples were sectioned to
`a 12-µm thickness in a cryostat device (2800 Frigocut N,
`Reichert-Jung; Leica Instrument, Germany) maintained at
`⫺20°C. Two sections were placed on opposite ends of a
`gelatin-coated slide, one for total binding (TB) and one for
`nonspecific binding (NSB), and then dried and stored at
`⫺80°C until used. Iodinated STa was used as the radioligand
`for this assay and was prepared by a modification of the
`method described by Krause and co-workers (35). The use of
`STa was necessary because iodination of guanylin (Tyr-9)
`may decrease the biological activity of this ligand (35) and
`uroguanylin has no tyrosine residue available for labeling.
`For the assay, each section on the slide was incubated with 30
`µl of DMEM containing 0.5% BSA to minimize background
`labeling (pH 5.5 at 37°C for 20 min). Total binding of
`125I-labeled STa was assessed by incubating 30 µl DMEM
`containing 1,000 cpm/µl 125I-STa on one section, and nonspe-
`cific binding was assessed by incubating 30 µl DMEM contain-
`ing both 125I-STa and unlabeled STa (1.0 µM), uroguanylin (10
`µM), or guanylin (10 µM) on the other section. After a 20-min
`incubation at 37°C, slides were washed five times with
`ice-cold phosphate-buffered saline solution and then air-
`dried. To verify labeling in the different intestinal segments,
`the slides were arranged in cassettes and exposed to Kodak
`X-OMAT AR (XAR 5) film overnight at ⫺80°C, and then the
`film was developed. Slides were then transferred in a dark
`room, coated with Kodak NTB-2 emulsion solution, and dried
`overnight. The emulsion-coated slides were sealed in light-
`tight boxes and stored at 4°C for 2–3 wk until they were
`developed. After development, fixation, and coverslipping, the
`radiolabeled TB and NSB sections were examined under
`bright- and dark-field microscopy.
`
`Materials
`
`Opossum uroguanylin (QEDCELCINVACTGC) and E. coli
`ST (CCELCCNPACTGC), STa13, were synthesized by the
`solid-phase method, using an Applied Biosystems peptide
`synthesizer, as previously described (25). Opossum urogua-
`nylin differs from murine uroguanylin (TDECELCINVAC-
`TGC) only in the sequence of the first three amino acids.
`Purified rat guanylin (PNTCEICAYAACTGC) was generously
`provided by Dr. Mark Currie (Searle Research and Develop-
`ment, St. Louis, MO). The iodination of E. coli ST (NSSNYC-
`CELCCNPACTGCY) was performed using a lactoperoxidase
`method as previously described (15, 16). Membrane-perme-
`able 8-bromo-cAMP (8-BrcAMP) and 8-bromo-cGMP (8-
`BrcGMP) were obtained from Research Biochemical Interna-
`tional (Natick, MA). All other chemicals were purchased from
`either Sigma Chemical (St. Louis, MO) or Fisher Scientific
`(Springfield, IL). Uroguanylin, guanylin, and E. coli STa were
`dissolved in deionized water at a stock concentration (s.c.) of 1
`
`mM. TTX was dissolved in 0.2% acetic acid at a stock
`concentration of 0.1 mM. Indomethacin (s.c., 10 mM), bu-
`metanide (s.c., 0.1 M), methazolamide (s.c., 1.0 M), DIDS (s.c.,
`0.3 M), and amiloride (s.c., 0.1 M) were dissolved in DMSO.
`
`Data Analysis
`
`Data are means ⫾ SE. Student’s t-test for paired or
`unpaired data or an ANOVA protected least-significant differ-
`ent test was used for comparisons of means among different
`intestinal segments and different treatment groups. In all
`cases, P ⬍ 0.05 was accepted as a statistically significant
`difference.
`
`RESULTS
`
`Segmental Responses to Uroguanylin in the
`Mouse Intestine
`Figure 1A shows the pattern of Isc responses to
`sequential treatment with specific agents on proximal
`duodenum. After the tissue was mounted, the addition
`of TTX (0.1 µM, serosal bath) resulted in a decrease in
`the baseline Isc to a stable value within 20 min.
