Background: Natriuretic peptides regulate Na+ and H2O transport in the cortical collecting duct (CCD). We have shown that natriuretic peptides have no effect on ion conductances or water transport of principal cells (PC) even though a cGMP-regulated K+ channel is located in the basolateral membrane of these cells. Methods: RT-PCR was used to screen for different guanylyl cyclases (GC) in CCD and to look for the expression of GC-1 and GC-A mRNA in CCD of male and female Wistar and Sprague–Dawley rats. Polyclonal antibodies were raised against the detected GC. BCECF was used to investigate the effects of ANP on intracellular pH in intercalated cells (IC). Results: GC-A and GC-1 were detected. GC-A was immunolocalized in the luminal membrane of IC while GC-1 was mainly found in the luminal membrane of PC. GC-1 is expressed in Sprague–Dawley and Wistar rats except for male Sprague–Dawley rats, while GC-A is expressed in all strains. ANP (160 nM, n=11), urodilatin (140 nM, n=6), which had no effect in PC, significantly decreased pHi by 0.02±0.01 and 0.03±0.01 Units in IC, respectively. ANP as well as urodilatin and 8-Br-cGMP decreased the pHi recovery after acidification by 30±6% (n=12), 37±7% (n=8), and 19±3% (n=8), respectively. Conclusion: GC-A is located in the luminal membrane of IC of rat CCD and ANP acts through this receptor when regulating pHi via an inhibition of the Na+/H+-exchanger. PC do not possess GC-A. GC-1 seems to be the only GC in these cells of most rat strains tested and therefore, it could be responsible for the regulation of K+ channels in the basolateral membrane via cGMP-dependent protein kinase.
Time for primary review 28 days.
Natriuretic peptides are structurally related peptides which share common features, such as tissue distribution, biosynthetic pathways, and pharmacologic effects in target organs. The atrial natriuretic peptide (ANP) and the brain natriuretic peptide (BNP) are mainly produced in the heart and in specialized areas of the nervous system while the C-type natriuretic peptide (CNP) is mainly found in peripheral blood vessels . A newer report demonstrated that the highly homologous form of ANP, urodilatin, and BNP and CNP are also generated in the kidney , making these circulating hormones act systemic, paracrine, and possibly autocrine. Natriuretic peptides control vascular tone and renal salt excretion, and therefore, modulate plasma volume and blood pressure. For ANP it has been shown to increase GFR by vasodilatation of afferent arterioles, reduce Na+ reabsorption from the proximal tubule by inhibiting sympathetic tone and action of catecholamines, and from the late distal tubules and collecting ducts by inhibiting the Na+ reabsorption and secretion of aldosterone [1,3]. Three natriuretic peptide receptors have been identified so far, GC-A, GC-B, and the clearance receptor (C-receptor) [4,5]. While GC-A and GC-B contain an intracellular guanylyl cyclase domain that is activated upon binding of these peptides, the C-receptor only has a short cytoplasmic domain which is not coupled to a guanylyl cyclase . Besides the soluble guanylyl cyclases (GC-S) several receptor-coupled guanylyl cyclases have been cloned so far. For some of them the agonist is known. GC-A binds ANP, BNP, and urodilatin, GC-B binds CNP. GC-C binds the heat-stable enterotoxin (STa), guanylin, and uroguanylin [4,5]. For GC-D, GC-E, GC-F, and GC-G, the agonists are still unknown [6–8]. GC-1, ksGC and GC-G are thought to be identical receptors [8–10].
One of the major renal tubular target sites for natriuretic peptides is the collecting duct of the mammalian kidney. For the inner medullary collecting duct it has been shown that ANP reduces Na+[11–13] and vasopressin-dependent water and urea reabsorption [14–16]. The effects of ANP on the cortical portion are less clear and quite contradictory. In two different studies it was shown that ANP reduces NaCl reabsorption in isolated perfused cortical collecting ducts (CCD) of rat  and in the M1 cell line of mouse , but three subsequent studies failed to confirm these results using cortical collecting ducts of rat [19–21]. Recently, we tested several cGMP-activating agonists, including ANP, BNP, CNP, urodilatin, guanylin, uroguanylin, to investigate their possible involvement in the regulation of basolateral K+ channels in rat CCD. These K+ channels had been shown to be regulated by a cGMP-dependent protein kinase . The tested cGMP-activating agonists were neither involved in the regulation of these K+ channels nor did they have any influence on electrogenic electrolyte or water transport in principal cells of rat CCD . To clarify these contradictory results we present a combined functional, molecular and histochemical approach to identify the distribution of receptor-coupled guanylyl cyclases in collecting ducts of rat using BCECF (2′,7′-bis(2-carboxyethyl)-5(6)-carboxyfluorescein) microfluorimetry, the RT-PCR method and immunocytochemistry.
