Characterization of chloride channels in membrane vesicles from the kidney outer medulla

Chloride channels in the loop of Henle

Cl- transport in the loop of Henle is responsible for reclamation of 25-40% of the filtered NaCl load and for the formation of dilute urine. Our understanding of the physiologic and molecular mechanisms responsible for Cl- reabsorption in both the thin ascending limb and thick ascending limb of Henle's loop has increased greatly over the last decade. Plasma membrane Cl- channels are known to play an integral role in transcellular Cl- transport in both the thin and thick ascending limbs. This review focuses on the functional characteristics and molecular identities of these Cl- channels, as well as the role of these channels in the pathophysiology of disease.

Imidazoline binding sites in human placenta: evidence for heterogeneity and a search for physiological function

1. An alpha 2-adrenoceptor antagonist, idazoxan, that binds to both alpha 2-adrenoceptors and to imidazoline sites (IR), has been used to characterize human placental IR. Human placenta is shown to be the richest source of IR (1800 +/- 100 fmol mg-1 protein; Kd 38.9 +/- 3.4 nM). 2. Primary cells derived from human placenta and grown in monolayers, also displayed a high density of receptors (3209 +/- 136 fmol mg-1 in cytotrophoblasts and 3642 +/- 144 fmol mg-1 protein in syncytiotrophoblast enriched cell culture). 3. [3H]-idazoxan did not show binding characteristics of alpha 2-adrenoceptors in human placental membranes or human trophoblastic cells, thus making it a ligand of choice to study the imidazoline site. The tissue appeared to be lacking alpha 2-adrenoceptors in that other alpha 2-adrenoceptor ligands, [3H]-rauwolscine and [3H]-clonidine, do not bind to alpha 2-adrenoceptors in human placenta. 4. IRs are localized on the cell surface, as determined by the release of bound [3H]-idazoxan from cells, when washed with high ionic/acidic medium. 5. Imidazoline receptors of human placenta display high affinity for amiloride (72 +/- 27 nM). The high affinity was used as a criterion to classify IR to IRa subtype (placenta, rabbit kidney, rabbit liver and rabbit adipose cells) as opposed to the IRb subtype which display low affinity for amiloride (greater than 2 microM, in all the other tissues).6. Several novel ligands comprising a guanido functional group attached to an aromatic residue (e.g. benziliden-amino-guanidine (BAG), guanido pyrole) display pronounced selectivity for IR over the M2-adrenoceptors as the affinity of BAG is about 40 fold higher (Kd= 18.9 +/- 13.8 nM in human placenta), than the affinity for M2-adrenoceptors (Kd = 768 +/- 299 nM in human platelets). Imidazoline sites bind selectively BAG and other guanido ligands thus indicating a distinct structural requirement at its site of binding.7. K+ channel blockers and monovalent ions (e.g. Cs' and NH4+) interfere with idazoxan binding to IR, indicating a possible involvement of IR in K+ transport.

Effects of NPPB (5-nitro-2-(3-phenylpropylamino)benzoic acid) on chloride transport in intestinal tissues and the T84 cell line

NPPB (5-nitro-2-(3-phenylpropylamino)benzoic acid) has been reported to block Cl- channels in isolated rabbit nephrons with high potency (IC50 = 80 nM). The effects of this compound on Cl(-)-mediated transport processes in intestinal tissues have been studied using agonist-stimulated short-circuit current (T84) in Ussing chamber experiments and 36Cl- fluxes in monolayers of a colonic cell line (T84). NPPB inhibited PGE1-stimulated Isc in rabbit distal colon and ileum at concentrations in the range 20 to 100 microM. However, NPPB at the same concentrations also inhibited glucose-stimulated Isc in rabbit ileum, suggesting that its effects were not restricted to those on Cl- transport. Consistent with this, exposure of rabbit distal colon to 100 microM NPPB was found to reduce endogenous ATP levels by 69%, implying that, at these concentrations, NPPB could impair active transport processes by an effect on cellular energy metabolism. Clear evidence for a direct effect of NPPB on epithelial chloride channels was found in studies on Cl- fluxes in T84 cell monolayers. NPPB inhibited VIP-stimulated Cl- uptake into T84 cells with an IC50 of 414 microM. NPPB (1 mM) also inhibited Cl- efflux from pre-loaded cells confirming its effect as a weak Cl- channel blocker in this system.