`Uroguanylin (1.0 µM, luminal bath) caused a rapid
`increase in Isc, which was sustained for a 40-min period.
`Subsequent addition of STa (1.0 µM, luminal bath)
`elicited a further increase in Isc. Bumetanide (0.1 mM,
`serosal bath) treatment resulted in a decrease in the Isc
`but to a level that was elevated relative to the Isc before
`the uroguanylin/STa treatment. The inhibitory effect of
`bumetanide on Isc typically reached steady state by 10
`min posttreatment, i.e., the percentage of bumetanide
`inhibition at 5 min was equal to 98 ⫾ 1.1% of the Isc at
`10 min postbumetanide (n ⫽ 15) in proximal duode-
`num. Glucose (10 mM, luminal bath) addition intended
`to stimulate Na⫹-coupled glucose transport caused a
`rapid increase in Isc. In contrast to apical treatment, the
`addition of either uroguanylin or STa to the serosal
`bath solution had no effect on the Isc (Fig. 1B).
`Cumulative data of the Isc responses to uroguanylin
`(1.0 µM) by different intestinal segments and the
`gallbladder are shown in Fig. 2. Uroguanylin elicited
`an Isc response in all intestinal segments but had no
`stimulatory effect on the gallbladder preparations.
`Proximal duodenum and cecum had the greatest mean
`Isc responses to uroguanylin (1.0 µM) and were not
`significantly different from each other. However, the Isc
`response in the proximal duodenum was significantly
`greater than the responses in the other small intestinal
`segments (jejunum and ileum). Likewise, the cecum
`had a greater Isc response than did the distal colon. The
`concentration of uroguanylin (1.0 µM) used in these
`experiments was not a maximally stimulatory dose,
`since subsequent addition of STa (1.0 µM) elicited
`further increases in the Isc in all intestinal segments
`that were tested (⌬Isc, in µA/cm2): 95.5 ⫾ 10.3 in
`proximal duodenum (n ⫽ 7), 112.0 ⫾ 18.5 in jejunum
`(n ⫽ 9), 64.5 ⫾ 10.5 in ileum (n ⫽ 9), 77.5 ⫾ 15.8 in
`cecum (n ⫽ 9), and 58.4 ⫾ 13.5 in distal colon (n ⫽ 7).
`Proximal Duodenum
`Effect of luminal acidic pH on uroguanylin action. To
`examine the effect of luminal acidity on uroguanylin
`
`Downloaded from journals.physiology.org/journal/ajpgi (165.001.202.250) on February 23, 2022.
`
`
`
`G636
`
`A
`
`-250
`
`-225
`
`-200
`
`-175
`
`-100
`
`-75
`
`-50
`
`-25
`
`0
`0
`
`TTX Uroguanylin
`
`1 1
`
`STa
`
`1
`
`Bumet GIc
`
`1 1
`
`20
`
`40
`
`80
`60
`Time (min)
`
`100
`
`120
`
`TTX Uroguanylin
`
`1 1
`
`-250
`
`-225
`
`-200
`
`-175
`
`STa
`
`1
`
`Bumet Glc
`
`1 1
`
`-100
`
`to
`
`-75
`
`-50
`
`-25
`
`100
`
`120
`
`20
`
`40
`
`80
`60
`Time (min)
`Fig. 1. Time course of short-circuit current (Isc) response to sequen-
`tial treatments in mouse intestine. Data are representative of 11
`separate experiments with proximal duodenum bathed in standard
`Krebs-Ringer-bicarbonate (KRB). A: uroguanylin (1.0 µM), Esch-
`erichia coli heat-stable enterotoxin (STa13; 1.0 µM), and glucose (Glc,
`10 mM) were applied to the luminal bath. B: uroguanylin (1.0 µM)
`and STa (1.0 µM) were added to serosal bath. TTX (0.1 µM) and
`bumetanide (Bumet; 0.1 mM) were added to serosal bath.