2.1 Isolation of cortical collecting ducts and cortical collecting duct cell clusters
Cortical collecting ducts and CCD cell clusters were freshly isolated from kidneys of female Wistar rats (Charles River Wiga, Sulzfeld, Germany) with a body weight of 100–200 g for the intracellular pH measurements. For RT-PCR experiments CCDs of male and female Wistar as well as Sprague–Dawley rats were used. The enzymatic isolation procedure was as described before in detail [23,24]. Isolated CCD clusters of 10–15 cells were fixed with a glass pipette in the thermostable and constantly perfused experimental chamber. The chamber was mounted on an inverted microscope (Axiovert 135, Zeiss, Jena, Germany) equipped with a 100× oil-immersion lens. The standard bath solution contained (in mM): NaCl 145, K2HPO4 1.6, KH2PO4 0.4, calcium gluconate 1.3, MgCl2 1, and d-glucose 5; pH was adjusted to 7.4. All experiments were performed in a running bath (0.5 ml volume) with an exchange rate of 20 times per minute at 37°C. All standard chemicals used were obtained in the highest available purity from Sigma (Deisenhofen, Germany) and Merck (Darmstadt, Germany).
2.2 Cloning of guanylyl cyclases in rat cortical collecting duct
The overall strategy of cloning guanylyl cyclase isoforms was to perform RT-PCR with degenerate primers positioned in conserved regions. A comparison of all rat GC-sequences cloned to date showed two highly conserved regions in the guanylyl cyclase catalytic domain in positions 912–928 and 984–1002 (positions refer to GC-A; amino acids accession number P18910; EMBL data base). In order to repress non-specific amplifications GC-specific cDNA was synthesized by reverse transcription using a primer positioned in the most conserved region of the 3′ end.
Total RNA was isolated from microdissected rat CCD using RNeasy-Kit (Qiagen, Hilden, Germany) and cDNA was synthesized using JH1 as primer (see below). For amplification, cDNA was mixed with PCR buffer containing 200 μM of each dNTP and 0.4 μM of each primer (JH2 together with JH4). The amplification was performed in a Perkin-Elmer Cetus thermocycler 9600 (Perkin-Elmer, Stuttgart, Germany) using the following parameters: 5 min of denaturation at 96°C; addition of 1 U Taq DNA-polymerase at 50°C (hot-start); 90 s of primer extension at 72°C. Then, 34 cycles of amplification were run with the following parameters: 20 s at 96°C; 10 s at 50°C; and 90 s at 72°C. After separation of the samples in a 1.5% ethidium bromide-stained agarose gel, PCR products with the expected size of 243 bp were isolated and purified using QIAquick gel extraction kit (Qiagen). The purified DNA was then cloned into a pGEM®-T vector (Promega, Heidelberg, Germany) and sequenced using a dye terminator sequencing kit (Perkin-Elmer). One of the sequences obtained matched the partial sequence of a guanylyl cyclase published as GC-1 . The full sequence was cloned later and renamed to GC-G . Therefore, we also cloned the 3′ terminus of GC-1 by rapid amplification of cDNA ends (3′ RACE). A specific GC-1 primer (GC1-1f) positioned within the cloned fragment was constructed and used for amplification together with a 3′-anchor primer (unip6) hybridizing to the oligo(dT)-anchor sequence (unip5) added through poly-(A)+-dependent reverse transcription. A single product was achieved which was cloned and sequenced as described above. Finally, the expression of GC-1 mRNA in rat CCD was verified using the highly specific primers GC1-1f and GC1-1r.
