INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE GLOMERULAR INFLAMMATORY response is characterized by a number of phenotypical changes, such as proliferation of mesangial cells (MCs) and increased deposition of extracellular matrix proteins in the mesangial space (1, 22). In addition, an increased synthesis of nitric oxide (NO) resulting from cytokine-mediated stimulation of the inducible NO synthase (iNOS) expressed by resident glomerular and inflammatory cells may contribute as a proinflammatory mechanism to the pathogenesis of glomerulonephritis. In cultured glomerular MCs, the expression of iNOS activity and mRNA on stimulation with cytokines or lipopolysaccharide (LPS) has been demonstrated by several investigators (27, 28, 35, 41). This enzyme, once expressed, releases large amounts of NO from the conversion of L-arginine plus oxygen to L-citrulline (reviewed in Ref. 23). A first report of induced NO synthesis being involved in glomerulonephritis came from a rat model of accelerated nephrotoxic nephritis (6). Molecular evidence for a participation of iNOS was provided with the identification of increased and activated iNOS enzyme in different rat models of glomerulonephritis (12, 21, 30). In certain models of autoimmune disease, inhibition of NO synthesis improved the accompanied glomerulonephritis (47). This indicated the potential proinflammatory role of induced NO synthesis.

With the destruction of resident or invading cells and with the activation of thrombocytes and their aggregation, all of which accompany glomerular inflammatory injury, large amounts of mononucleotides such as ATP and UTP are released. The nucleotides bind to cell surface receptors, which are designated as purinergic P2 receptors and expressed by many cell types. Several P2 receptors (5) have recently been cloned and can be divided into two groups: 1) P2X receptors are ATP-regulated ion channels, and 2) P2Y receptors are G protein-coupled heptahelical receptors. Each group is constituted by seven members, designated P2X1-P2X7 and P2Y1-P2Y7, respectively. The expression of P2Y2 receptors in MCs has been postulated in previous studies by functional criteria, since both ATP and UTP led to stimulation of phosphatidylinositol-specific phospholipase C (33, 40), increase in intracellular calcium (32), inhibition of cAMP accumulation (37), and activation of mitogen-activated protein kinases (18). Furthermore, it has been shown that extracellular ATP induces mitogenesis (18, 19, 39, 40) and contraction (32) in cultured rat MCs.

The aim of this study was to examine whether extracellular mononucleotides may regulate iNOS expression and activity in cultured rat MCs and which signal transduction pathways are involved in this potential regulatory pathway. Our results verify the presence of the purinergic P2Y2 receptor in cultured MCs by demonstrating its mRNA expression. Extracellular ATP and UTP strongly inhibited LPS/IFN--induced iNOS mRNA, protein, and activity by binding to purinergic P2Y2 receptors and subsequently activating protein kinase C (PKC). The release of nucleotides into the extracellular space may therefore critically regulate the generation of NO during the glomerular inflammatory response.

	MATERIAL AND METHODS

Top Abstract Introduction Methods Results Discussion References

Material. Dulbecco's modified Eagle's medium (DMEM), L-glutamate, insulin, penicillin, streptomycin, LPS (Escherichia coli serotype 026:B6), interferon- (IFN-), N^G-monomethyl-L-arginine (L-NMMA), ATP, adenosine 5'-O-(3-thiotriphosphate) (ATPS), ADP, AMP, adenosine, UTP, UDP, UMP, uridine, ,-methylene-ATP (APCPP), 3'-O-(4'-benzoyl)-benzoyl-ATP (BzATP), 2'-methylthio-ATP (2-MeS-ATP), phorbol 12-myristate 13-acetate (PMA), and 1-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H-7) were all purchased from Sigma (Deisenhofen, Germany). Ro-31-8220 was purchased from Biomol, and fetal calf serum (FCS) was from Boehringer (Mannheim, Germany). Tissue culture plastic was from Falcon (Becton-Dickinson, Heidelberg, Germany).

