Adrenergic regulation of salt and fluid secretion in human medullary collecting duct cells

【摘要】  Transepithelial salt and fluid secretion mediated by cAMP in initial inner medullary collecting ducts (IMCD i ) may be important for making final adjustments to urine composition. We examined in primary cultures of human IMCD i cells the effects of adrenergic receptor (AR) agonists and antagonists on intracellular cAMP levels, short-circuit current ( I SC ), and fluid secretion. Epinephrine (1 µM), norepinephrine (1 µM), and isoproterenol (10 nM) individually increased intracellular cAMP levels 57-, 2-, and 25-fold, respectively, and stimulated I SC 3.3-, 2.9-, and 3.4-fold, respectively. -AR activation increased net fluid secretion by cultured human IMCD i cell monolayers from 0.09 ± 0.04 to 0.26 ± 0.05 µl·h -1 ·cm -2 and freshly isolated rat IMCD i from 0.02 ± 0.01 to 0.09 ± 0.02 nl·h -1 ·mm -1. In monolayers, these effects were eliminated by blocking 2 -AR, but not 1 -AR. Activation of 2 -AR with guanabenz inhibited isoproterenol-induced I SC by 37% in human IMCD i monolayers and fluid secretion by 91% in rat IMCD i. Immunohistochemistry of human medullary tissue sections revealed greater expression of 2 -AR than 1 -AR; 2 -AR was localized to the basolateral membranes of human IMCD i. Immunoblots identified 2A -AR and 2B -AR in cultured human IMCD i cell monolayers. We conclude that 1 ) catecholamines stimulate cAMP-dependent anion and fluid secretion by IMCD i cells primarily through 2 -AR activation and 2 ) 2 -AR activation attenuates cAMP-dependent anion secretion.

【关键词】  chloride transport catecholamines epinephrine cystic fibrosis transmembrane conductance regulator NKCC

INNER MEDULLARY COLLECTING ducts (IMCD) are located at the terminal portion of nephrons, where salt transport, precisely controlled by hormones and neural factors, may contribute importantly to final urine volume and composition. Several transport proteins have been identified in IMCD cells that may participate in net NaCl absorption or secretion, depending on the physiological requirements for maintaining extracellular fluid homeostasis. Electrogenic Na + absorption by IMCD involves the activity of the basolateral Na + -K + -ATPase and apical epithelial Na + channels (ENaC) ( 46, 64 ) and cyclic-nucleotide gated nonselective cation channels (CNG) ( 27, 51 ), regulated in part by aldosterone ( 15, 34 ), angiotensin II ( 5 ), and atrial natriuretic peptide ( 45, 65 ). More recently, transport proteins known to participate in anion secretion by secretory epithelia, including CFTR, NKCC1, and anion exchangers, were also found to be expressed in IMCD cells ( 1, 16, 17, 20, 22, 33, 47, 50, 53, 55 ). Several studies have uncovered functional evidence of salt secretion by IMCD in situ and in vitro ( 20, 24, 25, 28, 39, 44, 50, 52, 53, 55, 58 ). In primary cultures of human IMCD i cells derived from initial inner medulla, we found that cAMP-stimulated, anion-coupled fluid secretion and CFTR Cl - channel inhibitors blocked anion secretion ( 53 ). NKCC1, a secretory Na + -K + -2Cl - cotransporter, and AE2, an anion transporter that exchanges HCO 3 - and Cl - across the plasma membrane, are present in human IMCD i cell monolayers and participate in Cl - entry across the basolateral membrane during cAMP-dependent Cl - secretion in vitro. AVP and PGE 2, classic renal cAMP agonists, stimulated anion secretion by human IMCD i cells ( 53 ). With the exception of AVP and PGE 2, the hormonal regulation of cAMP-dependent salt and fluid secretion by IMCD i is poorly understood. Considering that -adrenergic agonists mediate their responses via adrenergic receptors coupled to cAMP synthesis, we examined the role of the adrenergic system in anion secretion by human IMCD i cells.

Renal sympathetic nerve activity plays an important role in the renal control of salt and water excretion, a major determinant of extracellular fluid volume and blood pressure. The vast majority of in vivo and in vitro studies directed at the elucidation of the role of the adrenergic system have focused on the regulation of renal blood flow and glomerular filtration ( 11, 23, 30 ). Adrenergic receptors have been found on most nephron segments, including IMCD, indicating that epithelial function is under some degree of control by the adrenergic system ( 2, 11, 14 ). The abundance of the adrenergic receptors on rat collecting duct cells is greater than can be accounted for by direct innervation. Thus neural factors released from sympathetic nerve terminals into the interstitial fluid and catecholamines, either kidney derived or coming from the circulation, may be important modulators of collecting duct transport.

Knowledge of adrenergic signaling in IMCD has, to a large extent, come from studies of isolated IMCD and cultured cells of species other than human. In immunohistochemical studies, -adrenergic receptors ( -AR) and 2 -AR were found in the plasma membranes of rat and rabbit collecting duct cells ( 12, 48, 60 - 62 ). In cultured rat IMCD cells, -AR agonists activated the production of intracellular cAMP via a stimulatory G protein (G s ), and 2 -AR were coupled to an inhibitory G protein (G i ) and 2 -AR activation was shown to impair cAMP generation by the AVP and isoproterenol in IMCD cells ( 12, 21, 40, 43, 48, 49, 61 - 63 ). In this regard, neural factors and hormones that mediate their action via adrenergic receptors may play an important role in the regulation of cAMP-dependent salt and fluid secretion by IMCD i. The present study is the first to examine the functional link among adrenergic receptors, cAMP production, and salt-coupled fluid secretion in human IMCD i cells.

METHODS

Cell culture preparation. Detailed methods for preparing primary cultures enriched in human IMCD i cells are published elsewhere ( 53 ). In brief, normal regions of human kidneys were obtained from surgically excised renal carcinomas in collaboration with the Department of Urology at the Kansas University Medical Center (KUMC). Normal tissue, confirmed by histological examination, was placed in sterile ice-cold PBS and immediately brought to the laboratory. Other kidney tissues were obtained from the Midwest Transplant Network (MTN; Kansas City, KS), an organ retrieval agency. These kidneys had been perfused with an electrolyte preservation solution in preparation for transplantation. Kidneys that were deemed unsuitable for transplantation due to anomalous vasculature, but were otherwise normal, were maintained in ice-cold solution and delivered to the laboratory within 24 h from the time the renal vessels were clamped. We detected no difference in the quality or characteristics in the cultures obtained from the two sources. This protocol was approved by the Institutional Review Board at the University of Kansas Medical Center.

Medullary regions were isolated, and the adjoining cortical tissues were removed 2-3 mm below the corticomedullary boundary, leaving the inner medulla for study. Regions of the initial inner medulla from several papilla were minced and digested in a DMEM/F-12 (1:1) mixture containing 220 IU/ml collagenase (type IV; Worthington Biochemical, Lakewood, NJ) and 100 IU/ml penicillin G and 0.1 mg/ml streptomycin (P/S). Collagenase digestion was stopped by addition of FBS. Cells were rinsed twice and seeded within T75 flasks containing DMEM/F-12 supplemented with 5% FBS, 5 µg/ml insulin, 5 µg/ml transferrin, and 5 ng/ml sodium selenite (ITS; Collaborative Biomedical Products, Bedford, MA) and P/S. Cells were allowed to attach to the flask overnight, and then the unattached tissue and debris were removed and fresh medium was added to the flask. After the cells had reached 70-80% confluency, they were lifted from the plastic with a typsin-EDTA solution and counted with a hemocytometer. Cells were either frozen in culture medium containing 10% DMSO or seeded directly onto permeable supports. The morphology and electrical properties of human IMCD i cell monolayers have been described elsewhere ( 53 ).

cAMP measurements. Human IMCD i cells were seeded (7.3 x 10 5 cells/0.332 cm 2 ) on cell culture supports (6.5-mm diameter, Transwell-COL, CoStar, Cambridge, MA) and grown as confluent monolayers as previously described ( 3, 53 ). For each experiment, IMCD i monolayers were switched to a Ringer medium containing 10 µM benzamil in the apical reservoir for 15 min at 37°C before the experiment [similar to the short-circuit current ( I SC ) experiments]. Epinephrine and specific receptor agonists and antagonists were added to the apical or basolateral medium, and the monolayers were incubated at 37°C for 15 min. The media were removed, and a mixture of 80% methanol-20% water was used to extract cAMP from the cells. cAMP content was determined using an enzyme immunoassay system (Amersham Pharmacia Biotech, Buckinghamshire, UK).