`
`bioactivity in the native intestine, the pH of the luminal
`bath was reduced to 5.0–5.5 for 20 min before the tissue
`was treated with the peptide agonists (pH adjustment
`period). The proximal duodenum was chosen because
`this segment elicited pronounced increases in urogua-
`nylin-induced Isc (Fig. 2) and because this part of the
`small intestine has an acidic intraluminal environment
`during digestion (1). The intraluminal pH reduction
`caused a rapid and reproducible increase in the Isc of
`13.4 ⫾ 2.7 µA/cm2, reaching a new steady-state Isc
`
`PHYSIOLOGICAL FUNCTIONS OF UROGUANYLIN
`
`baseline in about 5 min (Fig. 3A, pH adjustment
`period). A recent study suggests that the acid-induced
`current may represent stimulation of electrogenic,
`⫺ secretion in the murine duode-
`CFTR-dependent HCO3
`num (30). Subsequent uroguanylin treatment at the
`acidic pH produced an approximately twofold greater
`(P ⬍ 0.05) increase in Isc than was observed at pH 7.4
`(Fig. 3, A and B). Again, the increased Isc elicited by
`uroguanylin was only partially inhibited by bumetanide
`treatment: 18.7 ⫾ 5.7% at pH 7.4 and 27.0 ⫾ 8.0% at pH
`5.0–5.5 (Fig. 3B). At both pH conditions, uroguanylin
`treatment slightly increased Gt, but the changes were
`not statistically different from each other: 2.0 ⫾ 0.4
`mS/cm2 at pH 7.4 and 3.1 ⫾ 0.3 mS/cm2 at acidic pH.
`The effect of luminal acidity on uroguanylin bioactivity
`in jejunal tissue was also examined. The reduction of
`intraluminal pH to 5.0–5.5 again increased basal Isc
`(19.6 ⫾ 4.8 µA/cm2), and subsequent treatment with
`1.0 µM uroguanylin at acidic pH produced a signifi-
`cantly larger increase in Isc than was observed at pH 7.4
`(66 ⫾ 4.2 vs. 44.6 ⫾ 5.8 µA/cm2, P ⬍ 0.05, n ⫽ 4). In a
`dose-response study, uroguanylin was more active (P ⬍
`0.05) in acidic luminal conditions than at pH 7.4 within
`the tested range of concentrations (Fig. 4). It was not
`possible to test higher concentrations of uroguanylin
`because the supply of the peptide was inadequate for
`the experiment.
`Effect of luminal acidity on the actions of 8-BrcGMP
`and 8-BrcAMP. To test whether the effect of luminal
`acidity on uroguanylin action resulted from an effect of
`low extracellular pH on the bioelectric properties or
`intracellular signaling pathways of the epithelial cells,
`the Isc responses elicited by membrane-permeable cy-
`clic nucleotides, 8-BrcGMP and 8-BrcAMP, were exam-
`ined at acidic and physiological pH. In the proximal
`
`60
`
`50-
`
`40—
`
`30—
`
`20—
`
`10-
`
`cs7. -
`
`E 0
`
`r..)
`cn
`<1
`
`a,b
`
`b,c
`
`0
`
`1
`
`1
`
`Ileum
`
`Cecum
`
`II
`Proximal Mid
`Distal Gallbladder
`Colon
`Duodenum Jejunum
`Fig. 2. Segmental responses to luminal uroguanylin in mouse intesti-
`nal preparations. All tissues were bathed with standard KRB solu-
`tion. ⌬Isc was calculated as baseline Isc minus maximal Isc response to
`uroguanylin. Large intestinal epithelia (cecum and distal colon) were
`pretreated with amiloride (0.1 mM, luminal bath) 20 min before
`uroguanylin addition. Values are means ⫾ SE obtained from proxi-
`mal duodenum (n ⫽ 11), midjejunum (n ⫽ 9), ileum (n ⫽ 9), cecum
`(n ⫽ 12), distal colon (n ⫽ 5), and gallbladder (n ⫽ 4). Means without
`a letter in common are significantly different; one-way ANOVA
`protected least-significant difference test, P ⬍ 0.05.
`
`Downloaded from journals.physiology.org/journal/ajpgi (165.001.202.250) on February 23, 2022.
