I. Novak

Properties of the luminal membrane of isolated perfused rat pancreatic ducts

The aim of the present study was to investigate by what transport mechanism does HCO3− cross the luminal membrane of pancreatic duct cells, and how do the cells respond to stimulation with dibytyryl cyclic AMP (db-cAMP). For this purpose a newly developed preparation of isolated and perfused intra-and interlobular ducts of rat pancreas was used. Responses of the epithelium to inhibitors and agonists were monitored by electrophysiological techniques. Addition of HCO3−/CO2 to the bath side of nonstimulated ducts depolarized the PD across the basolateral membrane (PDbl) by about 9mV, as also observed in a previous study [21]. This HCO3− effect was abolished by Cl− channel blockers or SITS infused into the lumen of the duct: i. e. 5-nitro-2-(3-phenylpropylamino)-benzoic acid (NPPB, 10−5 M) hyperpolarized PDbl by 8.2±1.6 mV (n=13); 3′,5-dichlorodiphenylamine-2-carboxylic acid (DCl-DPC, 10−5 M) hyperpolarized PDbl by 10.3±1.7 mV (n=10); and SITS hyperpolarized PDbl by 7.8±0.9 mV (n=4). Stimulation of the ducts with dbcAMP in the presence of bath HCO3−/CO2 resulted in depolarization of PDbl, the ductal lumen became more negative and the fractional resistance of the luminal membrane decreased. Together with forskolin (10−6 M), db-cAMP (10−4 M) caused a fast depolarization of PDbl by 33.8±2.5 mV (n=6). When db-cAMP (5×10−4 M) was given alone in the presence of bath HCO3−/CO2, PDbl depolarized by 25.3±4.2 mV (n=10). In the absence of exogenous HCO3−, db-cAMP also depolarized PDbl by 24.7±3.0 mV (n=10). The present data suggest that in the luminal membrane of pancreatic duct cells there is a Cl− conductance in parallel with a Cl−/HCO3− antiport. Dibutyryl cyclic AMP increases the Cl− conductance of the luminal membrane. Taking together our present results, and the recent data obtained for the basolateral membrane [21], a tentative model for pancreatic HCO3− transport is proposed.

Electrophysiological study of transport systems in isolated perfused pancreatic ducts: properties of the basolateral membrane.

In order to study the mechanism of pancreatic HCO3− transport, a perfused preparation of isolated intra-and interlobular ducts (i.d. 20–40 μm) of rat pancreas was developed. Responses of the epithelium to changes in the bath ionic concentration and to addition of transport inhibitors was monitored by electrophysiological techniques. In this report some properties of the basolateral membrane of pancreatic duct cells are described. The transepithelial potential difference (PDte) in ducts bathed in HCO3−-free and HCO3−-containing solution was −0.8 and −2.6 mV, respectively. The equivalent short circuit current (Isc) under similar conditions was 26 and 50 μA·cm−2. The specific transepithelial resistance (Rte) was 88 Ωcm2. In control solutions the PD across the basolateral membrane (PDbl) was −63±1 mV (n=314). Ouabain (3 mmol/l) depolarized PDbl by 4.8±1.1 mV (n=6) within less than 10 s. When the bath K+ concentration was increased from 5 to 20 mmol/l, PDbl depolarized by 15.9±0.9 mV (n=50). The same K+ concentration step had no effect on PDbl if the ducts were exposed to Ba2+, a K+ channel blocker. Application of Ba2+ (1 mmol/l) alone depolarized PDbl by 26.4±1.4 mV (n=19), while another K+ channel blocker TEA+ (50 mmol/l) depolarized PDbl only by 7.7±2.0 mV (n=9). Addition of amiloride (1 mmol/l) to the bath caused 3–4 mV depolarization of PDbl. Furosemide (0.1 mmol/l) and SITS (0.1 mmol/l) had no effect on PDbl. An increase in the bath HCO3− concentration from 0 to 25 mmol/l produced fast and sustained depolarization of PDbl by 8.5±1.0 mV (n=149). It was investigated whether the effect of HCO3− was due to a Na++-dependent transport mechanism on the basolateral membrane, where the ion complex transferred into the cell would be positively charged, or whether it was due to decreased K+ conductance caused by lowered intracellular pH. Experiments showed that the HCO3− effect was present even when the bath Na+ concentration was reduced to a nominal value of 0 mmol/l. Similarly, the HCO3− effect remained unchanged after Ba2+ (5 mmol/l) was added to the bath. The results indicate that on the basolateral membrane of duct cells there is a ouabain sensitive (Na++K+)-ATPase, a Ba2+ sensitive K+ conductance and an amiloride sensitive Na+/H+ antiport. The HCO3− effect on PDbl is most likely due to rheogenic anion exit across the luminal membrane.

