Background

-Adrenergic receptors (-ARs) are known to participate in the regulation of glomerular filtration, NaCl reabsorption, acid-base balance, and renin secretion; however, the precise histologic localization of -AR at putative signaling sites involved in these processes remains an open issue.

Methods

We used a set of subtype-specific rabbit antibodies to visualize ₁- and ₂-AR in rat kidney by immunohistochemistry and specified cells and segments of the nephron thought to be regulated by catecholamines. In addition, the relative proportion of -AR subtypes in cortical and medullary portions of rat kidney was determined by Western blotting and by competing [¹²⁵I]-cyanopindolol binding with the ₁- or ₂-selective antagonists bisoprolol or ICI 118,551, respectively.

Results

Immunoreactivity for ₁-AR was found in mesangial cells, juxtaglomerular granular cells, the macula densa epithelium, proximal and distal tubular segments, and acid-secreting type A intercalated cells of the cortical and medullary collecting ducts. Immunoreactivity for ₂-AR was predominantly localized in the apical and subapical compartment of proximal and, to a lesser extent, distal tubular epithelia (suggesting interactions with luminal fluid catecholamines). Both subtypes were dense in the membranes of smooth muscle cells from renal arteries. Concordant data were obtained by radioligand binding and immunoblotting of membranes prepared from cortical and medullary portions of the kidney.

Conclusion

Our data provide an immunohistochemical basis for the cellular targets of -adrenergic regulation of renal function. Moreover, they could help to devise therapeutic strategies directed at renal -ARs.

METHODS

Antibodies

Production and characterization of subtype-specific rabbit antibodies against the human ₁- and ₂-AR have previously been published¹³. Here, we used affinity-purified antibodies directed against the aminoterminal (extracellular) domains of human ₁- or ₂-AR. The receptors are highly conserved between rats^14,15 and humans^16,17 with a calculated amino acid sequence identity of 93% for the ₁-AR and 89% for the ₂-AR, respectively. To control for specificity, the antibodies were preincubated (12 hours, 4°C) with baculovirus-infected Sf9 insect cells expressing 1 to 2 10⁶ intact recombinant ₁- or ₂-AR per cell^13,18,19. The unbound antibodies remaining in the supernatant after sedimentation (900 g, 4°C, 10 min) of Sf9 cells expressing the corresponding -AR subtype or not were used for these experiments. Immunolabeling of ₁- and ₂-AR was assigned to specific segments and cell types of the nephron by simultaneous or comparative immunostaining of various marker proteins. Calbindin (cytosolic vitamin D-dependent Ca²⁺ binding protein) was detected with a mouse monoclonal antibody D-28K (Sigma, Deisenhofen, Germany). Tamm-Horsfall protein was detected with a rabbit polyclonal antibody (Biotrend, Köln, Germany). Renin was detected by a rat antiserum, kindly provided by Professor A. Kurtz (Institute of Physiology, University of Regensburg, Germany). Polyclonal rabbit antibodies against the endothelial nitric oxide synthetase were obtained from Professor H. Schmidt (Institute of Pharmacology, University of Würzburg, Germany). A mouse monoclonal antibody directed against synaptopodin was a gift from Dr. P. Mundel (Albert Einstein College, New York, NY, USA). Specificity of a polyclonal rabbit antibody against rat anion exchanger 1 (band 3 protein) has been published elsewhere²⁰.

Western blots analysis

The kidneys of anesthetized male Sprague-Dawley rats (350 to 400 g) were removed and dissected into cortical, outer, and inner medullary portions. The different portions were homogenized in ice-cold buffer [10 mmol/L K₂HPO₄, 1 mmol/L ethylenediaminetetraacetic acid (EDTA), 1 mmol/L EGTA] containing various protease inhibitors (5 g/mL leupeptine, 1.5 mmol/L benzamidine, 200 U/mL aprotinine, 2 g/mL pepstatine A). Samples of 60 g membrane protein of each of these portions were analyzed by Western blot using affinity-purified rabbit antibodies against ₁- or ₂-AR diluted 1:1000¹³. Immunoreactive bands were visualized by horseradish peroxidase-conjugated secondary antibodies (diluted 1:5000; Dianova, Hamburg, Germany) and enhanced chemiluminescence (ECL; Amersham, Little Chelford, UK). Baculovirus-infected Sf9 insect cells expressing recombinant ₁- or ₂-AR served as controls^13,18,19.

Radioligand binding

All binding experiments were performed using the nonselective -AR antagonist [¹²⁵I]-iodocyanopindolol (2200 Ci/mmol [¹²⁵I]-CYP; NEN, Zaventem, Belgium). Samples of 50 g membrane protein were incubated (30 min, 30°C) each with 400 pmol/L [¹²⁵I]-CYP. Nonspecific binding was determined in the presence of 10 mol/L unlabeled L-propanolol. To determine the relative proportion of either ₁- or ₂-AR, incubations with 100 pmol/L [¹²⁵I]-CYP were supplemented with various concentrations of unlabeled receptor antagonists selective for ₁- (bisoprolol; Merck, Darmstadt, Germany) or ₂-AR (ICI 118,551; RBI, Deisenhofen, Germany). The reactions were stopped by adding an ice-cold buffer (0.1 mol/L Tris-HCl, pH 7.5) and rapid filtration (Whatman GF/C soaked in 0.3% polyethylenimine). Filter-bound radioactivity was measured by gamma counting. Ligand binding curves were fitted to the data by computer-aided nonlinear regression analysis.

Immunohistochemistry

Large structures (that is, blood vessels and glomeruli) were investigated in acetone-fixed kidney sections: Rat kidneys were rapidly frozen in isopentane at -56°C and cryostat sections of 3 to 4 m were fixed with cold acetone (4 min, -20°C). To investigate tubular structures, the kidneys were fixed in situ by formaldehyde perfusion, which provides a better preservation of the cell structures. The kidneys of anesthetized rats were perfused through an aortic cannula (placed below the level of the kidneys), first with 4% formaldehyde in phosphate-buffered saline (PBS) for five minutes followed by 0.8 mol/L sucrose in PBS for another five minutes. Subsequently, the kidneys were removed and treated as described previously in this article. Fixed kidney sections were rinsed three times with ice-cold PBS, blocked (30 min, 21°C) with PBS containing 4% bovine serum albumin (BSA), and 2% normal goat serum (Sigma), and incubated (12 h, 4°C) with affinity-purified receptor antibodies diluted 1:200 in PBS containing 2% BSA and 2% goat serum. After thorough rinsing with PBS, tissue-bound antibodies were detected with appropriate species-specific secondary antibodies (Dianova) conjugated to CY3 or CY2 (diluted 1:800 or 1:400, respectively). After rinsing again with PBS (3 times 10 min, 21°C), the slides were mounted in mowiol (Sigma) and inspected at 200- to 630-fold magnification under an epifluorescence microscope (Carl Zeiss, Oberkochen, Germany). Images were photographed with Kodak TMAX 400 black and white negative films. High-resolution images of cellular structures were obtained with a cooled CCD camera (Sensys, Photometrics, München, Germany) equipped with an additional fourfold magnification lens, giving a final magnification of 2520-fold.

Statistics

Radioligand-binding experiments were analyzed by computer-aided nonlinear regression analysis (LIGAND-program)²¹. All data are given as mean SD. Where quantitative statistics could not be applied (immunofluorescence and immunoblotting), the examples given are representative of at least three independent experiments done on different days and with kidneys obtained from different animals. Unless otherwise stated, fluorescent images are representative of the respective area of a section inspected in at least ten separate fields of view. The schemes given throughout this article depict simplified nephrons not drawn to scale (modified from Kriz and Kaissling²²) and do not represent the exact anatomy of the rat nephron (that is, short looped vs. long looped).