`
`
`
`PHYSIOLOGICAL FUNCTIONS OF UROGUANYLIN
`
`G637
`
`TTX pH adjustment Uroguanylin
`
`1 1 1
`
`-+-- pH 5.0-5.5
`—c— pH 7.4
`
`Bumet Glc
`
`1 1
`
`A
`
`-200
`
`-180
`
`-160
`
`-140
`
`"E -120
`O
`
`-100
`3.
`
`-80
`
`-o-- pH 5.0-5.5
`-0- pH 7.4
`
`120
`
`100
`
`80
`
`60
`
`40
`
`20
`
`c.)
`
`0
`
`20
`
`60
`40
`Time (min)
`
`80
`
`100
`
`Uroguanylin (1.0 µM)
`+Bumet (0.1 mM)
`
`-60
`
`-40
`
`-20
`
`0
`
`0
`
`B
`
`125-
`
`100-
`
`25-
`
`pH 7.4
`pH 5.0-5.5
`Fig. 3. Effect of pH on Isc response to uroguanylin in mouse proximal
`duodenum. A: representative time course of Isc response to agents in
`proximal duodenum superfused with Ringer solution at either pH
`5.0⬃5.5 or pH 7.4 in luminal bath. Luminal pH was reduced to pH
`5.0–5.5 by addition of 1.0 N HCl (pH adjustment), and serosal pH was
`maintained at 7.4. TTX (0.1 µM), uroguanylin (1.0 µM), and glucose
`(10 mM) were added to luminal bath, and bumetanide (0.1 mM) was
`added to serosal bath. B: cumulative data showing maximal change
`in Isc (⌬Isc) from baseline during uroguanylin treatment and after
`sequential addition of bumetanide (⫹Bumet). Results are means ⫾
`SE from intact proximal duodena of 8 (pH 7.4 group) and 9 (acid-
`treated group) different mice. *Significantly different from urogua-
`nylin treatment (paired t-test, P ⬍ 0.05). †Significantly different from
`pH 7.4 group (unpaired t-test, P ⬍ 0.05).
`
`duodenum, membrane-permeable 8-BrcGMP (20 µM)
`stimulated the Isc more than did an equimolar concen-
`tration of 8-BrcAMP (P ⬍ 0.001) (Fig. 5). However, the
`Isc response elicited by 8-BrcGMP at pH 7.4 was similar
`to that observed under acidic luminal conditions. In
`contrast, 8-BrcAMP (20 µM) was significantly less
`effective under acidic conditions than at pH 7.4 (P ⬍
`0.05). The increased Isc elicited by either 8-BrcGMP or
`
`-8
`
`-5
`
`-7
`Log [Uroguanylin, M]
`Fig. 4. Effect of luminal bath pH on noncumulative concentration-Isc
`response curve for uroguanylin in intact mouse proximal duodenum.
`Luminal bath pH was reduced to 5.0–5.5 by addition of 1.0 N HCl,
`and basolateral pH was maintained at pH 7.4. ⌬Isc was calculated as
`baseline Isc minus maximal Isc response to uroguanylin. Values are
`means ⫾ SE obtained from 4–9 different mice. *Significantly differ-
`ent from pH 7.4 group (unpaired t-test, P ⬍ 0.01).
`
`8-BrcAMP was significantly inhibited (P ⬍ 0.05) by
`bumetanide (0.1 mM) treatment: 59.3 ⫾ 1.8% at pH 7.4
`and 65.7 ⫾ 3.4% at pH 5.0–5.5, or 61.5 ⫾ 5.9% at pH 7.4
`and 90.8 ⫾ 4.2% at pH 5.0–5.5, respectively.