Characterisation of the effects of anthranilic and (indanyloxy) acetic acid derivatives on chloride transport in membrane vesicles

The effects of the Cl- channel blockers, NPPB, IAA94/95 and a number of related compounds on 36Cl- transport in membrane vesicles from bovine kidney cortex and rabbit ileum mucosa brush borders have been studied. These vesicles have been previously shown to be enriched in Cl- channel and Cl-/anion cotransport activity, respectively. Chloride transport was assayed in both types of vesicles by measuring the uptake of 36Cl- in response to an outwardly-directed Cl- concentration gradient. In kidney microsomes, a large proportion of the observed 36Cl- uptake was mediated by an electrogenic uniport and could be substantially reduced by clamping the membrane potential at zero mV using K+ and valinomycin. Chloride uptake was inhibited by both NPPB and IAA94/95 with apparent IC50 values of around 10 microM under optimal conditions (i.e., 4 min uptake at 4 degrees C). Under other conditions (e.g., 10 min uptake at 25 degrees C), where uptake had reached a steady-state level, much higher concentrations of inhibitor were required to cause inhibition. Therefore, previous differences in the reported potency of these compounds may, in part, have been due to the conditions under which Cl- uptake was measured. In addition, both NPPB and, to a lesser extent, IAA94/95 were found to have other effects on the vesicles, in that, when added at a concentration of 100 microM, they induced a leakage of pre-accumulated 36Cl-. This was probably caused by either dissipation of membrane potential or damage to the vesicle membranes. The sulphonic acid derivatives of NPPB and IAA94/95 (NPPB-S and ISA94/95, respectively) blocked 36Cl- uptake with around the same potency as NPPB and IAA94/95, but did not cause any non-specific Cl- leakage, when added at concentrations up to 100 microM. Inhibition of 36Cl- uptake by all four compounds was almost completely reversible. However, when vesicles were incubated with the inhibitors in the presence of an outward Cl- concentration gradient, or if vesicles were freeze/thawed in the presence of the compounds, inhibition could be only partially reversed. In rabbit brush border membrane vesicles, 36Cl- uptake was not reduced when the vesicles were voltage clamped using valinomycin and K+, and was therefore probably mediated by Cl-/Cl- exchange. However, despite the lack of effect of valinomycin, 36Cl- uptake was inhibited by both NPPB (approx. 80% inhibition at 100 microM) and, to a lesser extent, by IAA94/95 (approx. 30% inhibition at 100 microM).(ABSTRACT TRUNCATED AT 400 WORDS)

Cl- channels in basolateral renal medullary membranes: III. Determinants of single-channel activity

We evaluated the effects of varying aqueous Cl- concentrations, and of the arginyl- and lysyl-specific reagent phenylglyoxal (PGO), on the properties of Cl- channels fused from basolaterally enriched renal medullary vesicles into planar lipid bilayers. The major channel properties studied were the anion selectivity sequence, anionic requirements for channel activity, and the effects of varying Cl- concentrations and/or PGO on the relation between holding voltage (VH, mV) and open-time probability (Po). Reducing cis Cl- concentrations, in the range 50-320 mM, produced a linear reduction in fractional open time (Po) with a half-maximal reduction in Po at cis Cl- approximately 170 mM. Channel activity was sustained by equimolar replacement of cis Cl- with F-, but not with impermeant isethionate. For trans solutions, the relation between Cl- concentration and Po was negatively cooperative, with 50% reduction in po at 10 mM Cl-. Reducing cis Cl- had no effect on the gating charge (Z) for channel opening, but altered significantly the voltage-independent energy (delta G) for channel opening. Phenylglyoxal (PGO) reduced Z and altered delta G for Cl- channel activity when added to cis, but not trans solutions. Furthermore, in the presence of cis PGO, reducing the cis Cl- concentration had no effect on Z but altered delta G. Thus we propose that cis PGO and cis Cl- concentrations affect separate sites determining channel activity at the extracellular faces of these Cl- channels.