RESULTS

Characterization of ₁- and ₂-adrenergic receptors in rat kidney by radioligand binding and immunoblotting

To define our experimental system, we determined the amount and subtype distribution of -ARs in rat kidney by conventional radioligand binding. For this purpose, we separated cortical, outer, and inner medullary portions of the kidney by macroscopic dissection, isolated the cell membranes from these portions, and measured the overall concentration of membrane-bound -AR by using the nonselective receptor antagonist [¹²⁵I]-cyanopindolol. In agreement with previous reports^7,23, we observed the highest density of -AR in cortical membranes (53%) and somewhat lower receptor densities in membranes prepared from the outer medulla (39%; Table 1). Because only a rather faint amount of -AR was found in the inner medulla (8%), in this portion, a further subtyping of the receptors was not meaningful. In the cortex and outer medulla, the proportion of ₁- and ₂-AR subtypes was determined by competing [¹²⁵I]-cyanopindolol binding with unlabeled subtype-selective receptor antagonists, that is, using bisoprolol for the ₁-AR²⁴ and ICI 118,551 for the ₂-AR subtype, respectively. In each case, there were biphasic displacement curves suggestive of a two-sited binding model with a high- and a low-affinity component, which is in perfect agreement with previous radioligand binding studies on rat tissues^25,26. The relative proportions of ₁- and ₂-AR derived from these curves differed slightly between cortex (₁:₂ 65:35%) and medulla (₁:₂ 70:30%; Table 1), with both subtypes being more frequent in cortical (₁ = 56% and ₂ = 62%) than in medullary membranes (₁ = 44% and ₂ = 38%; Table 1). It should be noted that within a same kidney zone, similar proportions of ₁- and ₂-AR were obtained with bisoprolol and with ICI 118,551.

Table 1 - Binding of -adrenergic receptor ligands to membranes prepared from rat kidney.

Full table

From the subtype-specific receptor antibodies that were raised against selected extramembraneous domains of the human ₁- or ₂-AR¹³, those directed against the amino-terminal domains exhibited the highest degree of specificity and subtype selectivity in cell systems expressing recombinant human -AR. Since these domains are highly conserved between humans^16,17 and rats^14,15, we expected that these antibodies would be suitable for immunohistological studies on rat tissues. However, before embarking on such studies, this assumption was tested by probing Western blots of membranes prepared from cortical and medullary portions of rat kidney. Antibodies against the ₁-AR (anti–₁-AR) labeled a single protein band with an apparent size of 68 kD (Figure 1, lanes 1 through 3), which is in good agreement with the molecular mass deduced from the amino acid sequence of the rat ₁-AR¹⁴, when taking into account some variations in electrophoretic mobility caused by glycosylation²⁷. The apparent decrease in signal intensity from the cortex to the inner medulla is in good correlation with the receptor densities in the corresponding kidney zones determined by radioligand binding Table 1 and reflects the different expression of this receptor subtype in the corresponding kidney zones. Antibodies against the ₂-AR (anti–₂-AR) labeled a protein band of slightly higher mobility (Figure 1, lanes 6 through 8), which is also plausible in the light of the (slightly shorter) amino acid sequence of the rat ₂-AR¹⁵ when taking protein glycosylation into account. In this context, it should be noted that -AR are known to be less glycosylated when expressed in Sf9 insect cells^13,18, giving protein bands with apparent sizes of 51 and 47 kD for the recombinant human ₁- and ₂-AR, respectively (Figure 1, lanes 4 and 9). These data indicated that these antibodies did recognize rat ₁- and ₂-AR, and thus they appeared to be suitable for an immunohistological study on rat kidney.

Figure 1.

Immunoblots of ₁- and ₂-adrenergic receptors (₁- and ₂-AR) in rat kidney. Membranes prepared from cortical (lanes 1 and 6), outer medullary (lanes 2 and 7), and inner medullary (lanes 3 and 8) portions of rat kidney were subjected to Western blotting and probed with antibodies against ₁-AR (lanes 1 through 5) or ₂-AR (lanes 6 through 10). Bound antibodies were detected by horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence. Membranes prepared from Sf9 cells expressing recombinant ₁- (49 kD; lane 4 and 10) or ₂-AR (47 kD; lane 5 and 9) served as specificity controls (note that receptor glycosylation is known to be different and heterogenous in Sf9 insect cells)^13,27. Apparent molecular sizes (kD) are denoted on the right margin.

Full figure and legend (20K)

Immunolocalization of ₁- and ₂-adrenergic receptors in rat kidney

Procedural strategy

Initial experiments were carried out on cryostat sections (3 to 4 mm) of rat kidney, which were subsequently fixed with acetone. Staining of such sections with antibodies against -AR led to a bright and distinct labeling of vascular and glomerular structures, which could readily be interpreted at the cellular level. Certain tubular structures within the cortical and medullary nephron segments were also labeled by the antibodies, but were not sufficiently conserved for an interpretation at the cellular level. Immunohistochemistry of these structures had to be carried out on sections of kidneys fixed in situ by formaldehyde perfusion. However, in such perfusion-fixed sections, the staining of vascular and glomerular structures was generally much less intense, most likely because these tissue compartments are more strongly fixed by vascular perfusion than extravascular tissues.

Blood vessels

The cell types most brilliantly stained by antibodies against ₁- and ₂-AR in rat kidney sections were the smooth muscle cells in the walls of large blood vessels. Figure 2a shows a representative oblique section through an arcuate artery stained with anti–₁-AR. Immunoreactivity of smooth muscle in a consecutive section was largely abolished by preadsorption with recombinant ₁-AR expressed in Sf9 insect cells Figure 2b. At higher magnification, the antibody label was seen to be mainly confined to small dot-like structures located at the cell surface (Figure 2d, arrows), which corresponds to the -AR staining pattern previously observed in cultured human epidermoid carcinoma A 431 cells¹³. Vascular smooth muscle cells were also stained by anti–₂-AR (albeit somewhat less intense; Figure 2c), and this staining could likewise be abolished by preadsorption with recombinant ₂-AR (data not shown).

Figure 2.

₁- and ₂-AR in large renal arteries. Consecutive sections (3 m) of rat kidney were fixed with acetone, incubated with antibodies against ₁- (a, b, and d) or ₂-AR (c) and counterstained with CY3-labeled secondary antibodies. Specificity of the staining was confirmed by preadsorption with recombinant -AR expressed in Sf9 cells (b). Images focus on the same arcuate artery of the renal cortex in consecutive sections. The enlargement (d) demonstrates the dot-like distribution pattern of ₁-AR in the membranes of smooth muscle cells (arrows).