`Effect of luminal acidity on guanylin and STa ac-
`tions. To investigate whether the pH dependence was
`specific for uroguanylin action, we examined the effects
`of acidic pH on the secretagogue actions of guanylin
`and STa (Fig. 6). Whereas uroguanylin is more effective
`in stimulating the Isc under acidic luminal conditions
`
`120-
`
`100-
`
`80-
`
`6
`
`E
`
`-1 40-
`
`20-
`
`CI pH 7.4
`pH 5.0-5.5
`
`*
`
`0
`
`8 Br-cAMP +Bumet
`8-Br-cGMP +Bumet
`Proximal Duodenum
`Fig. 5. Effect of luminal bath pH on Isc response to membrane-
`permeable cyclic nucleotides in mouse proximal duodenum. 8-
`BrcGMP (20 µM) or 8-BrcAMP (20 µM) was added to both luminal
`and serosal baths 20 min after pH adjustment. Bumetanide (0.1 mM)
`was added to serosal bath 30 min later. ⌬Isc was calculated as
`baseline Isc minus maximal Isc response during cyclic nucleotide
`treatment and after subsequent treatment with bumetanide. Data
`are means ⫾ SE; proximal duodena were obtained from 4–6 different
`mice in each cyclic nucleotide treatment group. *Significantly differ-
`ent from 8-BrcGMP-treated group (unpaired t-test; P ⬍ 0.001).
`†Significantly different from pH 7.4 (unpaired t-test; P ⬍ 0.05).
`
`Downloaded from journals.physiology.org/journal/ajpgi (165.001.202.250) on February 23, 2022.
`
`
`
`I
`
`I Uroguanylin (1.0 µM)
`+Bumetanide (0.1 mM)
`
`G638
`
`PHYSIOLOGICAL FUNCTIONS OF UROGUANYLIN
`
`50-
`
`40-
`
`30-
`
`20-
`
`10-
`
`Change from Basal Isc (RA/cm2)
`
`CI pH 7.4
`pH 5.0-5.5
`
`100-
`
`80-
`
`60-
`
`20-
`
`T
`
`0
`
`STa (20 nM)
`
`Uroguanylin (1.0 µM)
`Guanylin (1.0 µM)
`Proximal Duodenum
`Fig. 6. Effect of luminal bath pH on Isc response to uroguanylin,
`guanylin, and STa in intact mouse proximal duodenum. Each peptide
`was added to luminal bath 40 min after TTX treatment (20 min after
`pH adjustment period). ⌬Isc was calculated as baseline Isc minus
`maximal Isc response to a peptide. Results are means ⫾ SE; proximal
`duodena were obtained from 6–9 different mice for each peptide
`tested. *Significantly different from pH 7.4 group (unpaired t-test,
`P ⬍ 0.05).
`
`(P ⬍ 0.01), the Isc response to guanylin (1.0 µM) was
`significantly reduced at acidic pH compared with pH
`7.4 (P ⫽ 0.05). In contrast, the Isc responses to STa (20
`nM) did not appear to be pH dependent. The Isc
`responses induced by guanylin or STa were incom-
`pletely inhibited by serosal bumetanide treatment (25–
`35% at pH 7.4; 50% at acidic pH) (data not shown). Gt in
`STa-treated or guanylin-treated groups under acidic
`intraluminal conditions was increased by 3–4 mS/cm2,
`whereas at pH 7.4, ⌬Gt in this period was ⬃0.5–1.0
`mS/cm2.
`⫺ secretion by urogua-
`Stimulation of Cl⫺ and HCO3
`nylin. As stated, a major fraction of the Isc stimulated
`by uroguanylin across the proximal duodenum bathed
`in KRB solution (pH 7.4) was insensitive to serosal
`bumetanide treatment (Fig. 3B). Therefore, ion substi-
`tution experiments were used to investigate the ionic
`basis of the Isc stimulated by 1.0 µM uroguanylin (Fig.
`⫺-free Ringer solution containing the car-
`7). In HCO3
`bonic anhydrase inhibitor methazolamide (1.0 mM),
`the uroguanylin-induced Isc in proximal duodenum was
`completely abolished by bumetanide. In contrast, in
`⫺, the urogua-
`Cl⫺-free Ringer solution containing HCO3
`nylin-induced Isc was markedly reduced but bumetanide
`had no effect on the stimulated Isc. In the absence of
`⫺/CO2 and Cl⫺, uroguanylin had no stimula-
`both HCO3
`tory effect on the Isc. Amiloride (0.1 mM, luminal bath)
`pretreatment of the proximal duodenum to inhibit
`electrogenic Na⫹ absorption also did not affect the
`uroguanylin-induced Isc (data not shown).
