INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

NITRIC OXIDE HAS BEEN shown to play an important role in the regulation of renal blood flow and renin secretion (2, 3). Recently, a large body of evidence indicated that NO is also involved in regulation of tubule function in a variety of nephron segments (8, 9, 17, 19). Three types of nitric oxide synthase (NOS), inducible NOS (iNOS), endothelial NOS (eNOS), and neuronal NOS (nNOS), have been found to be expressed in the kidney (2, 11). It is generally believed that the constitutive NOS isoforms, nNOS and eNOS, are responsible for modulating the physiological function of cells (7).

We have previously shown that NO plays a key role in regulation of the basolateral small-conductance (28 pS) K⁺ channel in the cortical collecting duct (CCD) (8). Although several studies have shown that nNOS is expressed in the CCD (1, 18), it is not known in which type of cell nNOS is expressed in the CCD. In addition, even though mRNA encoding nNOS is detected in the CCD, whether nNOS can be detected in the protein level has not been well explored. In the present study, we used RT-PCR and immunocytochemical methods to determine the expression of constitutive NOS in the CCD.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Isolation of CCD. The CCDs were isolated from kidneys of pathogen-free Sprague-Dawley rats purchased from Taconic (Germantown, NY). The animals were kept on a high-K⁺ diet and anesthetized with pentobarbital sodium (10 mg/kg). The reasons for using the animals on a high-K⁺ diet were as follows: 1) we had previously studied the effect of NO on K⁺ channels in the CCD obtained from rats on a high-K⁺ diet; and 2) dissection of the tubules was easier in the animals on a high-K⁺ diet than those on either normal or low-Na⁺ diet. After the abdomen was opened, the kidneys were perfused with DMEM medium (Life Technologies, Grand Island, NY) and removed. Thin coronal sections were cut with a razor blade and incubated in the DMEM medium saturated with 5% CO₂-95% O₂ and containing 0.5 mg/ml type I collagenase (Sigma, St. Louis, MO) and 0.5 mg/ml protease XXV (Sigma), at 37°C for 10-25 min. The kidney slices were then rinsed three times with dissection solution composed of (in mM) 135 NaCl, 1 Na₂HPO₄, 5 KCl, 5 glucose, 1.5 MgSO₄, 2 CaCl₂, and 5 HEPES (pH 7.4). We microdissected CCDs in the dissection solution at 0°C. The tubules (1 mm length) were captured on an autoclaved glass microbead and transferred to a tube containing 10 µl lysis buffer with 5 mM dithiothreitol (Sigma), 2% (vol/vol) Triton X-100, and 1.6 U/µl RNasin in RNase-free H₂O. The samples were frozen at 80°C until they were utilized in the RT-PCR reaction.

RT reaction. RT reaction was performed using a Superscript RNase H-reverse transcription kit (Life Technologies). RT components (10 µl) were added to the reaction tubes, each of which contained 4 µl of 5× first-strand synthesis buffer, 1 µl oligo(dT)_12-18, 2 µl deoxynucleotide mixture (10 mM), 1 µl RNase inhibitor, and 2 µl RNase-free water. After we mixed oligo(dT) with sample, the tubes were heated up to 70°C for 10 min, then chilled on ice for 1 min. The reaction tubes were subsequently incubated at 42°C for 2 min, and 1 µl reverse transcriptase was then added into the reaction tubes. For each experiment, we included one RT-positive control tube provided in the kit and one negative control containing all components for the RT reaction except the reverse transcriptase. All tubes were incubated at 42°C for 50 min. The reactions were terminated by heating up to 70°C for 15 min, and then 1 µl RNase H was added to the sample and incubated at 37°C to digest RNA strand in the DNA-RNA hybristrand.