Two independent anion transport systems in rabbit mandibular salivary glands

Cholinergically stimulated Cl and HCO3 transport in perfused rabbit mandibular glands has been studied with extracellular anion substitution and administration of transport inhibitors. (i) In glands perfused with HCO3-free solutions, replacement of Cl with other anions supported secretion in the following sequence: Br=>Cl>I=>NO3>isethionate. Furosemide, 1.0 and 0.1 mmol/l, inhibited Cl-supported secretion by 97–99% and 70–78%, respectively. SITS, 0.1 mmol/l, had no effect and amiloride, 1.0 mmol/l, caused a 55–65% inhibition. Addition of SITS to amiloride-treated glands produced no further effect. (ii) In glands perfused with Cl-free solutions, but containing 25 mM HCO3, amiloride, 1.0 mmol/l, inhibited secretion by 95% and methazolamide, 0.1 mmol/l, by 55%. (iii) In glands perfused with solutions containing both HCO3 and Cl, furosemide had smaller effects than in glands perfused with solutions containing only Cl — a dose of 1.0 mmol/l inhibited 60% of the initial fast phase of secretion, and 90% of the later plateau phase, while a dose of 0.1 mmol/l inhibited 30% of the initial phase, but had no effect on the plateau. SITS, 0.1 mmol/l, actually stimulated secretion by about 30%, but when infused in addition to furosemide (0.1 mmol/l), it inhibited by about 20%. Amiloride (1.0 mmol/l) caused no inhibition.The results suggest that there are at least three distinct carriers in the rabbit mandibular gland. One is a furosemidesensitive Na-coupled Cl (probably Na−K−2Cl) symport, responsible for the bulk of normal secretion. The others are an amiloride-sensitive Na−H antiport and a SITS-sensitive Cl−HCO3 antiport.

Effect of bicarbonate on potassium conductance of isolated perfused rat pancreatic ducts

The aim of this study was to investigate the role of the K+ conductance in unstimulated and stimulated pancreatic ducts and to see how it is affected by provision of exogenous HCO3−/CO2. For this purpose we have applied electrophysiological techniques to perfused pancreatic ducts, which were dissected from rat pancreas. The basolateral membrane potential PDbl of unstimulated duct cells was between −60mV and −70mV, and the cells had a relatively large K+ conductance in the basolateral membrane as demonstrated by (a) 20–22 mV depolarization of PDbl in response to increase in bath K+ concentration from 5 mmol/l to 20mmol/l and (b) the effect of a K+ channel blocker, Ba2+ (5 mmol/l), which depolarized PDbl by 30–40mV. These effects on unstimulated ducts were relatively independent of bath HCO3−/CO2. The luminal membrane seemed to have no significant K+ conductance. Upon stimulation with secretin or dibutyryl cyclic AMP, PDbl depolarized to about −35 mV in the presence of HCO3−/CO2. Notably, the K+ conductance in the stimulated ducts was now only apparent in the presence of exogenous HCO3−/CO2 in the bath solutions. Upon addition of Ba2+, PDbl depolarized by 13±1 mV (n=7), the fractional resistance of the basolateral membrane, FRbl increased from 0.66 to 0.78 (n=6), the specific transepithelial resistance, Rte, increased from 52±13 Ω cm2 to 59±15 Ω cm2 (n=11), and the whole-cell input resistance, Rc, measured with double-barrelled electrodes, increased from 20 MΩ to 26 MΩ (n=3). These results are consistent with Ba2+ inhibition of the K+ conductance. Following removal of exogenous HCO3−/CO2 in the same ducts, stimulation led to a larger depolarization on PDbl to about −25 mV, and Ba2+ had a smaller effect on PDbl and no significant effect on the resistances. The individual resistances in the duct epithelium were estimated from equivalent circuit analysis. The luminal membrane resistance, R1 decreased from about 2000 Ω cm2 to 80 Ω cm2 upon stimulation. The basolateral membrane resistance, Rbl, remained at 90–120 Ω cm2, and the paracellular shunt resistance, Rs, at 50–80 Ω cm2. Ba2+ increased Rbl of stimulated ducts to about 200 Ω cm2, an effect present only if the ducts were provided with exogenous HCO3−/CO2. Taken together, the present results indicate that the basolateral K+ conductance of pancreatic ducts is sensitive to exogenous HCO3−/CO2, i.e. without HCO3−/CO2 the conductance becomes very low although the ducts are undergoing stimulation.

Effect of ATP, carbachol and other agonists on intracellular calcium activity and membrane voltage of pancreatic ducts.