Electrical measurements. Human IMCD i cells were seeded at a density of 2.5 x 10 5 cells/Snapwell (surface area = 1.13 cm 2, CoStar) as described previously ( 53, 54 ). IMCD i cell monolayers were incubated for 3 days in DMEM/F-12, ITS, P/S containing 5% FBS, and then the serum concentration was reduced to 1% for the remaining incubation period. A reduction in the amount of serum in the incubation medium was found to improve and stabilize transepithelial electrical resistance ( R TE ) of renal cells grown on permeable supports (Wallace DP, unpublished observations).

Snapwell supports containing IMCD i monolayers were inserted into Ussing chambers (Harvard Apparatus, Hollison, MA) and bathed in a bicarbonate Ringer solution maintained at 37°C and equilibrated with 5% CO 2 balanced with O 2 ( 53 ). Transepithelial potential difference, I SC, and R TE were monitored with two dual-voltage-clamp devices (Warner Instruments, Hamden, CT). I SC was continuously monitored and recorded using a chart recorder. Positive changes in I SC indicate either anion secretion (e.g., Cl - ) or cation (e.g., Na + ) absorption. We showed previously that a portion of the baseline current was sensitive to inhibition by benzmail, consistent with Na + absorption via ENaC. To reduce Na + absorption, benazamil was added to the apical medium in all experiments, unless noted otherwise.

Fluid transport measurements. Human IMCD i cells (0.8 x 10 6 ) were seeded on 24.5-mm-diameter Transwell-COL (4.52 cm 2, CoStar) and grown to confluency. The measurement of fluid transport across epithelial cell monolayers has been previously described ( 53, 56 ). Briefly, the medium bathing the apical (upper) surface of the cell monolayer was removed and replaced with 200 µl of DMEM/F-12 medium containing 1% FBS, ITS, and P/S. Sterile, water-saturated mineral oil was layered above the fluid to prevent evaporation. Transwell supports containing the cell monolayers were placed in a six-well culture plate containing 2.5 ml of basolateral medium. The fluid-oil mixture was collected after 24 h, and the volume of fluid was determined using calibrated microcapillary tubes (Drummond, Broomall, PA). The volume of fluid transported across the epithelium, expressed in microliters per hour per square centimeter, was determined from the change in volume during the experimental period.

Immunoassays. Cell membrane extracts were prepared from freshly collected initial regions of inner medullas from normal human kidneys and from cultured IMCD i cell monolayers. Tissues were isolated, rapidly frozen in liquid N 2, and kept frozen at -80°C until cell membrane extracts were prepared. Crude cell lysates and membrane extracts were prepared at 4°C. Tissues were homogenized using a Polytron homogenizer (Brinkman, Westbury, NY) in lysate buffer containing 50 mM Tris (pH 7.4 with HCl), 10 mM imidazole, 0.3 M sucrose, and protease inhibitors [including 104 mM 4-(2-aminoethyl) benzenesulfonyl fluoride, 0.08 mM aprotinin, 2.1 mM leupeptin, 3.6 mM bestatin, 1.5 mM pepstatin A, antipain, 100 µM benzamidine, 1 µM DTT, 1 mM PMSF], and cell lysates were further homogenized with a glass dounce. The homogenates were centrifuged at 10,000 g at 4°C for 15 min to remove nuclei and debris. The supernatants were collected and centrifuged at 100,000 g at 4°C for 90 min. The cell membrane pellets were resuspended in 50 mM Tris (pH 7.4) containing 1% NP-40, 0.1% SDS, 0.5% sodium deoxycholate, and protease inhibitors (same as above). Protein concentrations in each of the membrane lysates were determined by Pierce BCA protein assay (Rockford, IL).

To prepare membrane lysates from primary cultures of human IMCD i cells, cell monolayers were grown on plastic petri dishes and maintained in culture as previously described ( 53, 59 ). Culture media were removed, and the cells were washed three times in ice-cold PBS. Cells were scraped and collected, then homogenized with a glass dounce in lysate buffer (same composition as used for tissue). Isolation of membrane proteins was carried out as described above.

A Mini-PROTEAN III cell (Bio-Rad, Hercules, CA) was used to resolve membrane proteins by SDS-PAGE (7.5%) using a method previously described ( 53, 59 ). Protein samples (20 µg) were mixed with equal volume of 2 x sample buffer (132 mM Tris·HCl, pH 6.8, 4.2% SDS, 0.005% bromophenol blue, 21% glycerol, 0.72 M -mercaptoethanol, 20 mM dithiothreitol) and denatured at 95°C for 5 min. Proteins were transferred to a nitrocellulose membrane blocked for 30 min at room temperature with blotting buffer containing 5% nonfat milk in TBS-T (20 mM Tris·HCl, pH 8.0, 137 mM NaCl, and 0.05% Tween 20).

Protein expression of 1 -, 2 -, 2A -, and 2B -AR in cell membrane fractions of human initial inner medullas or cultured IMCD i cells was determined using antibodies purchased from commercial sources. Anti- 2A -AR (Ab-1, rabbit polyclonal) was purchased from Oncogene (Cambridge, MA). Anti- 2B -AR (H-96, rabbit polyclonal), anti- 1 -AR (V-19, rabbit polyclonal), and 2 -AR (H-20, rabbit polyclonal) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). An anti- 1 -AR (rabbit polyclonal) and two anti- 2 -AR antibodies (rabbit polyclonal) were purchased from Research Diagnostics (Flanders, NJ). Nitrocellulose membranes were incubated overnight at 4°C in the blotting buffer containing an antibody with or without blocking peptide (5- to 10-fold excess of peptide fragment of the receptor against which the antibody was raised). The membranes were washed several times in TBS-T and incubated in anti-mouse antibody conjugated to horseradish peroxidase diluted in 5% dry milk in TBS-T for 60 min. The membranes were washed several times and proteins were visualized using the enhanced chemiluminescence system (Amersham Life Sciences, Arlington Heights, IL).

Immunostaining. In situ localization of 1 -AR and 2 -AR were performed on freshly fixed human initial inner medulla tissue. The immunostaining procedure has been described previously ( 53 ). Briefly, the tissues (1-2 mm thick) were fixed overnight in 4% paraformaldehyde at 4°C and then embedded in paraffin. Paraffin sections (5 µm) were placed on microscope slides, deparaffinized, and endogenous peroxidase activity was quenched using 3% H 2 O 2 in methanol. Endogenous biotin was blocked using an avidin-biotin blocking kit (Vector, Burlingame, CA). The slides were rinsed and incubated in 1.5% normal serum for 1 h, then incubated in the primary antibody (diluted to 10 µg/ml in PBS containing 1 mg/ml BSA) overnight at 4° C or for 3 h at room temperature. The tissues were rinsed thoroughly in PBS and incubated in biotinylated secondary antibody (Santa Cruz Biotechnology) for 20 min. Using Vectastain ABC reagents, a complex of biotin and avidin conjugated to horseradish peroxidase was created and visualized with 3,3'-diaminobenzidine (brown precipitate). The sections were incubated in hematoxylin to counterstain the nuclei and cytoplasm, dehydrated in a graded series of ethanol, cleared in xylenes, and mounted with coverslips using permount. Antibodies to 2 -AR were found unsuitable for immunostaining of the human medullary tissue.