Cell culture and isolation of MCs. For preparation and culture of glomerular MCs from male Sprague-Dawley rats, standard techniques were used, as described previously (20, 25, 28). Briefly, the kidneys were excised, and the cortex was separated from the medulla and homogenized using razor blades. Glomeruli were isolated from the homogenate by sequential sieving and collected on a 75-µm sieve. After treatment with 500 U/ml collagenase (type IV, Sigma) in phosphate-buffered saline for 30 min, the glomerular remnants consisting predominantly of endothelial cells and MCs were seeded in tissue flasks (30,000 glomeruli/flask) containing DMEM supplemented with 20% FCS, 5 µg/ml bovine insulin, 100 U/ml penicillin, and 100 µg/ml streptomycin. After three to four passages using 0.05% trypsin-0.02% EDTA, stable and homogenous cultures of MCs were obtained. The outgrowing cells were characterized as MCs by positive immunocytochemical staining for Thy-1.1 (Serotec; Blackthorn, Bicester, UK), smooth muscle cell actin, and myosin showing typical MC morphology (25). MCs were further cultured in DMEM supplemented with 10% heat-inactivated (56°C for 1 h) FCS, 2 mM glutamate, 5 ng/ml insulin, 100 U/ml penicillin, and 1 mg/ml streptomycin. Cells were used for experiments at subconfluence or passaged at confluence with trypsin-EDTA (0.05%-0.02%, wt/vol). To obtain quiescent cells, MCs were maintained in medium containing 0.5% FCS for 3 days before cytokine treatment. MCs were used between passages 15 and 25. To stimulate iNOS induction, we used LPS (E. coli serotype 026:B6, 10 µg/ml medium) and IFN- (100 U/ml medium). Subconfluent quiescent MCs were coincubated according to the experimental conditions with ATP (10⁶-10³ M), ATPS (10⁷-10⁴ M), ADP, AMP, adenosine, UTP, UDP, UMP, uridine, APCPP, 2-MeS-ATP, BzATP (all at 10⁴ M), PMA (10⁸ M), H-7 (10⁵ M), and Ro-31-8220 (10⁷ M).

Measurement of NO production by detection of nitrite in cell culture supernatants. MCs were grown to subconfluence either in 24-well plates or to accommodate protein or RNA extraction on 60- and 100-mm plastic culture plates. The cells were conditioned in standard culture medium between 24 and 32 h, according to the experimental protocol. The nitrite content was measured using the Griess colorimetric method (13, 28). Briefly, 200 µl of the supernatant as well as control medium supplemented with known NaNO₂ concentrations serving as standard were mixed with 50 µl of Griess reagent (25 mM sulfanilamide and 25 mM naphtylethylenediamine) and 25 µl of 6 M HCl and incubated for 30 min in the dark. Optical density was measured at 550 nm with an ELISA reader, and the nitrite content of the conditioned medium was calculated according to the NaNO₂ standard curve obtained.

Preparation of radioactive cDNA probes (rat P2Y2, rat iNOS, and rat GAPDH). A probe for the rat P2Y2 receptor was provided by Dr. D. Julius (San Francisco, CA) (26). We used two sets of iNOS probes. First, we chose an EcoR I/Pst I-cut insertion of the plasmid piNOS B2 (48), a mouse iNOS cDNA clone. This 611-bp fragment is part of the 5'-coding region of the iNOS gene. Second, we used an 1,639-bp rat iNOS fragment, which we gained by homology-based reverse transcription-polymerase chain reaction (RT-PCR) (40 cycles, annealing temperature 60°C) of LPS/IFN--induced (24 h) cultured rat MCs, using the following primers (synthesized by Life Technologies): sense, 5' TGCATGGACCAGTATAAGGCAAGC 3' (position 1877-1900 of M84373); and antisense, 5' ACCTGCTCCTCGCTCAA 3' (position 3479-3495 of M84373) (added restriction sites are underlined).