Full figure and legend (222K)

Glomeruli

By immunohistochemistry, several locations within and adjacent to the glomerulus could be identified, where ₁-AR appeared to be clustered: a very strong immunostaining by anti–₁-AR was observed in the juxtaglomerular portion of the afferent arteriole Figure 3a. These immunofluorescent signals could be clearly assigned to the renin-producing juxtaglomerular granular cells themselves by staining of serial sections with renin-antibodies (Figure 3, compare circled areas in a and b). In addition, ₁-AR appeared to be densely expressed in the walls of arterioles entering the glomeruli, which could be assigned to the afferent arterioles by their internal elastic membrane (visible by green autofluorescence). Anti–₁-AR also highlighted banded structures in the glomerular tuft, which seemed to be interspaced between the capillaries Figure 4a, c, e, g, i. These signals seemed to be specific for ₁-AR in as much as they could be abolished by preadsorption with recombinant ₁-AR (Figure 4, compare c and d). Such structures were not labeled by anti–₂-AR (Figure 4, compare a and b), which reacted with only certain tubular structures adjacent to the glomerulus, as will be discussed later. Moreover, the banded pattern of ₁-AR in the glomerular tuft was clearly different from that obtained with antibodies against endothelial nitric oxide synthetase (eNOS) as a specific marker of capillaries (Figure 4, compare e with f) and could not be assigned to podocytes by counterstaining with synaptopodin antibodies (Figure 4, compare g with h). Considering that within the glomerular tuft, only three cell types are found (endothelial cells, podocytes, and mesangial cells)²² and that the signal obtained with anti–₁-AR could not be associated with endothelial cells or with podocytes, most likely it is associated with mesangial cells. This argument is further supported by high resolution images of the glomerular capillary tuft Figure 4i, k, suggesting that ₁-AR are situated in the membranes of mesangial cells that are clustered in the intercapillary space and extend processes joining the capillary walls. Most likely, the latter structures correspond to mesangial cell processes extending toward the peripherally located capillaries (Figure 4, compare i and k).

Figure 3.

Colocalization of ₁-AR and renin in the juxtaglomerular region. Consecutive sections (3 m) of rat kidney were fixed with acetone, incubated with antibodies against ₁-AR (a) or renin (b), and counterstained with CY3-labeled secondary antibodies. Images focus on an afferent arteriole entering its parent glomerulus. The renin-producing juxtaglomerular granular cells in the wall of the afferent arteriole are shown in the circle (R).

Full figure and legend (318K)

Figure 4.

Assignment of ₁- and ₂-AR to glomerular cell types. Each line shows the same glomerulus in two consecutive sections (3 m) of rat kidney fixed with acetone. One section of each horizontal pair was incubated with anti–₁-AR and is shown on the left (a, c, e, g, and i). The corresponding section on the right was incubated with anti–₂-AR (b), anti–₁-AR preadsorbed with recombinant ₁-AR from Sf9 cells (d, specificity control), antibodies against endothelial NO synthetase (f), or antibodies against synaptopodin (h). The preparations were counterstained with CY3- (a–g) or CY2-labeled (h) secondary antibodies. At high magnification (i), it can be seen that the ₁-AR staining appears to be restricted to mesangial cells [M] located in the intercapillary space, which are characterized by cell processes [P] that extend toward the capillaries [C]. The simplified scheme on the right (k, modified after⁷¹) shows three glomerular capillaries [C] attached to the centrally located mesangium. The podocytes and the endothelium are shown in white. The glomerular basement membrane is depicted in black, and the mesangial cell in dark gray (with a cell matrix-containing fibrillar structures). Tongue-like mesangial cell processes [P] form connections between the glomerular basement membrane at two opposing mesangial angles.

Full figure and legend (196K)

Proximal convoluted tubule

Proximal convoluted tubules (PCTs) in proximity and/or in direct connection to the glomerulus (that is, the tubular S1/2 segments) were distinguished from distal tubular segments by the absence of calbindin Figure 5b. Among these tubules, we obtained only a weak staining with anti–₁-AR, but rather strong signals with anti–₂-AR Figure 5a, which decreased gradually within the S3 segment (proximal straight tubule, located in the deep portion of the medullary rays) and in the outer stripe of the outer medulla (OS OM; Figure 5c). At high resolution, it can be seen that the labeling by anti–₂-AR was clearly more pronounced at the apical (that is, at the base of the epithelial brush border and within the subapical endocytotic compartment; Figure 5d) compared with the basolateral pole of PCT cells Figure 5e. The staining appeared to be highly specific because it could be preadsorbed with recombinant ₂-AR Figure 5g, but not ₁-AR Figure 5f. To exclude a nonspecific reactivity of anti–₂-AR with microvilli, Western blots and histologic sections of rat duodenum were probed (data not shown). Apical membranes of rat duodenum were not labeled by anti–₂-AR, thus confirming the specificity of the staining obtained with our antibodies. Anti–₁-AR gave a weak dot-like staining pattern on both basolateral and apical membranes of PCT cells, but the highest amount of ₁-AR appeared to be concentrated within the subapical endocytotic compartment Figure 5e, probably corresponding to internalized (apical) receptors, as will be discussed later. Taken together, it appears that the proximal tubular epithelia in rat kidney are dominated by ₂-AR rather than ₁-AR and that these receptors are preferentially located in the apical and subapical compartment of PCT cells [which is in agreement with recent functional studies on isolated rat proximal tubular epithelial (PTE) cells]^28,29.

Figure 5.

Localization of ₂-AR in the proximal tubules. Sections (4 to 5 m) of the outer and inner cortex were obtained from rat kidneys fixed in situ by perfusion with formaldehyde. The top row shows a double staining of the outer renal cortex with antibodies against ₂-AR (a; tubular S1/2 segments; PCT, proximal convoluted tubule; G, glomerulus) and calbindin (b; DCT, distal convoluted tubule). In the inner cortex (c; tubular S2/3 segments, transition between cortex and outer stripe of the outer medulla, OS OM), the ₂-AR staining decreases gradually along the proximal straight tubules (S3 segment). At high resolution (d), it can be seen that the ₂-ARs are preferentially expressed in the apical membranes of the cells constituting the proximal convoluted tubule, whereas only a faint amount of ₂-AR can be detected in the basolateral membranes. At the same scale, section (e) demonstrates the distribution of ₁-AR in PCT cells, with the highest receptor concentration in the subapical endocytotic compartment. (f and g) Consecutive sections of the cortex stained with anti–₂-AR after preadsorption with recombinant ₁- (f) or ₂-AR from Sf9 cells (g, specificity control).

Full figure and legend (205K)

Thick ascending limb of Henle's loop (TAL segment) and macula densa

Figure 6 shows a section through the border area between renal cortex and the OS OM (Figure 6 a and e for orientation markers). Three different nephron-segments are depicted: (1) the proximal tubules, which were strongly labeled by anti–₂-AR, as already described previously Figure 6d, (2) the thick ascending limb (TAL) segments identified by the expression of Tamm-Horsfall protein³⁰ Figure 6c, and (3) the distal nephron segments (that is, the distal convoluted and the connecting tubules), which are labeled by calbindin Figure 6b and also strongly labeled by anti–₁-AR Figure 6a.

Figure 6.

Localization of ₁- and ₂-AR in the thick ascending limb (TAL). Sections (4 through 5 m) were obtained from perfusion-fixed rat kidneys. The top row shows a double staining with antibodies against ₁-AR (a) and calbindin (b). The same area in two consecutive sections was stained with antibodies against Tamm-Horsfall protein (c) and ₂-AR (d), which were detected by CY3- (a, c, and d) or CY2-labeled (b) secondary antibodies. Images focus on the cortical and medullary TAL. Arrows in (a) and (c) highlight the position of ₁-AR in cortical TAL segments that are also positive for Tamm-Horsfall protein (c). The dotted line in (a) represents the border between the cortex and the outer stripe of the outer medulla (OS OM). (e) This corresponds to (a) with inversed colors and indicates the various nephron segments seen in the section that was positioned as schematically shown in (f). Abbreviations are as follows: G, glomerulus; MD, macula densa; PCT, proximal convoluted tubule; PST, proximal straight tubule; TAL, thick ascending limb; and DCT, distal convoluted tubule.