Measurements of fluid transport in freshly microdissected rat IMCD i. In experiments designed to examine the effects of AR agonists and antagonists on fluid transport in intact IMCD i, individual collecting ducts were dissected from the initial inner medulla of rat kidneys. Methods for tissue preparation and measurements of fluid transport in IMCD i have been described in detail ( 55 ). Male Sprague-Dawley rats (3-5 wk of age) were allowed free access to water containing 3% sucrose and 0.225% NaCl for 17 h. Animals were anesthesized by inhalation of isoflurane, the left kidney was removed, and a thin slice containing the inner medulla was obtained. Animals were killed by exsanguination following established guidelines. Individual IMCD i were dissected in DMEM/ F-2 supplemented with 5% FBS and randomly attached to culture plates (1 experimental condition/plate) coated with poly- L -lysine. The plates were warmed for 1 h before the beginning of the experiment and maintained at 37°C throughout. AR agonists and/or antagonists were added at time 0, and lumen diameters were measured at 15, 30, 60, and 120 min by video analysis. To enable rapid measurements, the largest diameter 500 µm in length was used to calculate lumen volume assuming cylindrical tubule geometry.

Statistics. Data are presented as means ± SE. Where appropriate, Student's t -test or one-way ANOVA and the Student-Newman-Keuls multiple comparison posttest were used to determine statistical significance. Data groups containing heterogeneous variances indicated by the Bartlett test were analyzed using the nonparametric tests, the Mann-Whitney U -test or the Kruskal-Wallis nonparametric ANOVA and Dunn's posttest. P < 0.05 was taken to indicate statistical significance.

RESULTS

-AR and -AR modulate intracellular cAMP in human IMCD i cells. Previously, we showed that AVP (100 mU/ml) and PGE 2 (25 ng/ml) increased cAMP in human IMCD i cells from 0.46 ± 0.04 pmol/monolayer in the control medium to 3.45 ± 0.62 and 12.35 ± 4.18 pmol/monolayer, respectively ( 53 ). In the present study, we found that 1 µM epinephrine ( Fig. 1 ) increased intracellular cAMP from 0.48 ± 0.04 to 27.17 ± 3.04 pmol/monolayer, nearly the same amplitude as with forskolin (29.26 ± 3.88 pmol/monolayer) ( 53 ). ICI-118, 551 (a 2 -AR antagonist) completely blocked the effect of epinephrine on cAMP production (cAMP content in this group was not significantly different from the control group). By contrast, the addition of atenolol to block 1 -AR did not affect cAMP stimulation by epinephrine. This indicates that the activation of cAMP production in IMCD i cells by epinephrine was mediated via 2 -AR, but not 1 -AR.

Fig. 1. Effect of epinephrine and 1 -adrenergic receptor (AR) and 2 -AR antagonists on intracellular cAMP levels in human initial inner medullary collecting ducts (IMCD i ) monolayers. Epinephrine (1 µM) increased intracellular cAMP in IMCD i cell monolayers ( n = 4 monolayers/group). Incubating the monolayers in 10 µM ICI-118, 551, a 2 -AR antagonist, for 15 min before the addition of epinephrine completely blocked cAMP production. By contrast, there was no effect on cAMP production by pretreatment with 10 µM atenolol, a 1 -AR antagonist. Values are means ± SE. * P < 0.001 compared with control. # P < 0.001 compared with epinephrine alone.

Several studies localized 2 -AR within plasma membranes of rat and rabbit IMCD cells ( 8, 11, 12, 14, 48, 49, 62 ). To determine whether activation of 2 -AR inhibits intracellular cAMP synthesis in human IMCD i cells, we examined the effect of guanabenz, an 2 -AR agonist, on isoproterenol-induced cAMP production. IMCD i cell monolayers were treated for 15 min with either control medium, 10 nM isoproterenol, or the combination of 10 µM guanabenz and isoproterenol ( Fig. 2 A ). Isoproterenol increased cAMP from 0.39 ± 0.05 to 9.50 ± 0.96 pmol/monolayer. Guanabenz reduced cAMP production by isoproterenol to 4.11 ± 0.30 pmol/monolayer.

Fig. 2. Effect of 2 -AR agonists and antagonists on cAMP production in human IMCD i monolayers. A : isoproterenol (10 nM), a -AR agonist, increased intracellular cAMP levels in human IMCD i cells. Guanabenz (10 µM), an 2 -AR agonist, added 15 min before the addition of isoproterenol, inhibited the production of intracellular cAMP. Values are means ± SE. * P < 0.05 compared with previous group ( n = 4 monolayers/condition). B : norepinephrine (1 µM) increased intracellular cAMP levels in human IMCD i monolayers. Blocking 2 -AR by prior incubation of the monolayers with 10 µM rauwolscine, an 2 -AR antagonist, enhanced cAMP production by norepinephrine. By contrast, prior incubation with prazosin (10 µM), an 1 -AR antagonist, had no effect on the norepinephrine response. Values are means ± SE. There were 5 monolayers/condition except for the norepinephrine group, in which an outlier (a value greater than 2 SD from the mean) was removed. * P < 0.05 compared with control. # P < 0.05 compared with norepinephrine alone. P < 0.01 compared with previous group.

Norepinephrine (NE), an agonist that binds to both -AR and -AR, caused a small but significant increase in cAMP ( Fig. 2 B ). The magnitude of the NE response was 2.4-fold greater in the presence of 10 µM rauwolscine, an 2 -AR antagonist. By contrast, cAMP production induced by NE was unaffected by prazosin (10 µM), an antagonist of the 1 -AR. Blocking -AR completely prevented NE stimulation of cAMP (data not shown). Thus NE stimulation of cAMP was likely mediated through -AR activation and opposed to some degree by 2 -AR activation. These data are consistent with -AR and 2 -AR in human IMCD cells having contrasting effects on the activity of adenylyl cyclase.

2 -AR and 2 -AR regulate transepithelial anion secretion by human IMCD i cells. IMCD i cell monolayers were grown for 7-10 days on permeable supports, and an I SC method was used to examine the effects of adrenergic receptor agonists and antagonists on electrical parameters. Previously, we showed that human IMCD i cells developed an R TE (258 ± 47 ·cm 2 ), a lumen negative potential difference of (-0.3 ± 0.1 mV) and a positive I SC (1.4 ± 0.5 µA/cm 2 ) ( 53 ). The baseline current was reduced by benzamil, an inhibitor of ENaC. cAMP agonists (e.g., AVP, PGE 2 ) and 8-bromo-cAMP stimulated anion secretion by the IMCD i monolayers. In the current study, benzamil was included in the apical medium to limit the contribution of Na + -absorptive pathways to I SC.

To investigate the effect of -AR activation on I SC in IMCD i cells, we incubated six pairs of monolayers in increasing concentrations of either epinephrine or isoproterenol ( Fig. 3 ). A significant increase in I SC was attained with concentrations of basolateral isoproterenol as low as 0.1 nM, and a near-maximal increase in I SC of 3.1 ± 0.4 µA/cm 2 above baseline was achieved with 10 nM isoproterenol. Epinephrine increased I SC in a similar dose dependency; the increases were significant at 1 nM and above. The concentrations of agonists to induce half-maximal stimulation (EC 50 ) were 10 nM for epinephrine and 1.9 nM for isoproterenol. There was no difference in the maximal I SC stimulation between the two agonists. In other experiments, norepinephrine (1 µM) increased I SC from 1.3 ± 0.3 to 3.8 ± 0.4 µA/cm 2 ( P < 0.0001, n = 10).