The glyceraldehyde-3'-phosphate dehydrogenase (GAPDH) probe was gained using an homology-based RT-PCR (30 cycles, annealing 50°C) of rat MC cDNA, using the following primers (synthesized by Life Technologies): sense, 5' AATGCATCCTGCACCACCAA 3' (position 469-488 of M17701); and antisense, 5' GTCATTGAGAGCAATGCCAGC 3' (position 919-939 of M17701).

³²P labeling was performed for all probes using the random prime labeling kit (Boehringer).

Northern blot hybridization analysis. Total RNA from rat MCs grown to subconfluence in 100-mm culture dishes was obtained, as described earlier (7, 28). Twenty micrograms of total RNA were electrophorectically size fractionated under denaturing conditions, using 1% agarose including 1.8% formaldehyde. The separated RNA was capillary transferred by 20× standard sodium citrate (SSC) to nylon membranes (Amersham) and fixed though ultraviolet crosslinking. Next, the RNA was prehybridized (5× Denhardt's solution, 5× SSC, 50 mM Na₃PO₄, 0.1% SDS, 250 mg herring sperm DNA, 50% formamide) for at least 2 h at 42°C and then hybridized with the appropriate ³²P-labeled cDNA-specific probes (P2Y2 probe, mouse macrophage iNOS probe followed after stripping by rat vascular smooth muscle iNOS probe) for 24 h, using the same conditions. The membrane was washed at 42°C twice for 15 min with 2× SSC, 0.1% SDS, for 30 min with 0.1× SSC, 0.1% SDS, and exposed to X-ray film (Kodak XAR-5, supplied by Sigma) at 80°C for at least 24 h. To rectify for RNA transfer and content, filters were stained with bromophenol blue (0.04% in 500 mM sodium acetate, pH 5.5) and rehybridized with a ³²P-labeled rat GAPDH probe.

Purinergic P2Y2 receptor RT-PCR. One microgram of total RNA of quiescent cultured rat MCs was subjected to an RT reaction using oligo(dT)₁₆, according to the protocol of the Perkin-Elmer Cetus GeneAmp RNA PCR kit. Specific primers for the homology-based PCR (30 cycles, annealing at 56°C) produced PCR products, which were size separated on ethidium bromide-stained agarose gels, revealing DNA bands at the expected 380 bp size. Negative controls for the RT and the PCR reagents remained negative. Primer (synthesized by Eurogentec) sequences were as follows: sense, 5' CAGCGTGAGAGGGACCCGAA 3' (positions 521-540); and antisense, 5' TGAGGTCAAGTGATCGGAAG 3' (positions 881-900).

iNOS protein Western blot analysis. Preparation of MC lysates and immunoblot analysis was performed using time-matched 60-mm plates of MCs. They were incubated with complete medium containing 0.5% FCS, medium supplemented with LPS (10 µg/ml medium)/IFN- (100 U/ml medium) or medium supplemented additionally with reagents according to the experimental protocol. After the incubation period, the medium was removed and assayed for the nitrite content, as described above. The plates were washed twice with ice-cold phosphate-buffered saline. Thereafter, 0.5 ml of boiling lysis buffer solution (1% SDS, 10 mM Tris, pH 7.4) was added to the cells. The cell lysates were collected in centrifuge tubes after boiling and spinning at 12,000 g for 5 min at 4°C. Protein contents were determined using the bicinchoninic acid protein assay reagent (BCA; Pierce, Munich, Germany). The samples (10 µg) were diluted in electrophoresis sample buffer (250 mM Tris, pH 6.8, 4% SDS, 10% glycerol, 0.006% bromophenol blue, 2% mercaptoethanol), boiled for 5 min, quenched on ice, and resolved by electrophoresis though 0.1% SDS-10% polyacrylamide gels. Proteins were electrophoretically transferred to nitrocellulose membranes (Amersham, Braunschweig, Germany), and the transfer efficiency was determined by staining the membranes with Ponceau S. After destaining in distilled water, the membranes were quenched in blocking solution [5% powdered dried low-fat milk in washing solution (10 mM Tris, pH 7.5, 100 mM NaCl, 0.1% Tween 20)] 60 min at room temperature. The blocking solution was decanted, and membranes were incubated for 60 min at room temperature with the primary antibody [anti-iNOS, rabbit, polyclonal (Affiniti, Nottingham, UK), 1:1,000]. Membranes were washed for 30 min at room temperature with several changes of the washing solution. The secondary detecting antibody was horseradish peroxidase-conjugated anti-rabbit IgG (1:2,000; Serva, Heidelberg, Germany), which was used with the enhanced chemiluminescence protocol (Amersham).