Full figure and legend (145K)

The TAL seems to be devoid of ₂-AR (Figure 6, compare c and d), whereas ₁-ARs show an inhomogeneous pattern of distribution within this nephron segment: The medullary portions of the TAL (Figure 6a, bottom, OS OM) were not stained, whereas the cortical portions were fluorescence positive (Figure 6a, top, arrows), albeit somewhat less intense than the distal nephron segments. Most notably, a dense cluster of ₁-AR staining was observed at sites of contact between the TAL segment and the vascular pole of the glomerulus (corresponding to the macula densa). Figure 7 focuses on the plaque of the macula densa adjacent to the extraglomerular mesangium; at the site of contact, the cells of the luminal membrane of the TAL segment lack Tamm-Horsfall protein (Figure 7b, arrows) and thus are characterized as macula densa cells³⁰. The corresponding immunofluorescent image of ₁-AR Figure 7a shows a clustered appearance of the receptors in the same cells. At high resolution (Figure 7a, insert), the expression of ₁-AR seems to be restricted to the apical surface of the macula densa cells. Figure 7c (₂-AR) and Figure 7d (Tamm-Horsfall protein) attest to the fact that the macula densa cells were not stained by anti–₂-AR (single arrow).

Figure 7.

Expression of ₁- and ₂-AR in the macula densa region. Sections (4 to 5 m) of perfusion-fixed rat kidney were incubated with antibodies against ₁- (a) or ₂-AR (c). Corresponding areas in consecutive sections stained with antibodies against Tamm-Horsfall protein are shown in (b) and (d), respectively. Bound antibodies were detected by CY3-labeled secondary antibodies. Arrows indicate the position of macula densa cells lacking the Tamm-Horsfall protein. The insert in (a) provides an example of the punctate ₁-AR staining in the apical membrane of a macula densa cell attached to the extraglomerular mesangium at higher magnification.

Full figure and legend (164K)

Distal convoluted and connecting tubules

Distal segments of the nephron were identified by calbindin to allow for a differentiation between distal convoluted tubule (DCT; light and uniform staining) and connecting tubules (CNT; strong staining, but interspersed with unstained intercalated cells)^31,32. Figure 8a and b show that the ₁-ARs colocalize with calbindin throughout the section. At a higher resolution, it can be seen that in the CNT (as opposed to the DCT), anti–₁-AR stained not only the calbindin-positive principal cells, but also the apical rather than basolateral membranes of the calbindin-negative intercalated cells (Figure 8a, b, bottom half, arrows). Specificity of the ₁-AR staining was confirmed as already described (Figure 8a, bottom half, insert). In contrast, no significant amounts of ₂-AR could be detected in the CNT, and only a very faint amount of ₂-AR appeared to be expressed in distal convoluted tubular epithelia [Figure 8c (₂-AR) and Figure 8d (calbindin), arrowheads].

Figure 8.

Localization of ₁- and ₂-AR in the distal convoluted (DCT) and connecting tubules (CNT). Sections (4 to 5 m) of perfusion-fixed rat kidney were incubated with antibodies against ₁- (a) or ₂-AR (c). Corresponding double stains with calbindin antibodies are shown in (b) and (d), respectively. The insert in (a) shows consecutive sections of rat DCT incubated with anti–₁-AR after preadsorption with recombinant ₂- (top) or ₁-AR (bottom; specificity control). Bound IgG was counterstained with CY2- (b and d) or CY3-labeled (a and c) secondary antibodies. Arrows in (a) and (b) highlight the position of ₁-AR in intercalated cells lacking calbindin. Arrowheads in (c) point to the small amount of ₂-AR detected in the membranes of DCT cells. The strong fluorescence in (c) corresponds to the staining of ₂-AR in the proximal convoluted tubule. Abbreviations are as follows: G, glomerulus; IC, intercalated cell.

Full figure and legend (154K)

The collecting duct system was identified by its irregularly shaped lumina. Figure 9a shows a collecting duct in the renal cortex strongly labeled by anti–₁-AR. Figure 9b shows the staining of a consecutive section with antibodies against the basolateral erythroid band 3 exchanger, which is a specific marker of acid-excreting type A intercalated cells (whereas the base-excreting type B intercalated cells lack band 3 protein)^33,34. Upon comparison of Figure 9 a and b, it can be seen that most of the cells constituting the cortical collecting duct (CCD) are positive for ₁-AR (with varying intensity), whereas only a fraction of these cells are also labeled by the anti-band 3 protein antibody. Interestingly, in the preceding paragraph, we described the expression of ₁-AR at the apical pole of intercalated cells in the CNT, where in rat kidney more than 90% represent type A cells³⁵. Thus, it seems likely that the fraction of ₁-AR–positive intercalated cells in the CCD also comprise the acid-excreting type A cells. When viewing the outer Figure 9c and inner medullary collecting ducts (IMCD; Figure 9d), the fraction of cells that express the ₁-AR decreases successively. A more precise analysis of the colocalization of ₁-AR Figure 10a and band 3 protein Figure 10b in consecutive sections of a medullary collecting duct shows that in this subsegment, only the acid-excreting type A intercalated cells are stained by anti–₁-AR. At higher magnification, again it can be seen that the expression of ₁-AR appears to be more pronounced at the apical (luminal surface and subapical endocytotic compartment) than at the basolateral pole of these cells (Figure 10c, IC). By using the anti–₂-AR antibody, only a very small amount of ₂-AR could be detected in the collecting duct system (data not shown). Fluorescent signals were clearly more pronounced in medullary than in cortical segments, with the highest receptor density in the IMCD. Again, not all of the cells present in this segment were labeled by the antibody. However, even in the IMCD, the signal intensities were too low to allow for a further differentiation of the specific cell types stained by anti–₂-AR.

Figure 9.

Localization of ₁-AR in the collecting duct. Sections of perfusion-fixed rat kidney were incubated with antibodies against ₁-AR (a, c, and d) or band 3 protein (b). Representative fluorescent images of cortical (a and b), outer (c), and inner medullary (d) collecting ducts were obtained after counterstaining with CY3-labeled secondary antibodies. The simplified schemes on the right margin indicate the respective position of the different sections. The insert in (d) shows the staining pattern obtained after preadsorption of anti–₁-AR with recombinant ₂- (top) or ₁-AR (bottom; specificity control).

Full figure and legend (120K)

Figure 10.

Localization of ₁-AR within type A intercalated cells of rat collecting duct. At a high magnification of consecutive sections of the medullary collecting duct stained with antibodies against the ₁-AR (a and c) or the band 3 protein (b), it can be seen clearly that only type A intercalated cells (with band 3 protein situated in the basolateral cell membrane) express a significant amount of ₁-AR. The enlargement (c) demonstrates that the large majority of the ₁-AR is clustered in the apical membrane and the subapical endocytotic compartment of the intercalated cell. Abbreviations are: IC, intercalated cell; PC, principal cell.

Full figure and legend (114K)

DISCUSSION

Technical considerations

Since the first reports on a direct innervation of the kidney by sympathetic nerves, evidence has accumulated that -AR might play an important role in the regulation of renal function^1,4. Both the ₁- and the ₂-subtype of these receptors, but not ₃-AR, have been found in the cortex and medullary portions of the kidney^10,12. However, the question of where precisely these receptors are located within the different segments and functional cell units of the nephron has not yet been answered in a definite manner. Here, we used a set of highly sensitive and subtype-specific receptor antibodies to analyze the localization of renal ₁- and ₂-AR by immunohistochemistry¹³. Most of our findings are compatible with those obtained in the past by other techniques, including radioligand binding^23,26, radioligand autoradiography^6,7, and RT-PCR^10,11,12. However, in this context, it should be mentioned that (1) the respective amount of -AR mRNA does not necessarily reflect the amount of intact receptor protein in a particular tissue probe, and that (2) RT-PCR requires microdissection of the tissue, which precludes direct imaging of -AR in situ. Although the sensitivity of the immunohistological approach is generally inferior to that of several other (most notably molecular) techniques, our results clearly indicate that it is much more precise in determining receptor localization Figure 11. Thus, our data might contribute to further elucidation of the role of each -AR subtype in the regulation of renal function.