The pancreatic duct has been regarded as a typical cAMP-regulated epithelium, and our knowledge about its Ca2+ homeostasis is limited. Hence, we studied the regulation of intracellular calcium, [Ca2+]i, in perfused rat pancreatic ducts using the Ca2+-sensitive probe fura-2. In some experiments we also measured the basolateral membrane voltage, Vbl, of individual cells. The resting basal [Ca2+]i was relatively high, corresponding to 263±28 nmol/l, and it decreased rapidly to 106±28 nmol/l after removal of Ca2+ from the bathing medium (n=31). Carbachol increased [Ca2+]i in a concentration-dependent manner. At 10 μmol/l the fura-2 fluorescence ratio increased by 0.49±0.06 (n=24), corresponding to an increase in [Ca2+]i by 111±15 nmol/l (n=17). ATP, added to the basolateral side at 0.1 mmol/l and 1 mmol/l, increased the fluorescence ratio by 0.67±0.06 and 1.01±14 (n=46; 12), corresponding to a [Ca2+]i increase of 136±22 nmol/l and 294±73 nmol/l respectively (n= 15; 10). Microelectrode measurements showed that ATP (0.1 mmol/l) hyperpolarized Vbl from −62±3 mV to-70±3 mV, an effect which was in some cases only transient (n=7). This effect of ATP was different from that of carbachol, which depolarized Vbl. Applied together with secretin, ATP delayed the secretin-induced depolarization and prolonged the initial hyperpolarization of Vbl (n=4). Several other putative agonists of pancreatic HCO3−secretion were also tested for their effects on [Ca2+]i. Bombesin (10 nmol/l) increased the fura-2 fluorescence ratio by 0.24±0.04 (n=8), neurotensin (10 nmol/l) by 0.25±0.04 (n=6), substance P (0.1 μmol/l) by 0.22±0.06 (n=6), and cholecystokinin (10 nmol/l) by 0.14±0.03 (n=7). Taken together, our studies show that Ca2+ homeostasis plays a role in pancreatic ducts. The most important finding is that carbachol and ATP markedly increase [Ca2+]i, but their different electrophysiological responses indicate that intracellular signalling pathways may differ.

Effect of secretin and inhibitors of HCO3−/H+ transport on the membrane voltage of rat pancreatic duct cells

The aim of the present study was to study the effect of secretin on the electrophysiological response of pancreatic ducts. Furthermore, we investigated the effects of lipid-soluble buffers and inhibitors of HCO3−/H+ transport. Ducts obtained from fresh rat pancreas were perfused in vitro. Secretin depolarized the basolateral membrane voltage, Vbl, by up to 35 mV (n=37); a halfmaximal response was obtained at 3×10−11 mol/l. In unstimulated ducts a decrease in the luminal Cl− concentration (120 to 37 mmol/l) had a marginal effect on Vbl, but after maximal secretin stimulation it evoked a 14±2 mV depolarization (n=6), showing that a luminal Cl− conductance GCl- was activated. The depolarizing effect of secretin on Vbl was often preceded by about a 6 mV hyperpolarization, most likely due to an increase in the basolateral GK+. Perfusion of ducts with DIDS (4,4′ — diisothiocyanatostilbene — 2,2′ — disulphonic acid, 0.01 mmol/l) or addition of ethoxzolamide (0.1 mmol/l) to the bath medium diminished the effect of secretin. Acetate or pre-treatment of ducts with NH4+/NH3 (10 mmol/l in the bath) depolarized the resting Vbl of −65±2 mV by 16±4 mV (n=7) and 19±3 mV (n=10), respectively. The fractional resistance of the basolateral membrane (FRbl) doubled, and the depolarizing responses to changes in bath K+ concentrations (5 to 20 mmol/l) decreased from 22±1 to 11±2 mV. The Na+/H+ antiporter blocker EIPA (5-[N-ethyl-N-isopropyl]-amiloride, 0.1 mmol/l) also depolarized Vbl by 10±1 mV, FRbl increased and the response to K+ concentration changes decreased (n=7). Effects of EIPA and ethoxzolamide on Vbl were greater in ducts deprived of exogenous HCO3−/CO2. Taken together, the present study shows that secretin increased the basolateral GK+ and the luminal GCl-. The depolarizing effect of secretin was diminished following inhibition of HCO3− transport (DIDS), or HCO3−/H+ generation (ethoxzolamide). Manoeuvres that presumably led to lowered intracellular pH (NH4+/NH3 removal, acetate, EIPA) decreased the basolateral GK+. The present data support our previously published model for pancreatic HCO3− secretion, and indicate that the basolateral membrane possesses a pH-sensitive GK+.