Fig. 3. Effect of -AR agonists epinephrine and isoproterenol on short-circuit current ( I SC ) of IMCD i cell monolayers. Values are means ± SE. (6 pairs of monolayers) for I SC after a 5-min incubation in benzamil to reduce the Na + current. Successive increments of the agonists were added to the basolateral medium after a steady-state was achieved with the previous concentration. Significant increases in I SC were observed with epinephrine (EC 50 = 10 nM) and isoproterenol (EC 50 = 1.9 nM). In other studies, no difference was observed in the peak response of 1 µM epinephrine when added to either the apical or basolateral medium. * P < 0.05 compared with zero.

To determine whether the effect of epinephrine was mediated by 2 -AR, we examined the effects of epinephrine in the presence and absence of the 2 -AR antagonist ICI-118, 551 (a typical response is shown in Fig. 4 ). Incubating monolayers in media containing 10 µM ICI-118, 551 for 15 min nearly eliminated the response to epinephrine. Moreover, ICI-118, 551 added to the basolateral medium secondary to the addition of epinephrine inhibited a sustained increase in I SC. The direct activation of adenylyl cyclase with forskolin restored I SC, demonstrating that the inhibitory locus of ICI-118, 551 was proximal to adenylyl cyclase. In a total of five experiments, 1 µM epinephrine increased I SC from 2.14 ± 0.46 to 6.26 ± 1.05 µA/cm 2 ( P < 0.001). The subsequent addition of ICI-118, 551 inhibited the sustained current to 2.63 ± 0.89 µA/cm 2 ( P < 0.01, compared with epinephrine alone).

Fig. 4. Effect of ICI-118, 551, a selective 2 -AR antagonist, on I SC induced by epinephrine in human IMCD i cell monolayers. Pretreating a monolayer with 10 µM basolateral ICI-118, 551 diminished the I SC induced by basolateral addition of 1 µM epinephrine ( ). Application of ICI-118, 551 to a sustained current secondary to the epinephrine stimulation also attenuated the I SC ( ). In both monolayers, inhibition with -AR inhibition was overridden by direct activation of adenylyl cyclase with forskolin. See text for composite values for a set of similar experiments.

To investigate further the cellular pathway of the catecholamine effect on anion secretion by human IMCD i monolayers, we examined the response to epinephrine (1 µM) in five pairs of monolayers incubated for 15 min in control media or media containing H-89, a PKA inhibitor. Epinephrine increased I SC from 2.07 ± 0.80 to 5.06 ± 1.06 µA/cm 2 ( P < 0.001, n = 5) in the untreated monolayers. Propranolol, a nonselective -AR antagonist, inhibited the current to 2.90 ± 1.00 µA/cm 2 ( P < 0.001), and forskolin increased the current to 5.53 ± 1.26 µA/cm 2 ( P < 0.001). By contrast, in the presence of H-89, epinephrine failed to increase I SC (0.50 ± 0.35 control vs. 0.80 ± 0.41 µA/cm 2; data not significantly different from control current), and the subsequent addition of forskolin had no effect on I SC, consistent with PKA inhibition by H-89 treatment (0.76 ± 0.35 µA/cm 2 ). The results of a typical experiment are shown in Fig. 5.

Fig. 5. Effect of PKA inhibitor H-89 on I SC induced by epinephrine in a human IMCD i cell monolayer. H-89 (10 µM), added to the apical and basolateral media for 15 min, diminished the current induced by 1 µM basolateral epinephrine ( ). The absence of a forskolin-induced rise in I SC confirmed that PKA was effectively inhibited. Basolateral application of propranolol (10 µM), a nonselective -AR antagonist, inhibited the sustained current induced by epinephrine ( ). Forskolin returned I SC to the same level as that recorded before the addition of propranolol.

To determine whether 2 -AR activation inhibits cAMP-dependent anion secretion by human IMCD i monolayers, we monitored I SC after the sequential additions to the bathing medium of 1 nM isoproterenol and 10 µM guanabenz ( Fig. 6 ). We found that isoproterenol increased I SC to a steady-state value 5.50 µA/cm 2 above control and guanabenz inhibited this current by 44%. This inhibitory effect was proximal to adenylyl cyclase because activating the enzyme with forskolin restored I SC to the prior level in the continued presence of guanabenz. In six similar experiments, 1 nM isoproterenol increased I SC from 2.58 ± 1.2 to 6.7 ± 1.2 µA/cm 2, P < 0.001. Guanabenz reduced the current to 5.17 ± 1.2 µA/cm 2, P < 0.05, whereas the subsequent addition of forskolin stimulated the current to 7.2 ± 2.2 µA/cm 2, P < 0.01.

Fig. 6. Effect of isoproterenol, a -AR agonist, and guanabenz, a 2 -AR agonist, on I SC. Isoproterenol (1 nM) stimulated I SC in an IMCD i monolayer. Activation of 2 -AR with the application of guanabenz (10 µM) caused a rapid inhibition of the isoproterenol-stimulated I SC. The subsequent addition of forskolin increased I SC, indicating that guanabenz inhibited the isoproterenol effect at the level of adenylyl cyclase, possibly through a G i -dependent mechanism. Benzamil was added to the apical medium for 15 min before the experiments to reduce the Na + current. See text for composite values for a set of similar experiments.

-AR agonists stimulate fluid secretion by human IMCD i cells. We determined whether activation of -AR by isoproterenol stimulates fluid secretion by IMCD cells. Cell monolayers were grown on Transwell-COLS for 14-24 days. The rate of fluid transport was determined over a 12-h incubation period ( Table 1 ). Group 1 was incubated in control medium consisting of DMEM/F-12 supplemented with 1% FBS, ITS, P/S, and 10 µM benzamil (to reduce absorption). Monolayers in group 2 were incubated in the same medium containing 100 nM isoproterenol, and those in group 3 contained isoproterenol and 10 µM ICI-118, 551. We found that in monolayers treated with control medium no significant transport of fluid occurred; however, monolayers incubated with isoproterenol secreted fluid. ICI-118, 551 eliminated fluid secretion induced by isoproterenol, indicating that the effect of isoproterenol was mediated by 2 -AR activation.

Table 1. Effect of isoproterenol and ICI-118, 551 on fluid secretion by human IMCD i cells

-AR expression in human IMCD i. Polyclonal antibodies that recognize portions of either human 1 -AR or 2 - AR were used to examine -AR expression in human IMCD i in situ and in cultured cells. By Western blot analysis ( Fig. 7 A ), we detected abundant immunoreactivity of 2 -AR at an apparent molecular mass of 55 kDa in extracted membranes of tissue collected from the initial region of the human inner medulla and in cultured IMCD i cell monolayers. This molecular mass is in agreement with the apparent molecular mass of the 2 -AR ( 4 ). The antibody also detected a weaker band at 76 kDa, suggesting the presence of a glycosylated form. In Fig. 7 B, using an anti- 1 -AR antibody, we found bands at 42, 62, and 75 kDa from membrane lysates of IMCD i cell monolayers. Similar bands were detected in fresh tissue from human initial inner medulla. 1 -AR have been reported to migrate to bands at 40, 50, and 67 kDa ( 6, 19 ). We have no explanation for the larger 75-kDa band, except for the possibility of alternate glycosylation of the protein. All bands were eliminated by a competing peptide.

Fig. 7. Western blot analysis for 2 -AR and 1 -AR proteins in membrane lysates from cultured human IMCD i cells and human initial inner medulla. Proteins (20 µg/lane) were separated by SDS-PAGE and probed using anti-human 2 -AR and 1 -AR antibodies. Left : apparent molecular mass of the proteins detected by the antibodies. A : an antibody specific for the 2 -AR (H-20, Santa Cruz Biotechnology) detected bands at 55 and 76 kDa in membranes extracted from IMCD i cells and tissue. This pattern of expression was observed in lysates from IMCD i cell cultures and tissues from 3 different kidney preparations and by an anti-human 2 -AR antibody from Research Diagnostics. CP, competing peptide. B : anti- 1 -AR antibody (V-19; Santa Cruz Biotechnology) detected bands at 42, 62, and 75 kDa in both the cultured IMCD cells and inner medullary tissue. A similar protein expression of these receptors was observed in membranes extracted from 4 cultures of IMCD i cells and 3 tissue preparations.