Figure 11.

Simplified scheme indicating the major locations of ₁- and ₂-AR in rat nephron, as determined by immunohistochemistry. Significant amounts of ₁-AR were found in renin-producing juxtaglomerular granular cells (in the wall of the afferent arteriole), mesangial cells, proximal tubular cells, cortical TAL segments (including the macula densa cells), all portions of the distal tubular nephron segments, and the intercalated cells of the collecting duct system (depicted as dark dots). ₂-ARs were found in the membranes of proximal tubular epithelia and, to a lesser extent, in the distal convoluted tubules and the medullary collecting ducts.

Full figure and legend (85K)

₁-Adrenergic control of renin release, sodium reabsorption, and glomerular filtration at the juxtaglomerular apparatus

In agreement with previous studies^2,7, we found a high amount of -AR in the glomerulus and the surrounding region. The glomerulus and juxtaglomerular apparatus are considered as cooperating structures involved in the regulation of systemic blood pressure. Strikingly, we found a selective expression of ₁-AR in almost all of the effector cells of this regulatory unit, suggesting that this receptor subtype might play an important role in the renal regulation of blood pressure.

Renin secretion and distal NaCl reabsorption

It is known that the -adrenergic system is involved in the stimulation of renin release^4,36. High levels of ₁-AR in the renin-producing juxtaglomerular granular cells suggest that these cells are directly activated by catecholamines. However, some previous findings point to the macula densa as another important site for the modulation of renin secretion^36,37,38. Renin is rapidly released in response to a decreased (luminal) NaCl concentration at the macula densa, which in turn is regulated by sodium reabsorption processes occurring in the premacula densa TAL. Functionally, the macula densa serves as a sensor of luminal NaCl: concentration-dependent NaCl uptake through the Na⁺/K⁺/2Cl^- cotransporter in the apical membrane of the macula densa cells initiates a sequence of events that regulates afferent arteriolar tone by altering the local balance of vasoconstrictors (angiotensin II) and vasodilators (NO), and thereby the glomerular filtration rate (referred to as tubuloglomerular feedback)³⁹. In addition, here we show that the macula densa cells and the premacula densa TAL cells near the macula densa express significant amounts of ₁-AR. This finding agrees with previous data obtained by RT-PCR¹⁰ and the results of a functional study on mouse TAL⁴⁰, demonstrating that in this highly innervated particular nephron segment⁴¹, catecholamines increase the active transport of sodium from the luminal side to the interstitium. Thus, by decreasing the NaCl concentration at the macula densa, the -adrenergic system might also contribute indirectly to the macula densa mechanism of renin release.

Glomerular perfusion and filtration

Vasoactive agents (that is, angiotensin, prostaglandins, and catecholamines) may decrease the effective glomerular filtration area by reducing blood flow to certain capillary loops⁴². In addition, it has been shown that both the glomerular ultrafiltration coefficient and the surface area of glomerular capillaries decrease substantially upon systemic application of isoprenaline^3,42. These effects of isoprenaline have been attributed to a contraction of glomerular mesangial cells in response to -AR–mediated stimulation of local mesangial renin activity and subsequent intrarenal angiotensin II generation^3,43. On the other hand, isoproterenol-mediated increases in intracellular cAMP result in decreased in myosin light-chain phosphorylation and tension development in mesangial cells⁴⁴, indicative of cellular relaxation. Here, we show that these specialized smooth muscle-like cells are richly endowed with ₁-AR, suggesting that this receptor subtype might be critically involved in the contraction/relaxation balance of mesangial cells. Moreover, a high density of ₁-AR in the afferent arterioles and in the smooth muscle cells of the larger renal arteries could be demonstrated. Thus, 1-adrenergic regulation might be envisioned to modulate the glomerular filtration area (via mesangial cells), perfusion of the glomerulus (via afferent arterioles), and perfusion of the kidney as a whole (via larger renal arteries).

-Adrenergic control of NaCl reabsorption in the proximal tubule

Whereas the presence and function of -adrenergic receptors in rat proximal tubules have been demonstrated by various means^45,46,47, the presence and role of -AR in this particular nephron segment have been controversial for a long time^45,48,49,50. This was partially due to considerable interspecies differences, but also to major differences in the techniques and/or methods employed. Microperfusion studies on rabbit proximal tubules provided evidence that, in this species, -receptor agonists stimulate volume flux and induce depolarization in a cAMP-independent manner^49,51. In rat proximal tubules, such effects have been attributed to -adrenergic receptors^52,53,54. However, more recently, the presence of -AR on isolated rat PTE cells could be demonstrated by radioligand binding. Functional studies on the same cells indicated that at least a fraction of these -AR were coupled to an as yet not thoroughly characterized cAMP-dependent cellular signaling pathway⁵⁵. These effects appeared to be mediated via ₁-AR localized on both apical and basolateral cell surfaces⁵⁵. However, signaling via apical ₁-AR required cytoskeletal-dependent endocytosis prior to activation of the basolaterally localized adenylate cyclase, which might explain our immunohistochemical results demonstrating a considerable amount of ₁-AR clustered in the subapical endocytotic compartment of PCT cells Figure 6e. Whereas the specific coupling of ₁-AR in rat proximal tubules still remains to be elucidated, ₂-AR activity has been shown to be transduced by a cAMP-independent cellular signaling pathway that acts on both apical Na⁺/H⁺ exchange and basolateral Na⁺,K⁺-ATPase activity^28,29. However, because most of the functional studies were done on cultured rat PTE cells, the precise localization of the corresponding signaling interface in relationship to the polarity of these cells in vivo remains to be demonstrated. Here, we detected high levels of ₂-AR in situ that were clearly more pronounced in apical rather than basolateral membranes of rat proximal tubular epithelia. These findings agree with recent functional data obtained on cultured rat PTE cells demonstrating that (1) selective stimulation of ₂-AR with metaproterenol increases both apical Na⁺/H⁺ exchange and basolateral Na⁺,K⁺-ATPase activity, and (2) that these effects are most probably mediated through apical rather than basolateral ₂-AR²⁹.

The S1 (and S2) segments of the PCT are known to be less densely innervated⁵⁶. Therefore, it has been suggested that catecholamines gain access to their receptors on the tubular epithelia mainly via the cellular interstitium and/or the circulation^1,52, suggesting a predominance of basolaterally localized -AR. However, our study shows that an important fraction of ₁-AR and the large majority of ₂-AR are located in the apical cell membrane. One possible explanation for this peculiar orientation would be that the receptors are meant to interact with catecholamines gaining access to the tubular fluid. This assumption is supported further by pharmacological studies on rat kidney, which have clearly shown that apical tubular surfaces are exposed to catecholamines (and their methylated derivatives) either through glomerular filtration or through direct secretion of these substances by organic cation transporters from the peritubular capillaries into the tubular luminal fluid^52,53,57. In addition, from functional studies on isolated rat PTE cells, it was concluded that cellular signaling through apical -AR is strictly mediated by the apical -AR themselves rather than by paracellular ligand leak or transcellular ligand transport⁵⁵. Taken together, it seems conceivable that both apical ₁- and ₂-AR are involved in the regulation of the NaCl and fluid balance in the proximal tubular nephron segments.