Polymerase chain reaction. A 5-µl aliquot of the first-strand cDNA was used for amplification of nNOS and eNOS, respectively, with the cDNA amplification reagent kit (Life Technologies). Each reverse-transcribed RNA sample (5 µl) was mixed with 91 µl of PCR master solution and 4 µl specific primer for rat nNOS and eNOS, respectively. For cDNA amplification, the primers used were 5' GAATACCAGCCTGATCCATGGAAC 3' (sense, position 2461-2484), and 5' TCCTCCAGGAGGGTGTCCACCGCA 3' (antisense, position 3039-3062) for rat nNOS (4); 5' CTGGCGGCGGAAGAGAAAGGAGTC 3' (sense, position 1878-1901) and 5' GGGGCTGGGTGGGGAGGTGATGTC 3' (antisense, position 2618-2641) for murine eNOS (6). To synthesize specific probes for Southern blot analysis, we used a second pair of primers for amplification of the region positioned inside of the first PCR amplification products. The primers were 5' GCTAATGGGGCAGGCCATGG 3' (sense, position 2577-2596) and 5' TCGGGAGCTAGAATAGGAGG 3' (antisense, position 2894-2913) for rat nNOS; and 5' AAGGCTGGAGGAGCTGGGCG 3' (sense, position 2011-2030) and 5' TGGGCACACACCTATGTGGT 3' (antisense, position 2394-2413) for mouse eNOS. The tubes were placed in a programmed thermocycle system (Progene, Technique), which was programmed as follows. Initial incubation was at 94°C for 2 min. Then 35 cycles of the following sequential steps were performed: 94°C for 1 min (melt); 55°C for 1 min (anneal); and 72°C for 1.5 min (extend). Finally, the tubes were heated to 72°C for 8 min. For each experiment, we also included two positive control samples (the RT of the control RNA provided in the kit and the cDNA encoding nNOS/eNOS) and two negative control samples (distilled water and the RT product made in the absence of reverse transcriptase). The probes were labeled by digoxigenin-11-dUTP (DIG) with the PCR DIG probe synthesis kit from Boehringer Mannheim (Indianapolis, IN).

Southern blot analysis of PCR product. The PCR products (55 µl) were analyzed by electrophoresis on a 1% agarose gel and staining with ethidium bromide (50 µg%, wt/vol). The PCR products separated on the agarose gel were denatured, neutralized, and blotted onto a nylon membrane (Boehringer Mannheim). The DNA was bound to the nylon membrane by ultraviolet cross linker (Hoefer Scientific Instrument), prehybridized, and hybridized with DIG-labeled eNOS or nNOS DNA probe. Subsequently, the nylon membrane was washed twice with 2× SSC, containing 0.1% SDS at room temperature and 0.5× SSC at 68°C, respectively. We followed the procedure described by the Boehringer Mannheim user's guide to detect the corresponding band. The chemiluminescent signal was captured by exposing the blot to X-ray film for 60-90 min at room temperature.

Antibody and immunoblot for nNOS. The anti-nNOS polyclonal antibody, which recognizes the peptide at positions 1400-1419 (nNOS-C antibody), was purchased from Santa Cruz (Santa Cruz, CA), and the second anti-nNOS polyclonal antibody, which recognizes the peptide at positions 724-739 (nNOS-N antibody), was obtained from Biomol (Plymouth Meeting, PA). Protein samples extracted from the kidney medulla were separated by electrophoresis on 8% SDS-polyacrylamide gels and transferred to nitrocellulose membrane. The membranes were blocked with 10% nonfat dry milk in Tris-buffered saline (TBS), rinsed and washed with 1% milk in Tween-TBS. The nNOS-C antibody and the nNOS-N antibody were diluted at 1:3,000 and at 1:1,000, respectively. The antibodies were diluted in this same buffer and hybridized to blots for 1 h. The total protein content used for immunoblot was 10-50 µg.