Statistical analysis. When suitable, means ± SE were determined. To test for statistically significant differences, we used Student's t-test or analysis of variance, if multiple comparisons were made against a single control, whenever applicable. Significance was assigned at P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

ATP and ATPS inhibited nitrite accumulation in supernatant from cultured rat MCs. Treatment of cultured rat MCs with LPS (10 µg/ml medium)/IFN- (100 U/ml medium) for 24 h led to a significant induction (P < 0.001) of nitrite production in these cells [control, 1.2 ± 0.4 µmol/ml medium; LPS/IFN- induction, 17.6 ± 0.7 µmol/ml medium (100%)], which was inhibited by the addition of the NOS inhibitor L-NMMA (10⁴ M) by 87.3 ± 4.9% (P < 0.001). This indicated that nitrite accumulation in our experimental system was due to NOS activity (43). When ATP was given together with LPS/IFN- dose dependently the nitrite production was reduced, significantly by 48.2 ± 6.3% at 10³ M (P < 0.001 vs. control) (Fig. 1). Addition of the stable ATP analog ATPS (10⁴ M) led to a pronounced inhibition of NO production by 50.3 ± 6.5% (P < 0.001). Neither ATP nor ATPS showed an effect by themselves. The inhibitory effect of ATP and ATPS was also observed with preincubation of the nucleotides for up to 8 h before adding LPS/IFN- (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   ATP inhibits nitrite accumulation in supernatant from cultured rat mesangial cells (MCs). ATP dose dependently inhibited synergistic upregulation of nitrite formation by lipopolysaccharide/interferon- (LPS/IFN-). Nitrite was measured in quiescent MCs incubated for 24 h with indicated agents [LPS (10 µg/ml medium)/IFN- (100 U/ml medium), 104 M NG-monomethyl-L-arginine (L-NMMA)]. Nitrite concentrations were determined in the supernatant using the Griess reaction. Results are expressed in % standard LPS/IFN- induction. Data are presented as means ± SE with assays performed in duplicate (n = 6-28); ind, induction of inducible nitric oxide synthase (iNOS) with LPS/IFN-. * P < 0.001 vs. LPS/IFN-.

Effects of different P2 receptor agonists on nitrite accumulation in supernatants from cultured rat MCs. The P2Y2 receptor agonists ATP, UTP, and ATPS reduced LPS/IFN--induced nitrite accumulation in cultured rat MCs by 36.4 ± 4.3, 29.9 ± 5.8, and 50.3 ± 6.5%, respectively (P < 0.001 vs. control). No inhibitory effects could be observed with ATP metabolites such as ADP, AMP, or adenosine. Furthermore, the P2Y1 agonist 2-MeS-ATP, the P2X1-P2X6 agonist APCPP, the P2X7 agonist BzATP, and the P2Y6 agonist UDP were without effect (Fig. 2).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Purinergic P2 receptor agonists affecting nitrite accumulation in supernatant from cultured rat MCs. ATP, UTP, and adenosine 5'-O-(3-thiotriphosphate) (ATPS) inhibited upregulation of nitrite formation by LPS/IFN-. These agonists had no effect on LPS/IFN--induced nitrite formation: ATP degradation products ADP, AMP, and adenosine; P2Y1 receptor agonist, 2'-methylthio-ATP (MeSATP); the P2Y6 receptor agonist, UDP; the P2X1-P2X6 receptors agonist, ,-methylene-ATP (APCPP); and P2X7 receptor agonist, 3'-O-(4'-benzoyl)-benzoyl-ATP (BzATP). All nucleotides have been applied at 104 M. Nitrite was measured in quiescent MCs incubated for 24 h with the indicated agents [LPS (10 µg/ml medium)/IFN- (100 U/ml medium)]. Nitrite concentrations were determined in the supernatant using the Griess reaction. Results are expressed in % standard LPS/IFN- induction. Data are presented as means ± SE with assays performed in duplicate (n = 6-28). * P < 0.001 vs. LPS/IFN- induction.