`Taken together, these findings suggest that urogua-
`⫺ secretory currents
`nylin stimulates both Cl⫺ and HCO3
`across the duodenum bathed in physiological Ringer
`solution, similar to guanylin action in the rat (23). To
`⫺ (base) secretion directly,
`estimate changes in HCO3
`pH-stat studies of proximal duodenum under voltage-
`
`10
`
`HCO3-free
`
`HCO3/CI-free
`
`Cl-free
`Ringer Solution
`Fig. 7. Effect of anion-substituted Ringer solutions on Isc response to
`uroguanylin in intact mouse proximal duodenum. Sodium bicarbon-
`ate was replaced with TES. Gluconate was substituted for Cl⫺ in
`⫺/Cl⫺-free solutions. Methazolamide (1.0 mM) was
`Cl⫺-free and HCO3
`added to both luminal and serosal bath solutions 20 min before
`⫺-free Ringer solutions.
`luminal uroguanylin application in HCO3
`Bumetanide (0.1 mM) was added to serosal bath 30 min after
`uroguanylin. Change from basal Isc was calculated as pretreatment
`Isc minus maximal Isc recorded after uroguanylin or 10 min after
`bumetanide treatment. Values are means ⫾ SE; proximal duodena
`were obtained from 6–9 different mice in each treatment group.
`*Significantly different from uroguanylin-treated group (paired t-
`test, P ⬍ 0.01).
`
`clamp conditions were used to measure the effect of
`⫺
`HCO3
`. The values of Isc in Table 1 are
`uroguanylin on J s=l
`given in microequivalents per square centimeter per
`⫺
`HCO3
`. As shown
`hour to facilitate the comparison with J s=l
`in Table 1, uroguanylin (1.0 µM) at pH 7.4 stimulated
`⫺
`HCO3
`and the Isc. At luminal pH 5.15, the basal
`the J s=l
`⫺
`HCO3
`and Isc were larger than the basal param-
`mean J s=l
`eters measured at pH 7.4 (P ⬍ 0.05, unpaired t-test). In
`part, this difference was due to spontaneous increases
`⫺
`HCO3
`and Isc with the concomitant appear-
`in both the J s=l
`ance of amorphous material (mucus) in the luminal
`bath during the application of pH 5.15. Several solution
`changes of the luminal bath proved to be useful in
`minimizing the spontaneous increases in the measured
`
`Table 1. Effect of luminal pH and uroguanylin on
`⫺
`HCO3
`and Isc across murine proximal duodenum
`J s=l
`
`Luminal pH 7.4
`Basal
`Uroguanylin
`⌬ (Basal ⫺ uroguanylin)
`Luminal pH 5.15
`Basal
`Uroguanylin
`⌬ (Basal ⫺ uroguanylin)
`
`⫺
`HCO3
`J s=l
`µeq · cm⫺2 · h⫺1
`
`Isc ,
`µeq · cm⫺2 · h⫺1
`
`0.78 ⫾ 0.07
`2.27 ⫾ 0.29*
`1.48 ⫾ 0.22
`
`1.82 ⫾ 0.13
`3.94 ⫾ 0.17*
`2.08 ⫾ 0.08†
`
`⫺1.28 ⫾ 0.55
`⫺3.80 ⫾ 0.81*
`⫺2.52 ⫾ 0.28
`
`⫺2.49 ⫾ 0.20
`⫺7.30 ⫾ 0.77*
`⫺4.81 ⫾ 0.91†
`
`⫺
`HCO3
`Values are means ⫾ SE (n ⫽ 4). J s=l
`, serosal-to-luminal flux of
`⫺. *Significantly different from basal value (paired t-test,
`HCO3
`P ⬍0.05). †Significantly different from pH 7.4 (unpaired t-test,
`P ⬍0.05).
`
`Downloaded from journals.physiology.org/journal/ajpgi (165.001.202.250) on February 23, 2022.