Purification of a stilbene sensitive chloride channel and reconstitution of chloride conductivity into phospholipid vesicles

A protein conferring passive chloride permeability was isolated from a N-octylglucoside solubilized extract of partially purified H(+)-transporting osteoclast cell membranes. Purification was achieved by binding of solubilized protein to an amine-linked 4,4'-diisothiocyanatostilbene-2,2'-disulfonate (DIDS) Sepharose 4B column and elution with 50 mM KCl. A major protein, with MR = 60 kD on 10% SDS-PAGE, was obtained, which was further purified to homogeneity by HPLC gel filtration. This protein introduced 36Cl- permeability when reconstituted in phospholipid membranes by equilibrium dialysis. The Cl- transport recovered in reconstituted membranes retained sensitivity to DIDS confirming the identity of the isolated protein as a stilbene-sensitive chloride channel.

NPPB inhibits the basolateral membrane K+ conductance in the isolated bullfrog cornea

The effects of the Cl- channel blocker, 4-nitro-2-(3-phenylpropylamino)benzoate (NPPB) on active transepithelial Cl- transport were measured in the isolated bullfrog cornea. With a Cl(-)-free Ringers, stromal-side 10(-5) M NPPB elicited a maximum depolarization of the membrane voltage from -72 +/- 6 to -48 +/- 9 mV (n = 6, P less than 0.05) and reduced the magnitude of the depolarization induced by a 10-fold increase in K+ concentration. Subsequent exposure to 10(-4) M ouabain decreased the membrane voltage from -41 +/- 6 mV to -25 +/- 2 mV (n = 6, P less than 0.05). After stimulation with 10(-5) M amphotericin B of a short-circuit current, Isc, largely accounted for by tear to stroma K+ diffusion, this Isc was effectively inhibited by 10(-5) M NPPB on the stromal-side. This decrease reflected a fall in basolateral membrane K+ conductance. In NaCl Ringers, inhibition of the essentially Cl(-)-originated Isc either on the tear- or stromal-sides required instead 10(-4) M NPPB. NPPB depolarized the membrane voltage from -55 +/- 7 to -38 +/- 6 mV (n = 14, P less than 0.05). The direction of the change in the fractional apical membrane resistance (fRo) depended upon its initial value; in those corneas with a lower value it increased whereas if they had a higher fRo, 10(-4) M NPPB consistently caused fRo to fall. However, following exposure to 5 x 10(-3) M Ba2+ and a fall in fRo, NPPB consistently caused fRo to increase significantly from 30 +/- 8 to 53 +/- 4% (n = 5). Therefore, inhibition of active Cl- transport by 10(-4) M NPPB may be associated with declines in: (1) a basolateral membrane K+ conductance that is distinct from a Ba2(+)-sensitive pathway; (2) an apical membrane Cl- conductance. Neither of these effects may be the result of a direct effect of NPPB on a conductance pathway because: (1) the drug was equipotent from either bathing solution; (2) following a one hour washout the Isc had not fully recovered to its control value.

Reconstitution of a kidney chloride channel and its identification by covalent labeling