Sequences of the primers used in 5′ to 3′ orientation:
2.3 Detection of GC-A and GC-1 in cortical collecting ducts using RT-PCR
Cortical collecting duct segments (300–400) were collected from male and female Wistar and Sprague–Dawley rats and lysed in a 4-M guanidinium chloride buffer. Total RNA was isolated using the RNeasy-kit (Qiagen). Isolated total RNA was incubated with 10 U DNase I (Promega) at 37°C for 1 h to digest isolated traces of genomic DNA. RNA and DNase I were then separated by an additional clean-up step using a new RNeasy column. cDNA first strand synthesis was performed in a total reaction volume of 30 μl containing 5 μg total RNA, 10 nM dNTP-Mix (Biometra, Göttingen, Germany), 1 nM p(dT)10 nucleotide primer (Boehringer Mannheim, Germany) and 200 U Moloney murine leukemia virus reverse transcriptase (MMLV-RT; Promega). Then 1/30 of each cDNA first strand reaction mixture was subjected to a 50-μl PCR reaction in a UNO II thermo cycler (Biometra) using 20 pmol of each primer and 1 Unit of Taq DNA polymerase (Qiagen). Reaction conditions were as follows: 3 min at 94°C, 30 s at 59°C and 1 min at 72°C, 1 cycle; 30 s at 94°C, 30 s at 59°C and 1 min at 72°C, 30 cycles; 30 s at 94°C, 30 s at 59°C and 10 min at 72°C, 1 cycle. PCR reaction products were analyzed by agarose gel electrophoresis.
All positive signals obtained from PCR experiments were sequenced by GATC (Konstanz, Germany).
The following PCR primers for GC-A and ksGC were used (listed in 5′ to 3′ direction). The sequence is followed by the expected fragment length for the respective sense and antisense primer:
GC-A sense:AAA AAT TGT GGA CGG CAC CTG AG 888 bp
GC-A antisense:AGG CAG TAT CGG GCC ATC TTT AG
GC-1 sense:GTC CTC TGG CTC TGT TTC CTA 447 bp
GC-1 antisense:AAT CCT TTC GTC ACC TTT ATC CT
2.4 Peptide synthesis and antibodies
Receptor segments of GC-A and GC-1 (ksGC, GC-G) were synthesized as multiple antigenic peptides (MAPs) on a manually prepared branched Fmoc8Lys7Ala-Wang polystyrene resin. Standard Fmoc chemistry  on a SMPS 350 automated multiple peptide synthesizer (Zinsser Analytics, Frankfurt, Germany) equipped with a 48-hole synthesis block (Multisyntech, Bochum, Germany) was used. Acylation of excess Fmoc amino acids (10 equivalents) was performed in the presence of 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU), diisopropyl-ethylamine and 1-hydroxybenzotriazole for 30 min . Fmoc groups were cleaved with 50% piperidine in N-methylpyrrolidinone. The assembled peptides were deprotected and cleaved from the resin with a mixture of trifluoroacetic acid/ethanedithiol/water (94:3:3 v/v) in 2 h. Crude peptides were precipitated with tert-butylmethylether, lyophilized from 5% acetic acid and purified by standard preparative reversed-phase HPLC on a Vydac C18 column. Homogeneity and identity of the purified peptides was checked by HPLC and amino acid analysis.
Polyclonal antisera were raised in New Zealand rabbits. Synthesized receptor segments of GC-A and GC-1 were coupled to hemocyanin/thyreoglobulin as carrier proteins. For the first booster injection, the receptor–hemocyanin/thyreoglobulin complex (500 μg of the coupled peptide in NaCl) was then emulsified with complete Freund's adjuvant and injected subcutaneously into the back of the rabbits. With time intervals of 4 weeks, subsequent booster injections were carried out with incomplete Freund's adjuvant (250 μg of the coupled peptide). Two weeks after each booster injection blood was drawn .
Respective peptide sequences were taken from extracellular (GC-A) and intracellular domains (GC-1) of each receptor. The sequence information from position 421–433 of the GC-1 receptor was taken from Swiss-Prot (locus KSGC_RAT, accession P55205, 1996). W at position 434 was deduced from cloned cDNA out of the cortical collecting duct of female Wistar rats.
GC-1 (rat):EFTEEEAKVPEILW (421–434)
2.5 Specificity of the GC-1 and GC-A antibody
Female Wistar rat kidney membrane fractions were prepared as described . SDS-Page and immunoblotting was done according to published procedures . In brief, 20 μg membrane protein was separated on a 10% gel and blotted semidry onto PVDF membranes (Polyscreen, NEN-DuPont, Bad Homburg, Germany) with 1 mA/cm2 for 75 min. The membranes were washed twice with tris-buffered saline containing 0.1% Tween 20 pH 7.6 (TBS-T), blocked for 1 h at room temperature with Rotiblock™ (Carl Roth, Karlsruhe, Germany) and incubated with polyclonal rabbit GC-1 antiserum K252 (1:1000) or polyclonal rabbit GC-A antiserum A245 (1:1000) overnight at 4°C in Rotiblock™. The next day the membranes were washed in TBS-T (3 times for 5 min each) and incubated for 1 h with anti-rabbit-IgG–HRP conjugate (1:100 000, New England Biolabs, Beverly, MA, USA). Then the membranes were washed again (five times for 5 min each), incubated for 1 min in WestDura chemiluminescence solution and exposed for various times to CLXposure chemiluminescence film (both from Pierce, Rockford, IL, USA). The signals could be blocked by preincubation with 1 μg/ml antigenic peptide.