By immunohistochemistry, we found 2 -AR are abundantly expressed in the IMCD i, with the highest level of expression in basolateral membranes ( Fig. 8 A ). This pattern of expression was observed using three anti- 2 -AR antibodies that recognize different regions of the receptor. The 2 -AR antibody binding was reduced by immunizing peptide competition ( Fig. 8 C ), and no staining was visible when the primary antibody was substituted with a nonspecific IgG from the same animal species from which the primary antibody was prepared (not shown). By contrast, we observed little 1 -AR staining in human IMCD i ( Fig. 8 B ), and the distribution was relatively diffuse.

Fig. 8. Expression of 2 -AR and 1 -AR in human initial inner medullary collecting ducts. A : an anti- 2 -AR antibody (H-20; Santa Cruz Biotechnology) detected abundant immunostaining for 2 -AR on the basolateral membranes of human initial IMCD (arrows). B : an anti- 1 -AR antibody (V-19; Santa Cruz Biotechnology) also recognized proteins on IMCD i; however, the intensity of the staining was much less and more diffuse. Competing peptides (5 x ) added to the incubation solutions containing the antibodies eliminated 2 -AR ( C ) and 1 -AR ( D ) staining.

2 -AR subtypes in human IMCD i cells. The presence of 2 -AR subtypes A and B in the membranes of cultured human IMCD i cells was assessed by Western blot analysis. Using an anti-human 2A -AR antibody, we detected a protein with a mobility of 64 kDa ( Fig. 9 ). Weaker bands were also detected at 45 and 57 kDa. We also found 2B -AR in membranes of cultured IMCD i cells. An antibody to the human 2B -AR detected a protein with a mobility of 66 kDa. In gels loaded with 40 µg protein, we detected a weak band at 48 kDa. In whole cell lysates, a predominant 43-kDa band was present. Although 2 -AR appear to be expressed in the human IMCD i cells by Western blot analysis, specific labeling of the receptors in IMCD i could not be detected with our antibodies by immunohistochemistry.

Fig. 9. Western blot analysis for 2A -AR and 2B -AR proteins in membrane lysates from cultured human IMCD i cells. Proteins (20 µg/lane) were separated by SDS-PAGE and probed using anti-human 2A -AR (Ab-1; Oncogene) and 2B -AR (H-96; Santa Cruz Biotechnology) antibodies. Numbers indicate the apparent molecular mass of the proteins detected by the antibodies. An antibody specific for the 2A -AR detected bands at 45, 57, and 64 kDa. Anti-human 2B -AR antibody detected bands at 66 and 86 kDa in the cultured IMCD cells. A similar protein expression of these receptors was observed in membranes extracted from 4 cultures of IMCD i cells. Competing peptides were not available. No bands were present when the primary antibody was substituted with a control IgG.

-AR and 2 -AR agonists regulate net fluid secretion in intact rat IMCD i. To examine the effects of -AR and 2 -AR agonists on fluid secretion in intact IMCD i, we dissected individual IMCD i from rat kidneys and determined the rates of fluid secretion from changes in lumen volume over a 2-h interval in media containing -AR and 2 -AR agonists and antagonists. Epinephrine (1 µM) stimulated fluid secretion from a control baseline rate of 0.018 nl·h -1 ·mm -1 ( n = 8) to 0.081 nl·h -1 ·mm -1 ( n = 9, P < 0.001). Blocking -AR with propranolol decreased the rate of secretion to 0.036 nl·h -1 ·mm -1 ( n = 8, P < 0.01 compared with epinephrine alone). 2 -AR stimulation with guanabenz blocked cAMP-dependent fluid secretion induced by -AR activation in rat IMCD i ( Fig. 10 ). This inhibitory effect of guanabenz was prevented with rauwolscine, an antagonist of 2 -AR. These data indicate that -AR and 2 -AR regulation of cAMP-dependent anion and fluid secretion by IMCD i is not limited to cells in culture.

Fig. 10. Effect of 2 -AR and 2 -AR agonists on fluid secretion in freshly microdissected rat IMCD i. Isoproterenol (1 µM) stimulated net fluid secretion in rat IMCD i. Pretreatment of IMCD i for 1 h in 10 µM guanabenz, an 2 -AR agonist, inhibited the effect of isoproterenol on fluid secretion. Blocking 2 -AR with the addition of 10 µM rauwolscine, an 2 -AR antagonist, partially restored isoproterenol-induced fluid secretion. Benzamil was added to all groups to reduce Na + absorption. The number of IMCD i is shown in parentheses below each bar. Values are means ± SE. Significant differences were determined by the Kruskal-Wallis nonparametric ANOVA and Dunn's multiple comparison posttest. * P < 0.01 compared with control. # P < 0.01 compared with isoproterenol alone and not significant from control. P < 0.05 compared with control.

DISCUSSION

Kidneys regulate salt excretion with extraordinary precision, assisted by neural and hormonal control of the glomerular filtration rate (GFR) and the rate of salt transport by various nephron segments. Over the last two decades, the collecting ducts have been recognized as potentially important sites for making the final adjustments to urine composition. In the present study, we determined the effects of catecholamines and specific 2 -AR and -AR agonists and antagonists on cAMP production and anion-coupled fluid secretion in human IMCD i cells in vitro and examined the expression and localization of adrenergic receptors in human IMCD i in situ. We found that -AR agonists, epinephrine and isoproterenol, were agonists of Cl - -dependent fluid secretion, mediated by 2 -AR stimulation of cAMP-dependent PKA. By contrast, activation of 2 -AR inhibited cAMP accumulation and anion secretion induced by -AR agonists in human IMCD i cells. This study is the first to demonstrate that salt and water secretion by human IMCD i cells is regulated, in part, by the adrenergic system.

Role of -AR in salt-coupled fluid secretion in IMCD i cells.

Kidneys are richly innervated by sympathetic nerves, providing a source of adrenergic agonists. In addition to the release of catecholamines from renal nerves, a significant fraction of urinary epinephrine may be derived from nonneural sources in the kidney ( 11, 66 ). Catecholamines have been shown to activate intracellular cAMP production in rat IMCD cells ( 29, 60, 61 ) and regulate ion transport in a number of secretory epithelia, including that of the airway ( 13 ), intestine ( 35 ), salivary gland ( 35 ), pancreatic ducts ( 36 ), and semicircular canal ducts ( 31 ).

Epinephrine and NE have multiple actions in the kidney (7-9, 11, 12, 14, 21, 37, 40, 48, 49, 61-63), and their effects are expressed by binding to specific adrenergic receptors. Renal -AR activate adenylyl cyclase via an interaction with G S. Both 1 -AR and 2 -AR have been detected in collecting ducts, with variable expression among the different regions and among different animal species ( 11 ). In the current study, we used polyclonal antibodies to examine the expression of -AR protein subtypes in human IMCD i. We found abundant immunostaining for 2 -AR in the basolateral membrane of IMCD i in situ ( Fig. 8 A ), whereas 1 -AR staining was much weaker and diffuse ( Fig. 8 B ).

