Photoaffinity Labeling of the Epithelial Sodium Channel*

Vol. 261, No . 6 , Issue of February 25, pp. 2839-2843,1986 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 01986 by The American Society of Biological Chemists, Inc.

Photoaffinity Labeling of the Epithelial Sodium Channel* (Received for publication, June 27, 1985)

Thomas R. KleymanS, Teresa Yulo, Cameron Ashbaugh,Donald Landry, Edward Cragoe, Jr.& Arthur Karlin, and &ais Al-Awqati From the Departments of Medicine, Physiology, Neurology, and Biochemistry,College of Physicians and Surgeons, Columbia University, New York, New York 10032 and CMerck, Sharp and Dohme Research Laboratories, West Point, Pennsylvania19486

Sodium enters tight epithelia across the apical plasma membrane through a sodiumchannel, a process inhibited by submicromolar concentrations of amiloride and benzamil. Using membrane vesicles from bovine kidney cortex, we found that sodium transport through the sodium channel was inhibited by benzamil with anICs0 of 4 nM. Amiloride (ICs0 = 400 nM) was a weaker inhibitor of sodium transport. [3H]Benzamil bound to the vesicles at a single class of high affinity binding sites witha &of 5 nM, the similarity of which to the ICso suggests that these binding sites are associated with the sodium channel. Amiloride displaced bound [3H]benzamilwith a &of 2,500 nM. Bromobenzamil is a photoactive amiloride analog with potency similar to benzamil in inhibiting sodium transport (ICs0 = 5 nM) and binding to the sodium channel (&= 6 nM). [3H]Bromobenzamil was specifically photoincorporated into three molecular weight classes of polypeptides with apparent M,values of 176,000,77,000, and 47,000. The photoincorporationof [3H]bromobenzamil into these three classes of polypeptides was blocked by addition of excess benzamil and by amiloride in a dosedependent manner. These data suggest that these polypeptides are components of the epithelial sodium channel.

the hormones aldosterone and vasopressin (1, 7, 8). Changes in net transepithelial sodium transport are due, at least in part, tochanges in thenumber and/or probability of opening of sodium channels in the apical plasma membrane (3,643). To gain further insight into the mechanisms of regulation of the sodium channel, we must identify and eventually purify itsprotein components. Inthis paper, we used amiloride analogs to characterize the sodium channel present in membrane vesicles obtained from bovine kidney cortex and used a photoactive amiloride analog to identify components of the channel. A preliminary report of this work has appeared (9). EXPERIMENTALPROCEDURES

Materials-Amiloride, benzamil, bromobenzamil, [ben~yl-~HIbenzamil, and [benzyl-3H]bromobenzamilwere prepared at Merck, Sharp and Dohme Research Laboratories by methods previously described (10). Carrier-free **NaClwas obtained from Amersham Corp.All other compounds used were reagent grade. Preparation of Bovine Kidney Cortical Membrane Vesicles-Fresh bovine kidneys were obtained from a kosher slaughterhouse and kept on ice. All subsequent steps were performed at 4 “C. Cortical tissue was removed, finely minced, and washed twice in a buffer containing 100 mM KC1, 1 mM EDTA, 10 mM KPO, (pH 7.4), 0.1 mM phenylmethylsulfonyl fluoride, and 1.0 pg/ml pepstatin A (homogenizing buffer). The tissue was suspended 1:lO (w/v) in homogenizing buffer and homogenized with 10 strokes in a Teflon-glass homogenizer at 800 rpm. The homogenate was spun at 1,000 X g for 10 min to pellet unbroken cells and nuclear debris. The supernatantwas spun at 6,000 X g for 5 min to pellet mitochondria and again at 15,000 X g for 20 The plasma membrane of epithelial cells contains distinct min. The supernatantand loose white outer pellet were then spun at apical and basolateral domains, whose series arrangement 40,000 X g for 1 h. The pellet was suspended in homogenizing buffer allows vectorial transport of solutes across the cell. In a “tight” and spun at 40,000 X g for 30 min. The final pellet was resuspended epithelium, sodium passively enters the cell across the apical in the same buffer and stored at -20 “C. Aliquots were thawed and for binding studies. plasma membrane through a sodium-selective channel, which used To measure sodium influx, we prepared vesicles using a procedure is inhibited by submicromolar concentrations of the diuretic similar to thatdescribed above with the following modifications. The amiloride and is electrophysiologically and pharmacologically buffer used contained 55 mM NaCl, 87.5 mM sucrose, 2.5 mM EDTA, distinct from the voltage-gated sodium channel. Sodium exits 12.5 mM imidazole (pH 7.0), 0.1 mM phenylmethylsulfonyl fluoride, the cell across the basolateral membrane through the Na,K- and 1.0 pg/ml pepstatin A (transport buffer). Following the spin a t ATPase (1).The amiloride-sensitive sodium channel is pres- 1,000 X g for 10 min, the supernatant was spun at 10,000 X g for 5 ent inepithelia from a variety of sources, including the kidney min. The supernatant and loose white outer pellet were spun at 30,000 X g for 1 h. The pellet was resuspended in transport buffer cortical collecting tubule,distal colon, trachea,skin,and and spun again at 40,000 X g for 30 min. The pellet was suspended urinary bladder (1,2). In these tight epithelia,the rate of net in buffer, stored at 4 “C, and used within 24 h for transport studies. transepithelial sodium transport is influenced by a variety of Transport Assay-Sodium transport in bovine kidney cortical factors including the activity of sodium in themedium bathing membrane vesicles was measured with a modification of the method the apical plasma membrane (3), the intracellular sodium of Garty et al. (ll),as described in the legend of Fig. 1. Binding Studies-binding of [3H]benzamil(4.7 Ci/mmol) to bovine activity (4)and calcium activity ( 5 ) , cell metabolism (6), and kidney cortical membrane vesicles wasassayed by equilibrium dialysis * This research was supported by Grants AM20999 and AM34742 with 12,000-14,000 molecular weight cutoff dialysis tubing. Memfrom the United States Public Health Service. The costs of publica- brane vesicles, containing approximately 250 pg of protein, and [3H] tion of this article were defrayed in part by the payment of page benzamil were placed in the dialysis tubing, which was then placed charges. This article must therefore be hereby marked “advertise- in atest tube with 7.5 ml of homogenizingbuffer containing the same ment” in accordance with 18 U.S.C. Section 1734 solely to indicate concentration of [3H]benzamil.The tubes were stirred on a flatbed rotary mixer (160 rpm) for at least 16 h at 4 “C, by which time this fact. $Supported by a Clinician-Scientist Award from the American equilibrium had been achieved (data not shown). Aliquots were reHeart Association with funds provided in parthy the New York Heart moved from the dialysis bag and dialysate to determine the [3H] Association. benzamil concentrations. Protein determinations were performed by

2839

2840

Photoaffinity Labeling of the Epithelial Sodium Channel

the method of Bradford (12), using bovine serum albumin as the standard. Nonspecific binding was determined in parallel experiments in which 1 p~ unlabeled benzamil was added to the vesicles and dialysate. Photoaffinity Labeling-Bovine kidney cortical membrane vesicles were diluted to a protein concentration of 0.125 mg/ml with homogenizing buffer containing 20nM [3H]bromobenzamil (21.2 Ci/mmol) and incubated for greater than 1 h a t 4 "C. The vesicle suspension (2 ml) was stirred a t room temperature and irradiated for 15 min with light from a mercury arc lamp (Zeiss HBO, 50 watts) filtered through 1345-nm long pass and 300-400-nm band pass filters (Oriel Corp., Stratford, CT). Two filters were used to eliminate all wavelengths