Immunocytochemical labeling was performed on sections of rat kidney that were fixed by perfusion with 4% paraformaldehyde and embedded in paraffin. The sections were labeled as follows: (1) free aldehyde sites were blocked with 50 mM glycine in phosphate buffered saline (PBS) for 15 min, (2) non-specific binding sites were blocked with 5% goat serum diluted in PBG (0.2% gelatin and 0.5% bovine serum albumin in PBS) for 30 min, (3) incubation with polyclonal antibodies against GC-A and GC-1 (diluted 1:4000) in PBG for 2 h, or AQP-3 (diluted 1:2000; kindly provided by Dr Jean-Marc Verbavatz) , (4) after several rinsing steps incubation with Cy3 coupled goat anti rabbit IgG (DIANOVA, Hamburg, Germany, diluted 1:1000) in PBG for 1 h to detect binding sites of the primary antibodies.
For control experiments the preimmune sera of the rabbits providing the primary polyclonal antibodies were used. For fluorescence microscopy specially designed filter systems (ATF, Tübingen, Germany) were employed that utilized the exact absorption and emission maxima of the Cy3 and FITC dye.
2.7 Measurements of intracellular pH (pHi)
The membrane permeable BCECF analog 2′-7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein-acetoxymethyl ester (BCECF-AM, Molecular Probes Europe BV, Leiden, The Netherlands) was dissolved in standard solution containing 0.1 g/l pluronic F-127 (Calbiochem, Bad Soden, Germany). The CCD cells were loaded with BCECF-AM (1 μM) by incubating them with this solution at room temperature in the dark for 45 min. Loaded cells were excited by passing light through 488 and 436 nm filters positioned in a filter wheel (Physiologisches Institut, Universität Freiburg, Germany) rotating at 10 Hz using a xenon-quartz lamp (XBO 75 W; Zeiss) as light source. Fluorescence at 520–560 nm was detected with a single photon counting tube (H3460-04; Hamamatsu, Herrsching, Germany), and ten consecutive data points were averaged, yielding a time resolution of 1 Hz. The autofluorescence and the system noise were measured before each experiment and subtracted from the original BCECF fluorescence signals. An iris diaphragm allowed the reduction of the area of measurement to approximately 10–15 cells of the isolated CCD clusters. The fluorescence ratio (488:436 nm) was recorded after a 15-min equilibration period in the continuously perfused bath. Experiments were controlled and data were analyzed with an AT-486 computer system and custom-designed software (U. Fröbe, Universität Freiburg, Germany).
The calibration of the BCECF fluorescence signal was done in separate experiments with the protonophore carbonyl cyanide m-chlorophenyl-hydrazone (CCCP; 1 μM) and external pH values between 6.5 and 8.0 as described in detail before . In this range, the dependence of the fluorescence ratio on the external pH was linear. All experiments were done in the absence of HCO3− to exclude HCO3−-dependent transport systems regulating cellular pH.
pHi data are presented as fluorescence ratios (488/436) in the figure representing an original recording or as mean values±S.E.M. (n representing the number of monolayers). A paired Student's t-test was used to test for statistical significance with each effect compared to its own control. P<0.05 was set as the significance level.
The investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996).
Guanylyl cyclase (GC)-specific cDNA from microdissected cortical collecting ducts (CCD) was amplified by PCR using degenerate primers positioned at two highly conserved regions in the GC-catalytic domain. As a result of this strategy we found two different receptor-coupled guanylyl cyclases, namely GC-A and GC-1 (Fig. 1). GC-A binds ANP, BNP, and urodilatin . The full receptor of which GC-1 builds most of the intracellular domain [9,10] has recently been fully cloned and was named GC-G . In this latter report, an American group claimed that GC-1 is not detectable in the kidney using the northern blot technique while the first two reports identified mRNA of this novel guanylyl cyclase in kidney using RT-PCR. Since one of the two Japanese groups  used female Wistar rats (personal communication), while the American group used male Sprague–Dawley rats , we isolated cortical collecting ducts from male and female Wistar and Sprague–Dawley rats and used RT-PCR to trace the signal of GC-1. GC-1 was found in CCDs of all rats except male Sprague–Dawley rats which could explain why the American group was never able to find GC-1 in the kidney. GC-A was detected in all rats independent of strain and gender (Fig. 2).