Immunocytochemistry. The isolated CCD was fixed with 4% paraformaldehyde for 15 min and rinsed three times with 0.1 M phosphate buffer. The tubules were treated with Triton X-100 for 3 min and incubated with H₂O₂ (0.3%) for 30 min, followed by three rinses in 0.1 M phosphate buffer containing 1% BSA (washing buffer). The tubules were blocked in washing buffer with 2% normal goat serum for 4 h and then were incubated in primary polyclonal anti-nNOS antibody (1:2,500) for 12 h at room temperature. The tubules were washed three times before the 1-h incubation in secondary antibody, biotinylated goat anti-rabbit IgG. For histochemical staining, tubules were washed and then incubated in the avidin and biotinylated horseradish peroxidase macromolecular complex (Vectastain Elite ABC Kit) in washing buffer for 30 min. The tubules were then rinsed twice in 0.1 M phosphate buffer and once in 0.1 M Tris buffer, pH 7.45 and then reacted with 3,3'-diaminobenzidine (1 mg/ml) in the presence of hydrogen peroxide. The tubules were then dehydrated and coverslipped.

To carry out immunocytochemical studies on the isolated split-open CCD, the CCD was placed in a chamber mounted on an inverted microscope (Nikon), and the tubules were superfused with HEPES-buffered NaCl solution. The CCD was cut open with a sharpened micropipette to form a monolayer-like open tubule. The CCDs were treated in the same way as described for the intact tubules.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Microdissected CCDs were analyzed for nNOS and eNOS mRNA content by RT-PCR. Fig. 1A shows that a faint band of the PCR product of 602 bp corresponding to nNOS was detected in the CCD on an ethidium bromide-stained agarose gel, and Southern blots made from the agarose gel also revealed the presence of a clear band corresponding to mRNA for nNOS in the CCD (Fig. 1B) in all 12 experiments. In contrast, we did not find the PCR product of 739 bp corresponding to eNOS in the CCD (0 of 12 experiments, data not shown), suggesting that nNOS may be the only constitutive isoform of NOS expressed in the CCD. We found two bands in the Southern blot in some experiments. It is not clear why a double band for nNOS was detected. One possible explanation is that the double band might be the result of an alternative splicing. It has been reported that alternative splicing specifically regulates nNOS µ-isoform (nNOS-µ) in striated muscle (14). Further study is needed to confirm that the alternative splicing is probably present in the CCD.

View larger version (45K): [in this window] [in a new window]   Fig. 1.   A: ethidium bromide-stained agarose gels for neuronal nitric oxide synthase (nNOS). Expected PCR product size of 602 bp is indicated by arrow. RT-PCR products from single microdissected cortical collecting ducts (CCDs) are indicated in lanes 1-5. B: RT-PCR products from single microdissected CCDs were analyzed by Southern blot using digoxigenin-11-dUTP-labeled probes recognizing nNOS.

Although we have detected the presence of nNOS in the CCD at mRNA level, it is not known whether nNOS can be identified at the protein level. However, it was difficult to carry out the immunoblot analysis in the tubules, since we could not obtain large amounts of the tubules to perform a Western blot. Therefore, we carried out immunocytochemical studies on the isolated CCD. To determine the specificity of the antibody, we carried out immunoblot analysis for nNOS with tissue obtained from the renal medulla. Figure 2, A and B, shows two representative Western blots showing immunoreaction to nNOS with the polyclonal nNOS-N antibody at 1:1,000 dilution and with the nNOS-C antibody at 1:3,000 dilution, respectively. It is apparent that both antibodies detect a band from the homogenates of brain and kidney medulla, respectively, and the bands are located at ~160 kDa, which corresponds to the size of nNOS. Using the nNOS-C antibody, we could also find a band with high molecular weight. This band may be the result of cross-reaction with nNOS antibody or, alternatively, it may be a new isoform of nNOS. However, this cross-reaction may not affect our immunocytochemical studies, since we have observed identical immunostaining of the tubules using either nNOS-C antibody (Fig. 3B) or nNOS-N antibody (Fig. 3C), which has no cross-reaction with other proteins. Accordingly, we pooled data.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Immunoreactivity of nNOS in rat kidney and brain. Nitrocellulose filter containing cell homogenates from the brain (10 µg) and from the kidney medulla (50 µg) was reacted with polyclonal nNOS-N antibody (1:1,000) (A) or polyclonal nNOS-C antibody (1:3,000) (B). Arrows indicate the molecular mass (in kDa).

