Proc. Natl. Acad. Sci. USA Vol. 83, pp. 4769-4773, July 1986 Cell Biology Atrial natriuretic peptide inhibits renin release from juxtaglomerular cells by a cGMP-mediated process ARMIN KURTZ*, ROBERTO DELLA BRUNA*, JOSEF PFEILSCHIFTER*, ROLAND TAUGNERt, AND CHRISTIAN BAUER* *Physiologisches Institut der Universitit Zfrich, Winterthurerstrasse 190, 8057 Zurich, Switzerland; and tPhysiologisches Institut der Universitat Heidelberg, Im Neuenheimer Feld 326, D-6900 Heidelberg, Federal Republic of Germany Communicated by Robert E. Forster II, March 3, 1986 ABSTRACT We have exmined the effect of a synthetic MATERIALS AND METHODS analogue of human et-atrial natriuretic peptide (ANP), APH, on renin release in cultured renal juxtaglomerular cells (JGA Cell Culture. Isolation ofjuxtaglomerular cells was essen- cells). Using cell cultures containing 80-90% renal juxtaglom- tially as described (8, 9). In brief, rat kidneys were perfused erular cells, we found that ANP (10-13-10-9 M) strongly in situ with citrate buffer. After exstirpation of the kidneys, inhibited renin release from the cells in a dose-dependent renal cortex was minced and subsequently incubated with a fashion (ki, 10 pM) to about 10% of control. Inhibition of renin collagenase-trypsin solution. The suspension was poured release by ANP was paralleled by an increase in cellular cGMP over a 22-ptm screen, and single cells passing the sieve were levels; while in the presence of the cGMP-phosphodiesterase washed with culture medium. As described (9), cell cultures inhibitor M&B 22948 (1 mM), concentrations of ANP lower by containing about 50% juxtaglomerular cells can be obtained a factor of 100 were required to obtain the same effects on renin by culturing these single cells. To improve the enrichment of release and cGMP levels. The guanylate cyclase inhibitor juxtaglomerular cells, we centrifuged the cells in a Percoll- methylene blue (10 IAM), on the other hand, shifted the density gradient. About 15 x 106 cells were mixed with 30 ml dose-response curves for renin release and cGMP levels to of an isotonic 25% (vol/vol) Percoll solution and centrifuged 100-fold higher concentrations of ANP. Neither the influx of at 15,000 x g in a vertical rotor (SV-288, Sorvall) for 30 min. 45Ca into the cells nor the intracellular quin-2 signal, which is Four different bands with densities of 1.02 g/ml, 1.05 g/ml, a measure for changes ofintracellular Ca concentration, was in 1.06 g/ml, and 1.13 g/ml were regularly obtained. Table 1 any way altered by ANP. Our results suggest that ANP inhibits gives typical values for the distribution ofprotein and specific renin release fromjuxtaglomerular cells by a cGMP-dependent renin activity within the gradient. Band III cells (1.06 g/ml) process that does not involve changes in intracellular calcium. were incubated with culture medium [RPMI 1640, 25 mM Hepes, streptomycin at 100 tug/ml, penicillin at 100 units/ml, Two major effects ofatrial natriuretic peptide (ANP) on renal 2% (vol/vol) fetal bovine serum] at 370C in 5% C02/95% air. function have been described, namely the enhancement of All experiments as described below were performed on the sodium and water excretion (cf. refs. 1 and 2) and the second day of culture. suppression of renin release (3-5). Inhibition of renin release Renin Demonstration by Immunofluorescence. Immuno- by ANP leading to a diminished formation of angiotensin II fluorescence staining for renin was done exactly as described (AI) would result in decreased vascular tone and impaired (9). In brief, cell cultures were fixed in 4% (wt/vol) aldosterone secretion. Both effects are of physiological paraformaldehyde/Dulbecco's phosphate-buffered saline importance during states of expanded extracellular volume, (PBS). After 5 min in 10% (vol/vol) normal swine in which enhanced release of ANP from the atrium is serum/0.1% bovine serum albumin/PBS, the cultures were observed (6, 7). incubated with rabbit antiserum against rat renin diluted The mechanism by which ANP inhibits renal renin release 1:200 in 10% (vol/vol) normal swine serum/PBS. After is as yet a matter of debate (2). In view of the increased rinsing in PBS, a 30-min incubation followed with tetra- filtration rate of NaCl caused by the ANP, which leads to an methylrhodamine B isothiocyanate (TRITC)-conjugated goat increased NaCl load at the macula densa, it has been anti-rabbit IgG (1:500, Sigma). The specificity of the im- speculated that ANP inhibits renin release fromjuxtaglomer- munoreaction was tested with the usual control procedures ular cells via the macula densa receptor (3, 4). However, a (9). Immunofluorescence was examined using a Polyvar direct effect of ANP on juxtaglomerular cells could not be fluorescence microscope (Reichert-Jung, Vienna, Austria). excluded on the