3.2 Immunoblotting and immunohistochemistry
The rabbit GC-1 serum K252 and the rabbit GC-A serum A245 detected a signal of approximately 125 kDa in female Wistar rat kidney membrane fractions, but not in cytosolic preparations (Fig. 3). The rabbit GC-A serum A245 also detected several smaller signals which are due to protein breakdown. After longer exposure of the film some smaller signals were detected as well with rabbit GC-1 serum. The specific signals were blocked by preincubation with 1 μg/ml of the antigenic peptide (Fig. 3). Based on these findings the antibodies of these series were used for immunohistochemistry.
We localized the distribution of these receptors in the collecting duct of rat (Fig. 4). Panel A shows the labelling of collecting duct cells using an antibody against aquaporin 3 (AQP-3) which stains the basolateral membrane of principal cells (red). In panel B, an antibody against GC-A (yellow) was used showing that only the luminal membrane of the round intercalated cells are stained that protrude into the lumen. The more flat principal cells in between the intercalated cells carried no staining. In Panel C, an antibody against GC-1 (yellow) was used to show the localization of GC-1, which is mainly the luminal membrane of principal cells. In addition to the localization of this guanylyl cyclase in the CCD, an even stronger immunohistochemical staining of GC-1 was found in the IMCD and to a lesser degree also in OMCD and terminal IMCD (data not shown).
3.3 Measurements of intracellular pH (pHi)
With the knowledge of the lack of effects of natriuretic peptides in principal cells of rat CCD  and the localization of GC-A/NPR-A in the luminal membrane of apparently only a sub population of CCD intercalated cells, we focussed on the measurements of pHi in those cells. Intercalated cells accumulate BCECF to a much greater extent than principal cells , thus, the fluorescence signal from these CCD clusters were dominated by the intercalated cells. ANP (160 nM), its analog urodilatin (140 nM), and their membrane-permeable second messenger 8-Br-cGMP (0.5 mM) all decreased pHi by 0.02±0.01 (n=11), 0.03±0.01 (n=6), and 0.03±0.01 Units (n=9), respectively. The resting pHi in 46 CCD segments was 7.22±0.08. After acidification with either propionate (20 mM) or NH4+ pulses (20 mM), ANP (30±6%, n=12) as well as urodilatin (37±7%, n=8) reduced the recovery rate of pHi. 8-Br-cGMP also reduced ΔpHi/min by 19±3% (n=8). In Fig. 5 an original recording demonstrates the inhibitory effect of ANP on pHi-recovery. Amiloride (1 mM) inhibited the recovery rate by 100±30% (n=14).
Natriuretic peptides are involved in the control of renal salt, water, and urea excretion [12,13,33]. For the cortical part of the collecting duct there have been diverse reports concerning the actions of natriuretic peptides. In an early study it was reported that ANP inhibited NaCl and net fluid reabsorption in isolated CCD of rat . Later studies with the same preparation could not confirm these previous findings [19–21].
In the basolateral membrane of principal cells of rat CCD, we identified two K+ channels that are regulated by a cGMP-dependent protein kinase . Since none of the known cGMP-activating agonists, including ANP, were able to hyperpolarize the membrane potential of CCD cells , we focussed in this study on the distribution of different guanylyl cyclase receptors in principal and intercalated cells of the cortical collecting duct of rat as well as the overall distribution in the collecting duct. Using RT-PCR of RNA from microdissected CCD of rat, we obtained products corresponding to the guanylyl cyclase A receptor (GC-A) which binds ANP and for the guanylyl cyclase 1 receptor, a novel guanylyl cyclase that was first described in chemosensory tissue of rat and cattle and was also cloned at the same time from a rat kidney cDNA library [9,10]. Amplified cDNA of rat GC-1 was also found in kidney, lung, skeletal muscle, heart, olfactory and tongue epithelia . The full length receptor was cloned recently and named GC-G . While this group was unable to find the message of GC-1/GC-G in the kidneys of male Sprague–Dawley rats, but Matsuoka et al.  and we detected the message for the receptor in female Wistar rats, we had a closer look at the distribution of the mRNA in these rat strains. It seems that the message of GC-1 is only missing in male Sprague–Dawley rats which explains the described discrepancy. Specific immunohistochemical labeling of GC-1 mainly showed staining of luminal membranes of a sub population of cells of the rat CCD. These cells with rather straight luminal borders were also marked by an antibody against aquaporin-3 (AQP-3) at the basolateral membrane identifying these cells as principal cells. There might also be some staining of intercalated cells with the antibody against GC-1. So far GC-1 seems to be the only known receptor guanylyl cyclase localized in the luminal membrane of principal cells of rat CCD which makes this guanylyl cyclase a candidate for the regulation of the cGMP-dependent K+ channels in the basolateral membrane of these cells. However, the agonist that activates this receptor-coupled cyclase is still unknown. In addition to the localization of this guanylyl cyclase in the CCD an even stronger immunohistochemical staining of GC-1 was found in the IMCD and to a lesser degree also in OMCD and terminal IMCD. These localizations indicate that this guanylyl cyclase is involved in final regulation of the urine composition, including possibly the regulation of urea transport via a so far unknown agonist in addition to that by ANP and cGMP .