View larger version (100K): [in this window] [in a new window]   Fig. 3.   Immunocytochemical staining of the CCD using two kinds of rabbit anti-nNOS antibody (1:2,500 dilution). A: a light micrograph (×300) revealing the immunoreaction to nNOS-C antibody in the CCD. Immunostaining with anti-nNOS-C antibody (B) and anti-nNOS-N antibody (C) shows the positive reaction in principal cells but not in intercalated cells (×770).

Figure 3A shows that immunoreaction for nNOS is heterogeneous in the CCD, with both positive and negative immunostaining cells. To determine the cell type, we studied nNOS immunoreactivity in the split-open tubule. We found that the positive immunoresponse is localized in the large and round-shaped cells (Fig. 3, B and C). Since we can detect the amiloride-sensitive Na⁺ channels in the patch-clamp studies using the above described morphological characteristics, the cells with positive immunoreaction to nNOS are most likely the principal cells. On the other hand, no immunoreaction can be observed in the cells with irregular shape and small size. Since we did not detect the Na⁺ channels in the patch-clamp studies, those cells should be most likely intercalated cells.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In the present study, we have confirmed the finding made by Terada et al. (18) that mRNA encoding nNOS is present in the CCD. We could not detect the mRNA encoding eNOS with our method. Although we could not exclude the possibility that failure to detect eNOS may result from using mouse eNOS sequence, this possibility is not supported by observations made by Bachmann et al. (1) in which no immunoreactivity can be detected in the renal tubules using anti-eNOS antibody. That nNOS is expressed in the CCD is further confirmed by our observations that immunoreactivity to an nNOS antibody was detected in principal cells of the CCD. McKee et al. (10) used the NOS-isozyme-independent marker NADPH-diaphorase (NADPH-d) to determine the possible distribution of NOS in that rat kidney, and they observed positive immunoreactivity to NADPH-d in some cells of the CCD, suggesting the possible presence of NOS in the CCD.

The CCD plays a key role in K⁺ secretion and hormone-regulated Na⁺ reabsorption (5, 13). Two types of cells, principal cell and intercalated cell, are present in the CCD. The principal cell is responsible for K⁺ secretion and Na⁺ reabsorption, and the intercalated cell is involved in K⁺ reabsorption, proton secretion, and bicarbonate secretion. Although nNOS has been shown to be present in the CCD, it is not known in which cell type nNOS is expressed. In the present study, we observed that the positive immunoreactivity with nNOS antibody was located only in the principal cell but not in intercalated cell. Thus the immunocytochemical data are in agreement with the functional observations that inhibition of NOS decreased the activity of the basolateral 28-pS K⁺ channel in the principal cell of the CCD (8). Interestingly, Mundel et al. (12) have demonstrated that ₂-subunits of soluble guanylate cyclase are present only in the principal cell of the CCD (12). Since NO has been shown to stimulate the soluble guanylate cyclase, colocalization of both nNOS and the guanylate cyclase strongly suggests that this NO-cGMP signal transduction pathway may play an important role in the regulation of tubule function in the CCD. Recently, it was demonstrated that NO is involved in regulation of Na⁺ and water transport (15-17). We have shown that NO has a dual effect on the low-conductance K⁺ channel (28 pS) in the basolateral membrane of the CCD. NO stimulates the basolateral 28-pS K⁺ channel via a cGMP-dependent pathway at low concentrations (8), whereas NO inhibits the channel through a cGMP-independent pathway at high concentrations (20). Since basolateral K⁺ channels participate in generating cell membrane potential and are involved in K⁺ recycling across the basolateral membrane, changes in the activity of the basolateral K⁺ channels should have an impact on Na⁺ transport.