The basolateral membrane of the thick ascending loop of Henle (TALH) of the mammalian kidney is characterized by its high content of Na+/K(+)-ATPase and a Cl- conductance, which function in parallel in salt reabsorption. In order to reconstitute the Cl- channels, TALH membrane vesicles were solubilized in 1% sodium cholate in buffer containing 200 mM KCl, followed by dilution with soybean lipids (final ratio of protein/detergent/lipid of 1:3:15 in mg) and removal of the detergent by gel filtration on Sephadex G-50. Cl- channel activity in the liposomes was determined by a 36Cl- uptake assay where the accumulation of the radioactive tracer against its chemical gradient is driven by the membrane potential (positive inside) generated by an outward Cl- gradient. The 36Cl- uptake by the KCl-loaded liposomes was dependent on the inclusion of membrane protein and was abolished by valinomycin, indicating the involvement of a conductive pathway. It was also inhibited by 36% by 100 microM 4,4'-diisothiocyanostilbene-2,2'-disulfonate (DIDS) and 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB). Solubilization of the Cl- channels in cholate was optimal in the presence of 200 mm KCl, but was found to decrease markedly at low ionic strength. SDS-PAGE analysis of the proteins extracted by cholate at high and low salt concentrations showed that the Cl- channel-containing high KCl extract was enriched in the 96 and 55 kDa alpha- and beta-subunits of the Na+/K(+)-ATPase (the major proteins in the membrane preparation) and several minor protein bands. Treatment of the membrane vesicles with the radioactive analogue of DIDS, [3H]2DIDS, labeled primarily a 65 and a 31 kDa protein. The solubilization of the 31 kDa protein by cholate depended markedly on the ionic strength and thus paralleled the solubilization pattern of Cl- channel activity. Furthermore, the labeling of the 31 kDa protein was prevented by nonradioactive DIDS and by NPPB but not by other compounds, indicating that it may be a Cl- channel component.

Cl- transport in basolateral renal medullary vesicles: I. Cl- transport in intact vesicles

This paper provides the results of studies which characterized conductive 36Cl- flux in basolaterally enriched membrane vesicles prepared from rabbit renal outer medulla. Conductive 36Cl- uptake was studied under two different experimental conditions. In the first, 36Cl- flux was driven by an inside positive voltage created with oppositely directed Cl- and gluconate gradients. In the second, an inwardly direct K+ gradient was used to drive 36Cl- uptake. By these two methods, voltage-sensitive 36Cl- uptake was shown to comprise about 45 and 65%, respectively, of the initial rates of total 36Cl- flux. Separate paired studies demonstrated that the conductive 36Cl- uptake was inhibited by the Cl- channel blocker diphenylamine-2-carboxylate (DPC) with an IC50 for DPC of 154 microM. The voltage-dependent 36Cl- uptake had an activation energy of 6.4 kcal/mole. This 36Cl- conductance had an anion selectivity sequence of I- greater than Cl- greater than or equal to NO3- much greater than gluconate.

Net efflux of chloride from cell suspensions measured with a K+ electrode

Under appropriate conditions (presence of cation ionophores) net KCl efflux measured with a K+ electrode can be used to estimate conductive Cl- fluxes, a sensitive procedure that allows continuous recording. The procedure was tested in human red cells by demonstrating effects of ionophores and of an anion transport inhibitor, and in dissociated MDCK cells by demonstration of cAMP and volume-activated Cl- fluxes.

Inhibition of prostaglandin E2 synthesis by a blocker of epithelial chloride channels

Arginine-vasopressin (AVP) elicits a variety of responses in cultured rat mesangial cells, among them stimulation of prostaglandin biosynthesis and activation of Cl- channels. AVP produced an 11-fold increase over basal levels in prostaglandin E2 release from cultured mesangial cells. This response was completely inhibited by 25 microM indomethacin and 82 +/- 5% inhibited by 25 microM 5-nitro-2-(3-phenylpropylamino)-benzoic acid (NPPB) which is a potent blocker of epithelial Cl- channels. The IC50 for NPPB inhibition of prostaglandin E2 release was 8 microM. Indomethacin and NPPB at 25 microM also inhibited AVP-stimulated cellular accumulation of prostaglandin E2 by 98% and 79 +/- 7% respectively. The inhibitory effect of NPPB was not due to interference with the cellular response to AVP since at 50 microM it did not block AVP-stimulated release of arachidonate metabolites from cells metabolically labeled with [3H]-arachidonic acid. It is suggested that NPPB inhibition of prostaglandin E2 synthesis is at the cyclooxygenase level on the basis of its structural similarity to the fenamic acid type of cyclooxygenase inhibitors.