basis of experimental evidence. Therefore, Rabbit antiserum against rat renin, used in the studies, was we wanted to investigate whether or not ANP directly a generous gift of M. Celio (Zurich) who obtained it from T. inhibits renin release by renal juxtaglomerular cells. Inagami (Nashville, TN). For immunocytochemical demon- We have developed a cell culture system that contains stration and localization with the protein A/gold technique, around 50% juxtaglomerular cells (8, 9). We now have pellets of 106 cells were fixed in 1% glutaraldehyde and improved this method by including a Percoll-density gradient embedded in London White resin (London Resin, Basing- in the cell preparation procedure. As a result, cell cultures stoke, England). Ultrathin sections were immunostained for can be obtained that contain regularly between 80 and 90% renin as described by Taugner et al. (10). Rabbit antiserum juxtaglomerular cells. against rat renin used in these studies was a generous gift We found that ANP strongly inhibits renin release from from E. Hackenthal. these cultured cells and, furthermore, obtained strong evi- Renin Release. Renin release from the cultured cells was dence that the inhibitory effect ofANP is mediated by cGMP determined exactly as described (8). In brief, the culture and does not involve an increase in intracellular calcium. medium was replaced with prewarmed, Hepes-buffered sa- line [132 mM NaCl, 5 mM KCl, 0.8 mM MgSO4, 2 mM CaCl2, The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" Abbreviations: ANP, atrial natriuretic peptide; Al, angiotensin I; in accordance with 18 U.S.C. §1734 solely to indicate this fact. All, angiotension II; [Ca]j, intracellular Ca2+ concentration. 4769 Downloaded by guest on September 29, 2021 4770 Cell Biology: Kurtz et al. Proc. Natl. Acad. Sci. USA 83 (1986) Table 1. Protein distribution and specific renin activity of cells separated by a 25% (vol/vol) Percoll gradient Specific renin activity, Density, Protein, ng of AI per hr Band g/ml mg per mg of protein I 1.02 2.01 ± 0.35 30 ± 11 II 1.05 3.36 ± 0.75 143 ± 60 III 1.06 0.51 ± 0.19 1170 ± 280 IV 1.13 3.27 ± 0.62 7 ± 5 Data are mean ± SEM of three experiments. 10 mM NaOAc, 2 mM NaH2PO4, 10 mM glucose, 20 mM Hepes (pH 7.2)]; and the culture dishes were placed on a heating block at 370C. The renin secretion rate was calculated from the linear increase of the renin activity of the cell- conditioned buffer before and 10, 20, and 30 min after addition of the agents. Renin Activity Assay. Renin activity was determined by its ability to generate angiotensin I (AI) from the plasma of bilaterally nephrectomized rats exactly as described (11). AI was determined by radioimmunoassay (Isotopen Dienst West, Teufen, Switzerland). Intracellular cGMP Levels. Intracellular cGMP concentra- tions were examined under the same experimental conditions as the renin release. Five minutes after the addition of the agents, the buffer was removed from the cultures, and the dishes were placed on an ice block. After the addition of 0.4 ml of ice-cold 5 mM potassium phosphate, 0.2 mM EDTA, 0.5 mM 3-isobutyl-methylxanthine, and 150 mM KCl (pH 6.8), the cells were scraped off with a Teflon policeman. The FIG. 1. Microphotograph of cultured cells on second day of cell suspension so obtained was sonicated, boiled for 5 min, culture. (Upper) Phase-contrast and (Lower) immunofluorescence and centrifuged. An aliquot was removed from the sonicated micrographs of cells treated with rabbit anti-rat renin and TRITC- cell suspension for protein determination. The supernatants labeled goat anti-rabbit y-globulin. (x2902.) were assayed for cGMP using a cGMP radioimmunoassay (New England Nuclear). have also shown elsewhere (9), the round cells that stain for 45Ca-Uptake. 45Ca-uptake into the cultured cells was de- renin contain prominent renin granules (Fig. 2). We, there- termined exactly as described (8). fore, conclude that these cells are juxtaglomerular cells. The Intracellular Ca2' Measurement. Intracellular Ca2+, [Ca]i, cellular renin activity of the cultured cells used in this study was measured using quin-2. About 2 x 107 cells were was between 100-200 ng ofAl per hr per mg ofprotein on the incubated with 25 uM quin-2 AM [tetrakis(acetoxymethyl) second day of culture. The renin activity in the culture ester of quin-2] in RPMI 1640 for 20 min followed by another medium was between 1 and 2.6 ,g of AI per hr per mg of 40-min incubation with 4 vol ofmedium. After the incubation cellular protein. period, aliquots of 1 x 106 cells were washed twice and Effect of ANP on the Cultured Cells. Fig. 3 shows the effect resuspended in Hepes-buffered saline. Fluorescence ofquin- of ANP on the spontaneous renin release from the cultured 2-loaded cells was measured at 37°C in a Perkin-Elmer cells.