Isoproterenol, epinephrine, and NE stimulated cAMP production ( Figs. 1 and 2 ), and transepithelial anion secretion reflected in an increase in benzamil-insensitive I SC ( Figs. 3 - 6 ) in cultured human IMCD i cell monolayers. Moreover, isoproterenol stimulated net fluid secretion in human IMCD i monolayers ( Table 1 ) and in freshly isolated rat IMCD i ( Fig. 10 ). We have made an enabling calculation to compare the rates of fluid secretion by human IMCD i cell monolayers and dissected rat IMCD i. For a dilated rat IMCD i with an inner diameter of 20 µm, a 1-mm length would have 0.000628 cm 2 (surface area = 2 r x L ) of epithelium. Thus a secretion rate of 0.26 µl·h -1 ·cm -2 by human IMCD i cell monolayers would equal 0.0027 nl·h -1 ·mm -1 (0.000628 cm 2 = 1-mm tubule length). The average rate of fluid secretion in rat IMCD i ( Fig. 10 ) treated with isoproterenol was 0.0015 nl·min -1 ·mm -1, which is on the same order of magnitude as for human IMCD i monolayers. Evidence for salt and fluid secretion has been previously reported in rat IMCD in vitro. Rocha and Kudo ( 39 ) found that the unidirectional isotopic fluxes of Na + and Cl - reversed from absorption in control medium to secretion with the addition of permeable cAMP ( 39 ). Wall et al. ( 52 ) also reported Cl - secretion (2.7 ± 0.9 pmol·min -1 ·mm -1 ) and measured a net fluid secretion rate of 0.022 ± 0.011 nl·min -1 ·mm -1 in isolated, perfused IMCD from rats treated with deoxycorticosterone. We do not have a unique explanation for the higher rate of fluid secretion observed by Wall and associates.

The results of -AR activation of cAMP production in human IMCD i cells are in agreement with those of Teitelbaum et al. ( 48 ) and Yasuda et al. ( 61, 63 ), who found that isoproterenol increased cAMP in cultured rat IMCD cells. ICI-118, 551, a 2 -AR antagonist, blocked both cAMP production ( Fig. 1 ) and anion secretion ( Fig. 4 ) induced by epinephrine and prevented fluid secretion induced by isoproterenol ( Table 1 ). By contrast, atenolol, a 1 -AR antagonist, had no effect on cAMP production induced by -AR agonists ( Fig. 1 ). In pharmacological studies, NE was found to be less potent than epinephrine in activating 2 -AR ( 32 ). We found that epinephrine had a much greater effect on cAMP and anion secretion compared with NE, consistent with a predominance of 2 -AR. Based on these data, we conclude that 2 -AR, present in the basolateral membranes of human IMCD, mediate cAMP-dependent anion and fluid secretion induced by catecholamines.

2 -AR activation decreases cAMP-dependent anion secretion by human IMCD cells.

-AR are classified into 1 -AR and 2 -AR ( 14 ). 1 -AR are coupled to phospholipase C (activating phosphatidylinositide hydrolysis and modulating intracellular Ca 2+ ) and phospholipase A 2 (for PGE 2 production) ( 14, 42 ) and participate in vasoconstriction of renal arterioles ( 10 ). On the other hand, 2 -AR modulate cAMP production through G i ( 21 ) and have been shown to decrease cAMP formation by AVP in cortical and medullary collecting duct ( 37, 48, 49 ). 2 -AR agonists are generally thought to exert their effects by a mechanism proximal to the generation of cAMP. In the current study, we found that guanabenz, an 2 -AR agonist, inhibited isoproterenol-induced cAMP production and anion secretion in human IMCD i cells ( Figs. 2 and 6, respectively) and net fluid secretion by isolated rat IMCD i ( Fig. 10 ). The inhibitory effect of guanabenz on anion secretion by human IMCD i monolayers recovered completely after direct activation of adenylyl cyclase with forskolin. Thus in human IMCD i cells, the inhibition of cAMP-mediated anion secretion by guanabenz occurs upstream of adenylyl cyclase and is likely due to 2 -AR that are functionally coupled to G i.

Wilborn et al. ( 57 ) used RT-PCR of RNA extracted from microdissected cortical collecting duct of rat kidneys to show the presence of both 2A -AR and 2B -AR. However, in IMCD, the expression of the 2 -AR subtypes is species specific. In the rabbit, 2A -AR are expressed in IMCD, whereas in rat IMCD 2B -AR are expressed ( 8, 62 ). Using a human kidney cDNA library, Regan et al. ( 38 ) cloned a gene using the sequence from human platelet 2 -AR receptor and expressed the protein (apparent molecular masses of 42, 67, and 75 kDa) in COS-7 cells. Ligand binding studies indicated that the subtype was 2B -AR. By Western blot analysis, we found that both 2A -AR and 2B -AR proteins are expressed in the membranes of cultured human IMCD i cells. 2A -AR appeared to migrate to a major band at 64 kDa and minor bands at 45 and 57 kDa. Our results are in accord with previous studies in human kidney ( 38 ), human platelet, and neonatal lung ( 26 ) showing that 2A -AR migrate to 65 and 45 kDa.

We also found that 2B -AR were expressed in human IMCD i cell membranes with apparent molecular masses of 48, 66, and 86 kDa. 2B -AR have been reported to migrate to 48 and 62 kDa in rat corical collecting duct cells ( 57 ). Sheeve et al. ( 41 ) also reported a 2 -AR at 85 kDa in rat liver; however, the subtype was not determined. Thus based on our immunoblot data, we conclude that both 2A -AR and 2B -AR are present in human IMCD cells and may contribute to the regulation of cAMP-dependent NaCl and fluid secretion.

Potential role of the adrenergic system in the regulation of solute and fluid secretion by IMCD.

Understanding the explicit role of the adrenergic system in the regulation of collecting duct electrolyte and fluid transport in situ has been difficult owing to 1 ) the complex interactions among renal nerves and circulating adrenergic hormones; 2 ) the multiple effectors within the kidney, including blood vessels, glomeruli, and various nephron segments; 3 ) the variable expression and regulation of the adrenergic receptors; and 4 ) the capacity for NE and epinephrine to bind and activate both -AR and -AR. Recently, investigators have used in vitro methods to examine intrinsic mechanisms within collecting duct segments that might implicate a role for the adrenergic system in situ. The current study has uncovered a previously unrecognized function in human medullary collecting duct cells (adrenergic regulation of electrolyte and fluid secretion) and has revealed some of the molecular mechanisms underlying this function. The examination of the respective roles of individual mediators in situ will require novel experimental approaches that fall outside the scope of the present study.

Based on historical information, we can only speculate about the physiological role of adrenergic agonists in mediating solute secretion by IMCD. It is generally accepted that epinephrine and NE induce antinatriuresis in intact animals primarily by reducing renal blood flow, decreasing GFR, and increasing the fractional reabsorption of solutes in nephron segments proximal to the collecting duct system. Severe dehydration is a condition in which a reduction in GFR and an elevation in fractional reabsorption of solutes and fluid minimize urinary water loss. In this context, solute secretion by IMCD into highly concentrated urine would lessen the increase in Na + concentration in extracellular fluids. Evidence supporting this hypothesis was presented by Luke ( 28 ), who found that withholding access to drinking water while maintaining NaCl intake constant caused rats to increase the fractional excretion of Na +. It is conceivable that during dehydration, catecholamines released from renal nerves coupled with increased circulating levels of adrenergic hormones may activate 2 -AR in IMCD i to induce cAMP-dependent solute secretion and lessen an elevation in plasma osmolality. It is also plausible that active solute secretion and the obligatory osmotic movement of fluid by the IMCD during conditions associated with a marked reduction in glomerular filtration could be important for propelling small amounts of unreabsorbed glomerular filtrate into the renal pelvis as a fail-safe mechanism to eliminate toxins and other products of metabolism ( 18 ). IMCD salt secretion ( 20, 25, 39, 44, 50, 52, 53, 55, 58 ) may be a default mechanism conserved throughout evolution serving to preserve some degree of urine formation in extreme circumstances. In this context, "survival" hormones, epinephrine and NE, may act on IMCD i to stimulate urinary salt secretion.