₁-Adrenergic receptors in the distal segments of the nephron

In agreement with previous data obtained by radioligand binding and autoradiography^6,7,12, we demonstrate here that ₁-ARs are predominant in all distal segments of the rat nephron. Again, the expression of this subtype was more pronounced at the apical surface of the cells forming the DCT. In the CNT, a similar level of ₁-AR was found in the membranes of both cell types that constitute this segment, that is, the principal cells and the intercalated cells, whereas in the initial portions of the CCD the ₁-AR seemed to be expressed preferentially in the intercalated cells. We also identified some ₂-AR in the membranes of DCT cells, but neither in the CNT nor in the initial portions of the CCD was a significant amount of this receptor subtype detected. Previous studies on cell lines derived from the DCT suggest that the -AR of these cells are functionally coupled to the adenylate cyclase and stimulate sodium reabsorption¹². Moreover, direct innervation of the distal tubule has been demonstrated^1,58, suggesting that the -ARs in these segments are stimulated by catecholamines released from synapses. However, in addition to the predicted basolateral localization, our study shows that the majority of these receptors appear to be expressed in the apical membranes of DCT cells, which again leads to speculations about a functional relevance of tubular fluid catecholamines.

-Adrenergic receptors in the collecting duct system

High amounts of ₁-AR were found in the collecting duct system, which is composed of the distal portions of the CCD, and the outer medullary collecting duct (OMCD) and IMCD, respectively. In the medullary collecting duct, we also detected some ₂-AR, but failed to detect this receptor subtype in the CCD. This observation seems to disagree with a previous study demonstrating equal amounts of ₁- and ₂-AR–specific mRNA in microdissected tissue samples of rat CCD¹¹. The discrepancy may be due to contaminations of the isolated CCD fragments used for RT-PCR with small vessels and/or capillaries rich in ₂-AR. Discrepancies between mRNA and protein levels might also arise from differential translation and/or protein turnover of ₁- and ₂-AR in the collecting duct. Finally, it should be noted that similar PCR studies carried out on CCD of hamster⁵⁰ and rat kidney⁹ failed to detect significant levels of ₂-AR specific mRNA.

Until recently, the only well-established catecholamine actions in the collecting duct were derived from functional studies on the rabbit. In these studies, -adrenergic agonists were shown to decrease the transepithelial potential difference⁵⁹, to inhibit K⁺ secretion⁶⁰, and to stimulate exchange and Cl^-/Cl^- self-exchange⁶¹ in a cAMP-dependent manner. In addition, it was suggested that isoprenaline increases excretion into the tubular fluid through stimulation of ₁-AR⁶². All of these processes are thought to occur in type B intercalated cells, which are more abundant in rabbit than in rat CCD^35,63. In the rat, little is known about the -adrenergic control of this particular nephron segment. Here, we show a high expression of the ₁-AR subtype in rat CCD. The receptors are predominantly localized on intercalated (rather than principal) CCD cells, including both the band 3-positive type A and very probably also the band 3-negative type B intercalated cells. Thus, in conjunction with functional data derived from the rabbit, our immunohistochemical results indicate that in rat CCD, the -adrenergic system might be involved in the regulation of both the acid-excreting type A and the base-excreting type B intercalated cells.

₁-Adrenergic control of acid excretion in the medullary collecting duct

The OMCD has been shown to be an important site for urine acidification, whereas the IMCD is known to represent the site of final adjustment of the urine composition. The numbers of intercalated cells in these segments are less than in the CCD, but they are all acid-excreting type A cells (identified by H⁺-ATPase in the apical, and band 3 protein in the basolateral membrane)³³. In the rabbit, there is increasing evidence that -ARs are critically involved in regulating OMCD function^64,65 and particularly in regulating the secretion of H⁺ into the tubular fluid. Recently, those studies were extended by Manger, Pappas and Koeppen⁶⁶, who demonstrated that in cultured rabbit type A intercalated cells isoprenaline activates both a cAMP-dependent (increase in basolateral exchange) and a cAMP-independent cellular signaling pathway (increase in apical H⁺-ATPase activity), which resulted in increased luminal H⁺ secretion. Our study revealed that in rats the ₁-AR subtype was clearly restricted to the acid-excreting type A intercalated cells, suggesting that -adrenergic stimulation of acid secretion in the collecting duct is mediated through this receptor subtype. Besides the ₁-AR subtype, a faint amount of ₂-AR in the medullary portions of the collecting duct was detected, with an apparent higher expression of this receptor subtype in the IMCD. Corresponding data were obtained by Yasuda et al, who demonstrated expression of both functional ₁-AR and, to a lesser extent, ₂-AR in the initial and terminal portions of rat IMCD⁶⁷.

Pharmacological considerations

One striking feature in this study is that an important fraction of the -AR expressed in the tubular cells of the nephron and in the intercalated cells of the collecting duct system is located apically (in addition to the predicted basolateral location) and thus is likely to be exposed to the tubular luminal fluid. Implications of this peculiar orientation for the delivery of the endogenous ligands have already been discussed. However, one should also consider implications for the systemic administration of synthetic -AR agonists and antagonists. Such considerations apply particularly to ₁-selective antagonists of the first and second generations, such as metoprolol and bisoprolol, which are standard agents in the treatment of hypertension, coronary heart disease, and heart failure^68,69. Although a majority of these water-soluble drugs are metabolized to inactive metabolites, up to 50% of an oral dose of bisoprolol (and about 3% in the case of metoprolol) is eliminated via the kidneys as an unchanged substance²⁴. It is likely that such agents will exert pharmacological effects during their passage through the nephron. For instance, they could block ₁-AR at the macula densa, in distal tubular epithelia, or in type A intercalated cells of the collecting duct and thus inhibit renin secretion, NaCl reabsorption, and also urine acidification. Such pharmacological effects have, to our knowledge, not been taken into consideration when using these compounds in clinical practice because the expression of -AR at the luminal surfaces of the nephron has not been known. In light of a recent report on the effects of metoprolol on renal NaCl handling in an experimental rat model of chronic heart failure⁷⁰ and the data presented here, it is tempting to speculate how the pharmacological potential of -receptor antagonists in tubular luminal fluid could be exploited further in therapy.