Expression of P2Y2 receptor mRNA in cultured rat MCs. Homology-based RT-PCR was performed with total RNA from quiescent cultured rat MCs providing PCR products of the expected size of 380 bp visualized on ethidium bromide-stained agarose gels (n = 5), as shown in Fig. 3. Negative controls for the RT and PCR remained negative. In parallel, Northern blot analysis on the same RNA sample has been performed after size fractioning by using a mouse cDNA probe hybridizing to the mRNA to give a signal at 3.0 kb during autoradiographic detection (n = 2) (gift by Dr. D. Julius).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Expression of P2Y2 receptor mRNA in quiescent cultured rat MCs. Left: 1.5% ethidium bromide-stained agarose gel of P2Y2 PCR products amplified from first-strand cDNAs or RNAs (left ctr lane) or PCR without cDNA (right ctr lane). MC lanes indicate first-strand cDNA as template from 2 of 5 independent experiments and RNA extractions from quiescent cultured rat MCs. A molecular weight standard is present in the right lane; ctr, control for contamination with genomic DNA (left ctr lane) and for contamination of PCR reagents with P2Y2 PCR products (right ctr lane). Right: expression of mRNA for P2Y2 receptor. Ten micrograms of RNA harvested from quiescent cultured rat MCs were subjected to size fractioning by 1% agarose gel electrophoresis, transfer to nylon membranes, and hybridization using 32P-labeled cDNA from the mouse (gift from Dr. D. Julius) (n = 2).

Coincubation with ATP reduced the cumulative iNOS mRNA expression after 24 h of stimulation with LPS/IFN- in cultured rat MCs. To further elucidate the inhibitory mechanism of ATP on LPS/IFN--induced NO production, we examined the effect of P2Y2 receptor activation on iNOS mRNA levels in MCs. Coincubation with ATP reduced the iNOS transcripts in resting cultured rat MCs stimulated with LPS/IFN- in 4 of 4 experiments. The reduced iNOS mRNA corresponded with the effect on NO production, as assessed by the nitrite accumulation in the cell culture supernatant (Fig. 4).

View larger version (69K): [in this window] [in a new window]   Fig. 4.   Coincubation with ATP reduced cumulative iNOS mRNA expression on Northern blot analysis after 24 h of stimulation with LPS/IFN- in cultured rat MCs. Treatment of cultured rat MCs for 24 h with either vehicle, LPS/IFN-, ATP, or LPS/IFN- + ATP [LPS (10 µg/ml medium), IFN- (100 U/ml medium) + 104 M ATP] was followed by harvesting total RNA, electrophoresis of 10 µg, transfer to nylon membranes, and hybridization using a 32P-cDNA of the murine iNOS clone piNOS B2. Top: variation of the RNA content of each lane was controlled by rehybridization with a 32P-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe (1 representative experiment is presented, n = 4). Bottom: densitometrical analysis of autoradiography detection is presented as the ratio of iNOS to GAPDH.

Coincubation with ATP reduced iNOS protein levels after 24 h of stimulation with LPS/IFN- in cultured rat MCs. To investigate the influence of the reduced iNOS mRNA levels during P2Y2 receptor activation on iNOS protein availability for NO production, we coincubated MCs with ATP or ATPS during LPS/IFN- induction. iNOS protein content was reduced in 9 of 11 experiments on Western blot analysis (Fig. 5).