The capacity for salt-coupled fluid secretion is evident in renal cyst formation. In autosomal dominant polycystic kidney disease, cysts develop from the outgrowth of individual tubule cells and fluid accumulates within the isolated sacs. cAMP stimulates anion and fluid secretion by human autosomal dominant polycystic kidney disease epithelial cells through activation of apical CFTR and basolateral NKCC1 ( 54 ). Thus catecholamines, mediated by an elevation in intracellular cAMP in the mural epithelial cells, may have adverse effects on renal cyst progression.

In summary, the current experiments demonstrate that adrenergic agonists acting via 2 -AR stimulate cAMP-dependent anion and fluid secretion by human IMCD i cells. By contrast, specific 2 -AR activation inhibits cAMP generation and anion secretion induced by -AR agonists. These findings are consistent with the view that human and rat IMCD i harbor intrinsic mechanisms, regulated by the adrenergic system, to induce salt secretion in pathophysiological conditions.

GRANTS

This work was supported by a grant from the National Institutes of Health Grant P20-RR-17686 (D. P. Wallace). Portions of this study were published in abstract form ( J Am Soc Nephrol.12: 43A, 2001).

ACKNOWLEDGMENTS

We thank Drs. Tamio Yamaguchi and Gustavo Blanco and Marcy Christensen for technical assistance and Drs. Jared Grantham and Larry Sullivan for helpful discussions.

【参考文献】
  Alper SL, Stuart-Tilley AK, Biemesderfer D, Shmukler BE, and Brown D. Immunolocalization of AE2 anion exchanger in rat kidney. Am J Physiol Renal Physiol 273: F601-F614, 1997.

Barajas L, Powers K, and Wang P. Innervation of the renal cortical tubules: a quantitative study. Am J Physiol Renal Fluid Electrolyte Physiol 247: F50-F60, 1984.

Belibi FA, Wallace DP, Yamaguchi T, Christensen M, Reif G, and Grantham JJ. The effect of caffeine on renal epithelial cells from patients with autosomal dominant polycystic kidney disease. J Am Soc Nephrol 13: 2723-2729, 2002.

Benovic JL, Shorr RG, Caron MG, and Lefkowitz RJ. The mammalian 2 -adrenergic receptor: purification and characterization. Biochemistry 23: 4510-4518, 1984.

Beutler KT, Masilamani S, Turban S, Nielsen J, Brooks HL, Ageloff S, Fenton RA, Packer RK, and Knepper MA. Long - term regulation of ENaC expression in kidney by angiotensin II. Hypertension 41: 1143-1150, 2003.

Boege F, Jurss R, Cooney D, Hekman M, Keenan AK, and Helmreich EJ. Functional and structural characterization of the two 1 -adrenoceptor forms in turkey erythrocytes with molecular masses of 50 and 40 kilodaltons. Biochemistry 26: 2418-2425, 1987.

Brodde OE and Michel MC. Adrenergic receptors and their signal transduction mechanisms in hypertension. J Hypertens Suppl 10: S133-S145, 1992.

Clarke D and Garg LC. 2 Adrenergic receptors in inner medullary collecting duct cells of the rabbit kidney. J Pharmacol Exp Ther 264: 879-888, 1993.

Cohen HT and Katz AI. Adrenergic receptors and catecholamine effects on sodium transport in the mammalian nephron. Semin Nephrol 11: 110-119, 1991.

Cooper CL and Malik KU. Prostaglandin synthesis and renal vasoconstriction elicited by adrenergic stimuli are linked to activation of 1 adrenergic receptors in the isolated rat kidney. J Pharmacol Exp Ther 233: 24-31, 1985.

DiBona GF and Kopp UC. Neural control of renal function. Physiol Rev 77: 75-197, 1997.

Edwards RM, Stack EJ, Gellai M, and Brooks DP. Inhibition of vasopressin-sensitive cAMP accumulation by 2 -adrenoceptor agonists in collecting tubules is species dependent. Pharmacology 44: 26-32, 1992.

Frizzell RA, Rechkemmer G, and Shoemaker RL. Altered regulation of airway epithelial cell chloride channels in cystic fibrosis. Science 233: 558-560, 1986.

Garg LC. Actions of adrenergic and cholinergic drugs on renal tubular cells. Pharmacol Rev 44: 81-102, 1992.

Garty H. Regulation of the epithelial Na + channel by aldosterone: open questions and emerging answers. Kidney Int 57: 1270-1276, 2000.

Ginns SM, Knepper MA, Ecelbarger CA, Terris J, He X, Coleman RA, and Wade JB. Immunolocalization of the secretory isoform of Na-K-Cl cotransporter in rat renal intercalated cells. J Am Soc Nephrol 7: 2533-2542, 1996.

Glanville M, Kingscote S, Thwaites DT, and Simmons NL. Expression and role of sodium, potassium, chloride cotransport (NKCC1) in mouse inner medullary collecting duct (mIMCD-K2) epithelial cells. Pflügers Arch 443: 123-131, 2001.

Grantham JJ and Wallace DP. Return of the secretory kidney. Am J Physiol Renal Physiol 282: F1-F9, 2002.

Graziano MP, Moxham CP, and Malbon CC. Purified rat hepatic 2 -adrenergic receptor. Structural similarities to the rat fat cell 1 -adrenergic receptor. J Biol Chem 260: 7665-7674, 1985.

Husted RF, Volk KA, Sigmund RD, and Stokes JB. Anion secretion by the inner medullary collecting duct. Evidence for involvement of the cystic fibrosis transmembrane conductance regulator. J Clin Invest 95: 644-650, 1995.

Jeffries WB and Pettinger WA. Adrenergic signal transduction in the kidney. Miner Electrolyte Metab 15: 5-15, 1989.

Kaplan MR, Plotkin MD, Brown D, Hebert SC, and Delpire E. Expression of the mouse Na-K-2Cl cotransporter, mBSC2, in the terminal inner medullary collecting duct, the glomerular and extraglomerular mesangium, and the glomerular afferent arteriole. J Clin Invest 98: 723-730, 1996.

Katz MA and Shear L. Effects of renal nerves on renal hemodynamics. I. Direct stimulation and carotid occlusion. Nephron 14: 246-256, 1975.

Keeler R and Azzarolo AM. Sodium excretion and renal precession of sodium over inulin. Am J Physiol Regul Integr Comp Physiol 260: R1183-R1187, 1991.

Kizer NL, Vandorpe D, Lewis B, Bunting B, Russell J, and Stanton BA. Vasopressin and cAMP stimulate electrogenic chloride secretion in an IMCD cell line. Am J Physiol Renal Fluid Electrolyte Physiol 268: F854-F861, 1995.

Lanier SM, Homcy CJ, Patenaude C, and Graham RM. Identification of structurally distinct 2 -adrenergic receptors. J Biol Chem 263: 14491-14496, 1988.

Light DB, Corbin JD, and Stanton BA. Dual ion-channel regulation by cyclic GMP and cyclic GMP-dependent protein kinase. Nature 344: 336-339, 1990.

Luke RG. Natriuresis and chloruresis during hydropenia in the rat. Am J Physiol 224: 13-20, 1973.

Maeda Y, Terada Y, Nonoguchi H, and Knepper MA. Hormone and autacoid regulation of cAMP production in rat IMCD subsegments. Am J Physiol Renal Fluid Electrolyte Physiol 263: F319-F327, 1992.

Malpas SC and Leonard BL. Neural regulation of renal blood flow: a re-examination. Clin Exp Pharmacol Physiol 27: 956-964, 2000.

Milhaud PG, Pondugula SR, Lee JH, Herzog M, Lehouelleur J, Wangemann P, Sans A, and Marcus DC. Chloride secretion by semicircular canal duct epithelium is stimulated via 2 -adrenergic receptors. Am J Physiol Cell Physiol 283: C1752-C1760, 2002.