References

1. DiBona, GF & Kopp, UC: Neural control of renal function. Physiol Rev 1997 77: 75–197,  | PubMed | ISI | ChemPort |
2. Engel, G, Maurer, R & Perrot, K, et al: -Adrenoceptor subtypes in sections of rat and guinea-pig kidney. Naunyn Schmiedebergs Arch Pharmacol 1985 328: 354–357,  | Article | PubMed | ISI | ChemPort |
3. Pelayo, J, Tucker, B & Blantz, R: Effects of -adrenergic stimulation with isoproterenol on glomerular hemodynamics. Am J Physiol 1989 257: F866–F873,  | PubMed | ISI | ChemPort |
4. Torretti, J: Sympathetic control of renin release. Ann Rev Pharmacol Toxicol 1982 22: 167–192,  | ISI | ChemPort |
5. Malbon, CC, Moxham, CP & Brandwein, HJ: Antibodies to -adrenergic receptors, in The -Adrenergic Receptors, 1991, edited by Perkins JP, Baltimore, Humana Press, pp 181–261
6. Healy, D, Münzel, P & Insel, P: Localization of ₁- and ₂-adrenergic receptors in rat kidney by autoradiography. Circ Res 1985 57: 278–284,  | PubMed | ISI | ChemPort |
7. Summers, R, Stephenson, J & Kuhar, M: Localization of -adrenoceptor subtypes in rat kidney by light microscopic autoradiography. J Pharmacol Exp Ther 1985 232: 561–569,  | PubMed | ISI | ChemPort |
8. Amenta, F, Cavallotti, C & De Rossi, M, et al: -Adrenoceptors in the rat kidney. Immunohistochemical study. Naunyn Schmiedebergs Arch Pharmacol 1983 324: 94–98,  | Article | PubMed | ISI | ChemPort |
9. Meister, B, Dagerlind, A & Nicholas, A, et al: Patterns of messenger RNA expression for adrenergic receptor subtypes in the rat kidney. J Pharmacol Exp Ther 1994 268: 1605–1611,  | PubMed | ISI | ChemPort |
10. Elalouf, J, Buhler, J & Tessiot, C, et al: Predominant expression of ₁-adrenergic receptor in the thick ascending limb of rat kidney: Absolute mRNA quantitation by reverse transcription and polymerase chain reaction. J Clin Invest 1993 91: 264–272,  | PubMed | ISI | ChemPort |
11. Mandon, B, Siga, E & Champigneulle, A, et al: Molecular analysis of -adrenergic receptor subtypes in rat collecting duct: Effects on cell cAMP and Ca²⁺ levels. Am J Physiol 1995 268: F1070–F1080,  | PubMed | ISI | ChemPort |
12. Gesek, F & White, K: Molecular and functional identification of -adrenergic receptors in distal convoluted tubule cells. Am J Physiol 1997 272: F712–F720,  | PubMed | ISI | ChemPort |
13. Jahns, R, Siegmund, C & Jahns, V, et al: Probing human ₁- and ₂-adrenoceptors with domain-specific fusion protein antibodies. Eur J Pharmacol 1996 316: 111–121, 10.1016/s0014-2999(96)00654-1 | Article | PubMed | ISI | ChemPort |
14. Machida, CA, Bunzow, JR & Searles, RP, et al: Molecular cloning and expression of the rat ₁-adrenergic receptor gene. J Biol Chem 1990 265: 12960–12965,  | PubMed | ISI | ChemPort |
15. Gocayne, J, Robinson, DA & Fitz, GM, et al: Primary structure of rat cardiac -adrenergic and muscarinic cholinergic receptors obtained by automated DNA sequence analysis: Further evidence for a multigene family. Proc Natl Acad Sci USA 1987 84: 8296–8300,  | PubMed | ChemPort |
16. Frielle, T, Collins, S & Daniel, KW, et al: Cloning of the cDNA for the human ₁-adrenergic receptor. Proc Natl Acad Sci USA 1987 84: 7920–7924,  | PubMed | ChemPort |
17. Kobilka, BK, Dixon, RA & Frielle, T, et al: cDNA for the human ₂-adrenergic receptor: A protein with multiple membrane-spanning domains and encoded by a gene whose chromosomal location is shared with that of the receptor for platelet-derived growth factor. Proc Natl Acad Sci USA 1987 84: 46–50,  | PubMed | ChemPort |
18. Reiländer, H, Boege, F & Vasudevan, S, et al: Purification and functional characterization of the human ₂-adrenergic receptor produced in baculovirus-infected insect cells. FEBS Lett 1991 282: 441–444,  | PubMed |
19. Jahns, R, Boivin, V & Siegmund, C, et al: Autoantibodies activating human ₁-adrenergic receptors are associated with reduced cardiac function in chronic heart failure. Circulation 1999 99: 649–654,  | PubMed | ISI | ChemPort |
20. Drenckhahn, D, Schlüter, K & Allen, DP, et al: Colocalization of band 3 with ankyrin and spectrin at the basal membrane of intercalated cells in the rat kidney. Science 1985 230: 1287–1289,  | PubMed | ISI | ChemPort |
21. Munson, PJ & Rodbard, D: LIGAND: A versatile computerized approach to characterization of ligand binding systems. Anal Biochem 1980 107: 220–239,  | Article | PubMed | ISI | ChemPort |
22. Kriz, W & Kaissling, B: Structural organisation of the mammalian kidney, in The Kidney: Physiology and Pathophysiology, 1992, edited by Seldin DW, Giebisch G, New York, Raven Press, pp 707–777
23. Snavely, MD, Motulsky, HJ & Moustafa, E, et al: -Adrenergic receptor subtypes in the rat renal cortex. Circ Res 1982 51: 504–513,  | PubMed | ISI | ChemPort |
24. Leopold, G & Kutz, K: Bisoprolol: Pharmacokinetic profile. Rev Contemp Pharmacother 1997 8: 35–43,  | ChemPort |
25. Brodde, O-E: The pharmacology of bisoprolol. Rev Contemp Pharmacother 1997 8: 21–23,  | ChemPort |
26. Michel, MC, Wang, XL & Schlicker, E, et al: Increased ₂-adrenoceptor density in heart, kidney, and lung of spontaneously hypertensive rats. J Auton Pharmacol 1987 7: 41–51,  | PubMed | ISI | ChemPort |
27. Boege, F, Jürss, R & Cooney, D, et al: Functional and structural characterization of the two ₁-adrenoceptor forms in turkey erythrocytes with molecular masses of 50 and 40 kilodaltons. Biochemistry 1987 26: 2418–2425,  | Article | PubMed | ISI | ChemPort |
28. Singh, H & Linas, S: ₂-Adrenergic function in cultured rat proximal tubule epithelial cells. Am J Physiol 1996 271(1 Pt 2):F71–F77,  | PubMed | ISI | ChemPort |
29. Singh, H & Linas, S: Role of protein kinase C in ₂-adrenoceptor function in cultured rat proximal tubule epithelial cells. Am J Physiol 1997 273: F193–F199,  | PubMed | ISI | ChemPort |
30. Hoyer, JR, Sisson, SP & Vernier, RL: Tamm-Horsfall glycoprotein: Ultrastructural immunoperoxidase localization in the rat kidney. Lab Invest 1979 41: 168–173,  | PubMed | ISI | ChemPort |
31. Taylor, AN, McIntosh, JE & Bourdeau, JE: Immunocytochemical localization of vitamin D-dependent calcium-binding protein in renal tubules of rabbit, rat, and chick. Kidney Int 1982 21: 765–773,  | PubMed | ISI | ChemPort |
32. Kaissling, B & Kriz, W: Morphology of the loop of Henle, distal tubule, and collecting duct, in Handbook of Physiology: Renal Physiology, 1992, edited by Windhager EE, New York, Oxford University Press, pp 109–167
33. Clapp, WL, Madsen, KM & Verlander, JW, et al: Intercalated cells of the rat inner medullary collecting duct. Kidney Int 1987 31: 1080–1087,  | PubMed | ISI | ChemPort |
34. Alper, SL, Natale, J & Gluck, S, et al: Subtypes of intercalated cells in the rat kidney collecting duct defined by antibodies against erythroid band 3 and renal vacuolar H⁺-ATPase. Proc Natl Acad Sci USA 1989 86: 5429–5433,  | PubMed | ChemPort |
35. Verlander, JW, Madsen, KM & Tisher, CC: Effect of acute respiratory acidosis on two populations of intercalated cells in rat cortical collecting duct. Am J Physiol 1987 253: F1142–F1456,  | PubMed | ISI | ChemPort |