View larger version (57K): [in this window] [in a new window]   Fig. 5.   Coincubation with ATP for 24 h reduces iNOS protein levels in cultured rat MCs stimulation with LPS/IFN- as assessed by Western blot analysis. Treatment of cultured rat MCs for 24 h with either vehicle, LPS/IFN-, or LPS/IFN- + ATP [LPS (10 µg/ml medium), IFN- (100 U/ml medium), 104 M ATP] was followed by preparing total cell lysates, electrophoresis of 10 µg protein by SDS-polyacrylamide gel electrophoresis, transfer to nitrocellulose membranes, and probing with an anti-iNOS mouse polyclonal antibody and peroxidase-conjugated rabbit anti-mouse IgG. Top: signal was detected by the enhanced chemiluminescence method (1 representative experiment is presented, n = 11). Bottom: densitometrical analysis of autoradiography detection.

Inhibition of iNOS induction by ATP is dependent on PKC activation. The PKC inhibitors Ro-31-8220 and H-7 have been evaluated to determine whether they alter LPS/IFN--induced nitrite production in the presence or absence of ATP or the PKC activator PMA. Dose-response experiments have been performed for all drugs used, and the results have been compared with common dosages provided in the literature (data not shown). The PKC inhibitors alone or in combination with LPS/IFN- did not influence NO production. Also PMA alone did not lead to an iNOS induction. The inhibitory effect of ATP (P < 0.001) was completely reversible by Ro-31-8220 (P < 0.02) and partially reversible by H-7 (P = 0.05). Furthermore, the decrease in nitrite accumulation provoked by PMA (P < 0.001) was significantly opposed (P < 0.02) by Ro-31-8220 (Fig. 6).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Inhibition of LPS/IFN--induced NO production by ATP depends on protein kinase C (PKC) activation and is paralleled by the PKC activator phorbol 12-myristate 13-acetate (PMA). Nitrite was measured in quiescent MCs incubated for 24 h with indicated agents [LPS (10 µg/ml medium) and IFN- (100 U/ml medium), Ro-31-8220 (Ro31), 1-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H-7), and PMA]. Concentrations of reagents are provided. Nitrite concentrations were determined in the supernatant using the Griess reaction. Results are expressed in % standard LPS/IFN- induction. Data are presented as means ± SE with assays performed in duplicate (n = 6-28). * P < 0.001 vs. LPS/IFN-, + P < 0.02 vs. LPS/IFN- + 104 M ATP, § P = 0.05 vs. LPS/IFN- + 104 M ATP.

Coincubation of LPS/IFN--induced MCs with ATP led to reduced iNOS protein expression, as has been discussed before. A similar effect on iNOS enzyme abundance was observed when PMA was used instead of ATP. The inhibitory effect of ATP and PMA on the expression of iNOS protein was reversed by adding either the PKC inhibitor Ro-31-8220 or H-7, as shown in Fig. 7 (n = 3). The addition of Ro-31-8220 or H-7 without ATP or PMA during the induction period did not affect the iNOS protein expression (n = 2).

View larger version (39K): [in this window] [in a new window]   Fig. 7.   Reduction in iNOS protein expression by ATP depends on PKC activation. Treatment of cultured rat MCs for 24 h with either vehicle, LPS/IFN-, or LPS/IFN- + ATP [LPS (10 µg/ml medium) and IFN- (100 U/ml medium), 104 M ATP, 108 M Ro-31-8220 (Ro), 105 M H-7, 108 M PMA] was followed by preparing total cell lysates, electrophoresis of 10 µg protein by SDS-polyacrylamide gel electrophoresis, transfer to nitrocellulose membranes, and probing with an anti-iNOS mouse polyclonal antibody and peroxidase-conjugated rabbit anti-mouse IgG. Top: signal was detected by enhanced chemiluminescence method (1 representative experiment is presented, n = 3). Bottom: densitometrical analysis of autoradiography detection.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Injury of glomerular cells and aggregation of thrombocytes occur in many types of glomerular disease, both in experimental animals and in humans (22). Nucleotides are present in the cytosol of all cell types, as well as in the dense granules of thrombocytes. These molecules are thought to be released into the extracellular space during glomerular injury. The concentration of ATP in the cytosol of most cell types is 2-5 mM. However, in the dense granules of thrombocytes, the ATP level is even higher and reaches the molar range (11). Therefore, the concentration of ATP is very likely in the millimolar to micromolar range in vivo in the vicinity of injured cells and aggregated thrombocytes.