Calcium influx pathways in rat pancreatic ducts

A number of agonists increase intracellular Ca2+ activity, [Ca2+]i, in pancreatic ducts, but the influx/efflux pathways and intracellular Ca2+ stores in this epithelium are unknown. The aim of the present study was to characterise the Ca2+ influx pathways, especially their pH sensitivity, in native pancreatic ducts stimulated by ATP and carbachol, CCH. Under control conditions both agonists led to similar changes in [Ca2+]i. However, these Ca2+ transients, consisting of peak and plateau phases, showed different sensitivities to various experimental manoeuvres. In extracellular Ca2+-free solutions, the ATP-induced [Ca2+]i peak decreased by 25%, but the CCH-induced peak was unaffected; both plateaus were inhibited by 90%. Flufenamate inhibited the ATP-induced peak by 35%, but not the CCH-evoked peak; the plateaus were inhibited by 75-80%. La3+ inhibited the ATP-induced plateau fully, but that induced by CCH by 55%. In resting ducts, an increase in extracellular pH, pHe, by means of HEPES and HCO3/CO2 buffers, increased [Ca2+]i; a decrease in pHc had the opposite effect. In stimulated ducts the pH-evoked effects on Ca2+ influx were more pronounced and depended on the agonist used. At pHe 6.5 both ATP- and CCH-evoked plateaus were inhibited by about 50%. At pH 8.0 the ATP-stim-ulated plateau was inhibited by 27%, but that stimulated by CCH was increased by 72%. Taken together, we show that CCH stimulates Ca2+ release followed by Ca2+ influx that is moderately sensitive to flufenamate, La3+, depolarisation, it is inhibited by low pH, but stimulated by high pH. ATP stimulates Ca2+ release and probably an early Ca2+ influx, which is more markedly sensitive to flufenamate and La3+, and is both inhibited by low and high pH. Thus our study indicates that there are at least two separate Ca2+ influx pathways in pancreatic ducts cells.

Effect of vasoactive intestinal peptide, carbachol and other agonists on the membrane voltage of pancreatic duct cells

The regulation of pancreatic exocrine secretion involves hormonal, neural and neurohormonal components. Many agonists are known to be effective in pancreatic acinar cells, but less is known about the ducts. Therefore, we wanted to investigate the influence of various agonists on isolated perfused pancreatic ducts and, as a physiological response, we measured the basolateral membrane voltage of the duct cells (Vbl) with microelectrodes. Pancreatic ducts were dissected from pancreas of normal rats and bathed in a HCO3−-containing solution. Under control conditions, the average Vbl was between -50 and -70 mV. Vasoactive intestinal peptide (VIP) and carbachol (CCH) reversibly depolarized Vbl when applied to the bath. VIP (9×10−9 mol/l) depolarized Vbl from -72±3 mV to -53±3 mV (n=20) and CCH (10−5 mol/l) from -62±3 to -35±4mV (n=10). Furthermore, a decrease of the Cl− concentration in the lumen led to an increase of VIP-induced depolarization of Vbl, suggesting that a luminal Cl− conductance was increased. Cholecystokinin (CCK, 10−10-10−7 mol/l) and bombesin (10−8, 10−5 mol/l), which stimulate pancreatic exocrine secretion in acini or whole glands, showed no significant effect on Vbl of the duct cells tested in our preparation (n=7, 6). Neurotensin (10−8 mol/l) had a marked depolarizing effect in two out of ten cases; Vbl depolarized from about -65 mV to-29 mV and the effect was reversible. Substance P (2×10−7 mol/l), alone or in combination with secretin, had no effect on Vbl of the tested duct cells (n=11). We propose that the basolateral membrane of pancreatic duct cells possesses receptors for VIP, acetylcholine and neurotensin. CCK, bombesin and substance P had no detectable effects on Vbl of the duct cells tested, which could be due to the lack of corresponding receptors on these cells, or due to the absence of electrophysiologically detectable effects, in spite of receptor presence.

A New Preparation of Pancreatic Ducts for Patch-Clamp Studies

Pancreatic HCO–3 secretion is ascribed to small intra- and interlobular ducts. Although the general transport mechanisms for ducts obtained from the rat pancreas are known, the i

Intracellular pH in Rat Pancreatic Ducts

In order to study the mechanism of H+ and HCO3- transport in a HCO3- secreting epithelium, pancreatic ducts, we have measured the intracellular pH (pHi) in this tissue using the pH sensitive probe BCECF. We found that exposures of ducts to solutions containing acetate/acetic acid or NH4+/NH3 buffers (20 mmol/l) led to pHi changes in accordance with entry of lipid-soluble forms of the buffers, followed by back-regulation of pHi by duct cells. In another type of experiment, changes in extracellular pH of solutions containing HEPES or HCO3-/CO2 buffers led to significant changes in pHi that did not seem to be back-regulated efficiently by duct cells. The sensitivity of pHi to the inhibitor HOE 694 and to changes in Na+ gradients, indicate that the Na+/H+ exchanger is present in this epithelium. Similarly, the sensitivity to Cl- and HCO3- gradients indicated the presence of the Cl-/HCO3- exchanger. Under some conditions, these exchangers can be invoked to regulate cell pH.