36. Osborn, JL, Thames, MD & DiBona, GF: Role of macula densa in renal nerve modulation of renin secretion. Am J Physiol 1982 242: R367–R371,  | PubMed | ISI | ChemPort |
37. Skott, O & Briggs, JP: Direct demonstration of macula densa-mediated renin secretion. Science 1987 237: 1618–1620,  | PubMed | ISI | ChemPort |
38. Schnermann, J & Briggs, JP: The macula densa is worth its salt. J Clin Invest 1999 104: 1007–1009,  | PubMed | ISI | ChemPort |
39. Wright, FS & Schnermann, J: Interference with feedback control of glomerular filtration rate by furosemide, triflocin and cyanide. J Clin Invest 1974 53: 1695–1708,  | PubMed | ISI | ChemPort |
40. Bailly, C, Imbert-Teboul, M & Roinel, N, et al: Isoproterenol increases Ca, Mg, and NaCl reabsorption in mouse thick ascending limb. Am J Physiol 1990 258: F1224–F1231,  | PubMed | ISI | ChemPort |
41. Barajas, L & Powers, K: Innervation of the thick ascending limb of Henle. Am J Physiol 1988 255: F340–F348,  | PubMed | ISI | ChemPort |
42. Schor, N, Ichikawa, I & Brenner, BM: Mechanisms of action of various hormones and vasoactive substances on glomerular ultrafiltration in the rat. Kidney Int 1981 20: 442–451,  | PubMed | ISI | ChemPort |
43. Dzau, VJ & Kreisberg, J: Cultured glomerular mesangial cells contain renin: Influence of calcium and isoproterenol. J Cardiovasc Pharmacol 1986 8(Suppl 10):S6–S10,  | PubMed | ISI | ChemPort |
44. Kreisberg, JI, Venkatachalam, MA & Radnik, RA, et al: Role of myosin light-chain phosphorylation and microtubules in stress fiber morphology in cultured mesangial cells. Am J Physiol 1985 249: F227–F235,  | PubMed | ISI | ChemPort |
45. Sundaresan, P, Fortin, T & Kelvie, S: - and -adrenergic receptors in proximal tubules of rat kidney. Am J Physiol 1987 253: F848–F856,  | PubMed | ISI | ChemPort |
46. Clark, JD, Cragoe, EJ & Limbird, LE: ₂-adrenergic receptors regulate Na⁺-H⁺ exchange via cAMP-dependent mechanism. Am J Physiol 1990 259: F977–F985,  | PubMed | ISI | ChemPort |
47. Feng, F, Pettinger, WA & Abel, PW, et al: Regional distribution of ₁-adrenoceptor subtypes in rat kidney. J Pharmacol Exp Ther 1994 258: 263–268,
48. Jacobs, W & Chan, Y: Evidence for the presence of functional -adrenoceptor along the proximal tubule of the rat kidney. Biochem Biophys Res Commun 1986 141: 334–339,  | Article | PubMed | ISI | ChemPort |
49. Kudo, K, Kondo, Y & Abe, K, et al: Evidence for presence of functional -adrenoceptor in rabbit S2 proximal straight tubules. Am J Physiol 1991 261: F393–F399,  | PubMed | ISI | ChemPort |
50. Taniguchi, S, Watanabe, T & Nakao, A, et al: Distribution of ₂-adrenergic receptor mRNA expression along the hamster nephron segments. FEBS Lett 1993 318: 65–70,  | Article | PubMed | ISI | ChemPort |
51. Bello-Reuss, E: Effect of catecholamines on fluid reabsorption by the isolated proximal convoluted tubule. Am J Physiol 1980 238: F347–F352,  | PubMed | ChemPort |
52. Chan, YL: The role of norepinephrine in the regulation of fluid absorption in the rat proximal tubule. J Pharmacol Exp Ther 1980 215: 65–70,  | PubMed | ISI | ChemPort |
53. Weinman, EJ, Sansom, SC & Knight, TF, et al: and -adrenergic agonists stimulate water absorption in the rat proximal tubule. J Membr Biol 1982 69: 107–111,  | PubMed | ISI | ChemPort |
54. Gesek, FA & Schoolwerth, AC: Hormonal interaction with the proximal Na⁺-H⁺ exchanger. Am J Physiol 1990 258: F514–F521,  | PubMed | ISI | ChemPort |
55. Hanson, A & Linas, S: -Adrenergic receptor function in rat proximal tubule epithelial cells in culture. Am J Physiol 1995 268: F553–F560,  | PubMed | ISI | ChemPort |
56. Barajas, L & Powers, K: Innervation of the renal proximal convoluted tubule of the rat. Am J Anat 1989 186: 378–388,  | Article | PubMed | ISI | ChemPort |
57. Baines, AD, Craan, A & Chan, W, et al: Tubular secretion and metabolism of dopamine, norepinephrine, methoxytyramine and normetanephrine by the rat kidney. J Phamacol Exp Ther 1979 208: 144–147,  | ChemPort |
58. Barajas, L, Powers, K & Wang, P: Innervation of the renal cortical tubules: A quantitative study. Am J Physiol 1984 247: F50–F60,  | PubMed | ISI | ChemPort |
59. Iino, Y, Troy, JL & Brenner, BM: Effects of catecholamines on electrolyte transport in cortical collecting tubules. J Membr Biol 1981 61: 67–73,  | PubMed | ISI | ChemPort |
60. Kimmel, PK & Goldfard, S: Effects of isoproterenol on potassium secretion by the cortical collecting tubule. Am J Physiol 1984 246: F804–F810,  | PubMed | ISI | ChemPort |
61. Tago, K, Schuster, VL & Stokes, JB: Regulation of chloride self exchange by cAMP in cortical collecting tubule. Am J Physiol 1986 251: F40–F48,  | PubMed | ISI | ChemPort |
62. Hayashi, M, Yamaji, Y & Iyori, M, et al: Effect of isoproterenol on intracellular pH of the intercalated cells in the rabbit cortical collecting ducts. J Clin Invest 1991 87: 1153–1157,  | PubMed | ISI | ChemPort |
63. Schuster, VL, Bonsib, SM & Jennings, ML: Two types of collecting duct mitochondria-rich (intercalated) cells: Lectin and band 3 cytochemistry. Am J Physiol 1986 251: C347–C355,  | PubMed | ISI | ChemPort |
64. Burnatowska-Hledin, MA & Spielman, WS: Immunodissection of mitochondria-rich cells from rabbit outer medullary collecting tubule. Am J Physiol 1988 254: F907–F911,  | PubMed | ChemPort |
65. Hays, S, Kokko, JP & Jacobson, R: Hormonal regulation of proton secretion in rabbit medullary collecting duct. J Clin Invest 1986 78: 1279–1286,  | PubMed | ISI | ChemPort |
66. Manger, T, Pappas, C & Koeppen, B: -Adrenergic regulation of H⁺ secretion by cultured outer medullary collecting duct cells. Am J Physiol 1992 263: F1011–F1019,  | PubMed | ISI | ChemPort |
67. Yasuda, G, Sun, L & Lee, H, et al: The ₁- and ₂-adrenoceptor subtypes in cultured rat inner medullary collecting duct cells. Am J Physiol 1996 271: F762–F769,  | PubMed | ISI | ChemPort |
68. MERIT-HF Study Group: Effect of metroprolol CR/XL in chronic heart failure: Metroprolol CR/XL randomised intervention trial in congestive heart failure (MERIT-HF). Lancet 1999 353: 2001–2007, 10.1016/s0140-6736(99)04440-2
69. Cibis-Ii Investigators and Committees: The cardiac insufficiency bisoprolol study II (CIBIS-II): A randomised trial. Lancet 1999 353: 9–13, 10.1016/s0140-6736(98)11181-9
70. DiBona, GF & Sawin, LL: Effect of metoprolol administration on renal sodium handling in experimental congestive heart failure. Circulation 1999 100: 82–86,  | PubMed | ISI | ChemPort |
71. Kriz, W, Elger, M & Lemley, K, et al: The structure of the glomerular mesangium: A biomechanical interpretation. Kidney Int 1990 38(Suppl 30):S2–S9,  | ISI | ChemPort |

Acknowledgments

The work was supported by the Deutsche Forschungsgemeinschaft (DFG Ja 706/2-1; Bonn, Germany). Valérie Boivin was recipient of a grant from the Alexander-von-Humboldt Foundation (Bonn, Germany). A portion of this study was presented at the European Conference on Management of Coronary Heart Disease, Nice, France, April, 2000.