GC-A could be immunolocalized in the luminal membrane of another sub population of cells of rat CCD. These cells were almost round in shape and protruded into the lumen which is characteristic for intercalated cells [35–38]. Furthermore, these cells were not stained by the AQP-3 antibody. The apparent absence of any staining of principal cells with this GC-A specific antibody supports those reports that could not detect any ANP effects in principal cells of rat CCD [19–21]. GC-A had been localized before in rat CCD using RT-PCR in individual microdissected tubules . Ritter et al.  have also described immunolocalization of GC-A in rat CCD. In their study GC-A was also localized mainly in the apical pole of intercalated cells. In another study with microdissected glomeruli and collapsed tubules in suspension specific binding of radiolabeled ANP was found exclusively in glomeruli, but not in cortical collecting duct, where ANP could not enter the lumen . Also, in that study a signal for GC-A in the luminal membrane of the thin limbs of Henle's loop was missed that was later detected by immunolocalization .
A localization of GC-A in intercalated cells fits well to the results found on pHi-regulation in this study. BCECF is accumulated in intercalated cells to a much higher degree than in principal cells . In the present study the fluorescence was registered from 10 to 15 cells of CCD clusters. Thus, the signal obtained was highly dominated by the intercalated cells. The inhibitory effect of ANP on pHi-regulation in these cells was displayed in the absence of HCO3− in the experimental solutions, thus, Cl−/HCO3− exchange was not active. Under these conditions pHi recovery after acidification could be completely blocked by amiloride indicating that a Na+/H+ exchanger was responsible for the acid extrusion. Therefore, it is most likely that ANP actually regulates a Na+/H+-exchanger in intercalated cells of rat CCD. The presence of a Na+/H+ exchanger in the basolateral membrane of intercalated cells of rabbit CCD was recently demonstrated . NHE-1 is a likely candidate for these effects and can be found in the basolateral membrane of several epithelia . This conclusion is further supported by the observation that only the basolaterally localized NHE-1 and not the luminally localized NHE-2 isoform is activated by acidification . ANP is known to inhibit Na+/H+ exchange in aortic vascular smooth muscle cells  and proximal tubules of rat . The regulation of such an acid extrusion mechanism could allow a control of the cellular pH while acid/base transport occurs across the same cells as was suggested before . In addition, inhibition of this exchanger in intercalated cells would lead to an increase in pHi and thus, to an increase in H+ secretion. Further studies are necessary to elucidate the role of ANP and related peptides on acid/base transport in the CCD of the rat and to identify the physiological agonist and function of GC-1 in the collecting duct.
This work was supported by the Deutsche Forschungsgemeinschaft (Schl 277/5-1 to 5-3) and a grant from the Max-Planck-Society and the Alexander von Humboldt Foundation. Mogens Kruhøffer was supported by the Danish Centre for Respiratory Physiological Adaptation. The authors gratefully acknowledge the expert technical assistance of Dorothee Bergmeier, Sabine Haxelmans, Ingrid Kleta, Melanie Klingenberg, and Heike Stegemann. Furthermore, we would like to thank Dr Wilhelm Kriz and Dr Marlies Elger for the preparation and fixation of rat kidney for the immunohistochemical localization of guanylyl cyclase receptors. We would also like to thank Dr Jean-Marc Verbavatz for providing us with an antibody against aquaporin-3. We thank W. Posselt for excellent photographic work.