Molinoff PB. - And -adrenergic receptor subtypes properties, distribution and regulation. Drugs 28, Suppl 2: 1-15, 1984.

Morales MM, Falkenstein D, and Lopes AG. The cystic fibrosis transmembrane regulator (CFTR) in the kidney. An Acad Bras Cienc 72: 399-406, 2000.

Naray-Fejes-Toth A and Fejes-Toth G. The sgk, an aldosterone-induced gene in mineralocorticoid target cells, regulates the epithelial sodium channel. Kidney Int 57: 1290-1294, 2000.

Nellgard P, Bojo L, Jonsson A, and Cassuto J. Influence of adrenoceptor agonists and antagonists on fluid secretion in small bowel obstruction. Eur J Gastroenterol Hepatol 7: 1059-1063, 1995.

Novak I. -Adrenergic regulation of ion transport in pancreatic ducts: patch-clamp study of isolated rat pancreatic ducts. Gastroenterology 115: 714-721, 1998.

Pettinger WA, Smyth DD, and Umemura S. Renal 2 -adrenoceptors, their locations and effects on sodium excretion. J Cardiovasc Pharmacol 7: S24-S27, 1985.

Regan JW, Kobilka TS, Yang-Feng TL, Caron MG, Lefkowitz RJ, and Kobilka BK. Cloning and expression of a human kidney cDNA for an 2 -adrenergic receptor subtype. Proc Natl Acad Sci USA 85: 6301-6305, 1988.

Rocha AS and Kudo LH. Factors governing sodium and chloride transport across the inner medullary collecting duct. Kidney Int 38: 654-667, 1990.

Rouch AJ and Kudo LH. 2 -Adrenergic-mediated inhibition of water and urea permeability in the rat IMCD. Am J Physiol Renal Fluid Electrolyte Physiol 271: F150-F157, 1996.

Shreeve SM, Fraser CM, and Venter JC. Molecular comparison of 1 - and 2 -adrenergic receptors suggests that these proteins are structurally related "isoreceptors." Proc Natl Acad Sci USA 82: 4842-4846, 1985.

Slivka SR and Insel PA. 1 -Adrenergic receptor-mediated phosphoinositide hydrolysis and prostaglandin E 2 formation in Madin-Darby canine kidney cells. Possible parallel activation of phospholipase C and phospholipase A 2. J Biol Chem 262: 4200-4207, 1987.

Smyth DD, Umemura S, and Pettinger WA. 2 -Adrenoceptor antagonism of vasopressin-induced changes in sodium excretion. Am J Physiol Renal Fluid Electrolyte Physiol 248: F767-F772, 1985.

Sonnenberg H. Secretion of salt and water into the medullary collecting duct of Ringer-infused rats. Am J Physiol 228: 565-568, 1975.

Stanton BA. Molecular mechanisms of ANP inhibition of renal sodium transport. Can J Physiol Pharmacol 69: 1546-1552, 1991.

Stokes JB and Sigmund RD. Regulation of rENaC mRNA by dietary NaCl and steroids: organ, tissue, and steroid heterogeneity. Am J Physiol Cell Physiol 274: C1699-C1707, 1998.

Stuart-Tilley AK, Shmukler BE, Brown D, and Alper SL. Immunolocalization and tissue-specific splicing of AE2 anion exchanger in mouse kidney. J Am Soc Nephrol 9: 946-959, 1998.

Teitelbaum I, Strasheim A, and Berl T. Adrenergic control of cAMP generation in rat inner medullary collecting tubule cells. Kidney Int 35: 647-653, 1989.

Umemura S, Marver D, Smyth DD, and Pettinger WA. 2 -Adrenoceptors and cellular cAMP levels in single nephron segments from the rat. Am J Physiol Renal Fluid Electrolyte Physiol 249: F28-F33, 1985.

Vandorpe D, Kizer N, Ciampollilo F, Moyer B, Karlson K, Guggino WB, and Stanton BA. CFTR mediates electrogenic chloride secretion in mouse inner medullary collecting duct (mIMCD-K2) cells. Am J Physiol Cell Physiol 269: C683-C689, 1995.

Vandorpe DH, Ciampolillo F, Green RB, and Stanton BA. Cyclic nucleotide-gated cation channels mediate sodium absorption by IMCD (mIMCD-K2) cells. Am J Physiol Cell Physiol 272: C901-C910, 1997.

Wall SM, Sands JM, Flessner MF, Nonoguchi H, Spring KR, and Knepper MA. Net acid transport by isolated perfused inner medullary collecting ducts. Am J Physiol Renal Fluid Electrolyte Physiol 258: F75-F84, 1990.

Wallace DP, Christensen M, Reif G, Belibi F, Thrasher B, Herrell D, and Grantham JJ. Electrolyte and fluid secretion by cultured human inner medullary collecting duct cells. Am J Physiol Renal Physiol 283: F1337-F1350, 2002.

Wallace DP, Grantham JJ, and Sullivan LP. Chloride and fluid secretion by cultured human polycystic kidney cells. Kidney Int 50: 1327-1336, 1996.

Wallace DP, Rome LA, Sullivan LP, and Grantham JJ. cAMP-dependent fluid secretion in rat inner medullary collecting ducts. Am J Physiol Renal Physiol 280: F1019-F1029, 2001.

Wallace DP, Tomich JM, Eppler JW, Iwamoto T, Grantham JJ, and Sullivan LP. A synthetic channel-forming peptide induces Cl - secretion: modulation by Ca 2+ -dependent K + channels. Biochim Biophys Acta 1464: 69-82, 2000.

Wilborn TW, Sun D, and Schafer JA. Expression of multiple -adrenoceptor isoforms in rat CCD. Am J Physiol Renal Physiol 275: F111-F118, 1998.

Wingo CS. Active and passive chloride transport by the rabbit cortical collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 258: F1388-F1393, 1990.

Yamaguchi T, Pelling JC, Ramaswamy NT, Eppler JW, Wallace DP, Nagao S, Rome LA, Sullivan LP, and Grantham JJ. cAMP stimulates the in vitro proliferation of renal cyst epithelial cells by activating the extracellular signal-regulated kinase pathway. Kidney Int 57: 1460-1471, 2000.

Yasuda G and Jeffries WB. Regulation of cAMP production in initial and terminal inner medullary collecting ducts. Kidney Int 54: 80-86, 1998.

Yasuda G, Sun L, Lee HC, Umemura S, and Jeffries WB. The 1 - and 2 -adrenoceptor subtypes in cultured rat inner medullary collecting duct cells. Am J Physiol Renal Fluid Electrolyte Physiol 271: F762-F769, 1996.

Yasuda G, Sun L, Umemura S, Pettinger WA, and Jeffries WB. Characterization of prazosin-sensitive 2 B-adrenoceptors expressed by cultured rat IMCD cells. Am J Physiol Renal Fluid Electrolyte Physiol 261: F760-F766, 1991.

Yasuda G, Umemura S, and Jeffries WB. Effect of epinephrine on cAMP accumulation in cultured rat inner medullary collecting duct cells. Am J Physiol Renal Physiol 272: F192-F197, 1997.

Zeidel ML. Hormonal regulation of inner medullary collecting duct sodium transport. Am J Physiol Renal Fluid Electrolyte Physiol 265: F159-F173, 1993.

Zeidel ML. Renal actions of atrial natriuretic peptide: regulation of collecting duct sodium and water transport. Annu Rev Physiol 52: 747-759, 1990.

Ziegler MG, Aung M, and Kennedy B. Sources of human urinary epinephrine. Kidney Int 51: 324-327, 1997.

作者单位：1 Kidney Institute and Departments of 2 Internal Medicine and 3 Urology, University of Kansas Medical Center, Kansas City, Kansas 66160