`
`
`
`PHYSIOLOGICAL FUNCTIONS OF UROGUANYLIN
`
`G639
`
`parameters. Nevertheless, subsequent application of
`uroguanylin at pH 5.15 produced pronounced increases
`⫺
`HCO3
`and Isc. The magnitude of the changes at
`in both J s=l
`pH 5.15 was significantly greater than the uroguanylin-
`⫺
`HCO3
`and Isc in duodena bathed at
`induced changes of J s=l
`pH 7.4. This finding was particularly significant be-
`cause it eliminated the possibility that the increased
`action of uroguanylin at acidic pH was the result of
`⫺ se-
`increasing the electrochemical gradient for HCO3
`cretion. In the initial studies (see Figs. 3 and 4), the
`luminal pH was reduced by adding HCl directly to the
`⫺-buffered KRB. This maneuver essentially re-
`HCO3
`⫺ from the luminal solution and thereby
`moves HCO3
`⫺ secre-
`increases the electrochemical gradient for HCO3
`tion. However, all pH-stat studies were performed in
`⫺; thus the electrochemical
`the absence of luminal HCO3
`⫺ efflux across the luminal mem-
`driving force for HCO3
`brane was essentially equivalent at both pH 5.15 and
`7.4. The total Gt of the proximal duodenum exposed to
`either pH 7.4 or 5.15 increased during uroguanylin
`treatment, but the changes were not significantly differ-
`ent from each other (⌬Gt ⫽ 20.3% at pH 7.4 and 26.3%
`at pH 5.15; not significant). The increased Gt was
`greater than that recorded at the peak ⌬Isc in the
`earlier experiments, but this may be a reflection of the
`longer experimental period. Furthermore, inspection of
`the data from the individual experiments did not reveal
`⫺
`HCO3
`, i.e.,
`an obvious correlation between Gt and J s=l
`greater increases in Gt were not associated with propor-
`⫺
`HCO3
`.
`tional increases in J s=l
`Uroguanylin stimulates Isc in duodenum of CF mice.
`Although it is well documented that intestinal epithelia
`from CFTR knockout mice have virtually no Isc re-
`sponse to agents that elevate intracellular cAMP or
`cGMP (4, 7, 21), it was of interest to examine the effects
`of uroguanylin on the Isc of proximal duodenum in
`CFTR knockout mice. Surprisingly, as shown in Fig.
`8A, uroguanylin (1.0 µM) resulted in a small but
`reproducible increase in Isc of 7.7 ⫾ 0.6 µA/cm2 (mean ⫾
`SE) in the proximal duodenum of cystic fibrosis (CF)
`mice. Basolateral bumetanide (0.1 mM) application did
`not inhibit the Isc response elicited by uroguanylin.
`Luminal glucose addition stimulated the Isc similarly
`across the duodenum from CF compared with normal
`mice. Dose-dependent increases in Isc by uroguanylin in
`the proximal duodenum of CFTR knockout mice are
`shown in Fig. 8B. A maximal response of 13.5 ⫾ 0.2
`µA/cm2 was elicited by 10 µM uroguanylin.
`Uroguanylin/guanylin receptors in the murine duode-
`num and other intestinal segments. To characterize
`uroguanylin, guanylin, and STa receptors in the mouse
`intestine, we used 125I-STa receptor autoradiography
`and an in vitro cGMP bioassay to estimate the density
`of functional receptors in the intestinal mucosa. 125I-
`STa binding was apparent in both villus and crypt
`epithelia of the intestinal preparations (Fig. 9, A and
`B). In contrast, 125I-STa binding sites were not found in
`the tissue layers of the intestinal wall, including the
`lamina propria, submucosal elements, or smooth muscle
`
`TTX
`
`pH 7.4 Uroguanylin
`
`1 1 1
`
`Bumet GM
`
`11
`
`A
`
`N
`E O
`...4
`O U)
`
`-80
`
`-60
`
`-40
`
`-20
`
`0
`
`0
`
`20
`
`60
`40
`Time (min)
`
`80
`
`100
`
`B
`
`16-
`
`12.5 -
`
`fir
`
`M
`
`10-
`
`eT
`E 0
`..,
`...
`0
`cn
`-,t1 5.0-
`
`7.5
`
`2.5
`
`1.0 µM
`
`10 µM
`
`3.0 µM
`[UROGUANYLIN]
`Fig. 8. Effect of uroguanylin on Isc across proximal duodenum from
`cystic fibrosis transmembrane conductance regulator (CFT