Glomerular MCs are thought to play a critical role in the progression of glomerular disease (42). It has been proposed by several authors that MCs are targets for the effects of extracellular nucleotides. Extracellular ATP induced MC proliferation in cell culture (18, 19, 40). Furthermore, studies in animal models of experimental glomerulonephritis support the concept that extracellular nucleotides are proinflammatory mediators in glomerular disease (36).

To further investigate the possible role of extracellular nucleotides in the glomerular inflammatory response, we examined whether extracellular nucleotides may regulate the expression of iNOS, which is thought to play an important role in the pathogenesis of glomerulonephritis (12, 21, 30). We found that the nucleotides ATP, UTP, and ATPS dose-dependently inhibited nitrite production in cultured rat MCs during coincubation with LPS/IFN-, which are known inducers of iNOS. The reduced nitrite accumulation paralleled the reduction in iNOS mRNA and iNOS protein, indicating that nucleotides primarily decrease iNOS mRNA levels, which resulted in diminished iNOS enzyme and NO production.

In additional experiments, we sought to determine the P2 receptors and signal transduction pathways involved in the inhibitory effect of extracellular ATP on iNOS induction in cultured MCs. In recent years, fourteen different P2 receptors (P2X1-P2X7 and P2Y1-P2Y7), which are stimulated by extracellular nucleotides, have been cloned (3, 10). Although specific antagonists for the different receptor types are lacking, the agonistic potency of various nucleotides at each P2 receptor has been determined (3, 10). Based on these pharmacological data, we were able to demonstrate that the inhibitory effect of ATP on iNOS expression was due to activation of the P2Y2 receptor (previously designated as P2U receptor) but not activation of any of the other known P2 receptors. The observed effect was not due to P2X receptor activation, since APCPP, which stimulates P2X1-P2X6 receptors, and the P2X7 agonist BzATP were without effect. Furthermore, the lacking effect of 2-MeS-ATP, ADP, and UDP excluded P2Y1, P2Y3, and P2Y6 receptors, respectively. The P2Y4 receptor was excluded since it is stimulated only by UTP but not ATP. P2Y5 and P2Y7 receptors are ATP binding molecules, which never proved to be functional receptors (16, 24). Therefore, it is concluded that ATP-mediated inhibition of iNOS activation is due to activation of P2Y2 receptors, which have been shown to be stimulated by ATP, UTP, and ATPS (26). Correspondingly, we found that MCs expressed mRNA for the P2Y2 receptor using homology-based RT-PCR and Northern blot analysis.

We excluded effects of ATP degradation products by comparing the ATP effect with the stable analog, ATPS, which also inhibited LPS/IFN--induced nitrite production in MCs. By contrast, the degradation products of extracellular ATP (ADP, AMP and adenosine) did not mimic the inhibitory effect of ATP.

P2Y2 receptors belong to the family of G protein-coupled heptahelical receptors, which have been shown to activate phospholipase C and PKC in MCs (33, 34, 40). Therefore, we investigated the potential involvement of PKC during the signal transduction events leading to suppression of iNOS. We demonstrated that the inhibitory effect of ATP on iNOS protein expression and nitrite production could be reversed by different PKC inhibitors, suggesting an involvement of PKCs. The ATP-dependent decrease of LPS/IFN--induced NO production and iNOS expression was mimicked by coincubation with PMA. This effect was reversible on addition of the PKC inhibitors. In other cell types, contrasting effects of PKCs on iNOS expression have been described. In avian osteoclasts, which also do enhance iNOS expression in response to inflammatory stimuli, acute administration of PMA led to an increase in iNOS mRNA, protein, and nitrite (44). In contrast, a number of reports have been published on either interleukin-1- or LPS-stimulated cell culture models. There, using, e.g., MCs or macrophage cell lines, acute treatment with PMA reduced iNOS expression (29, 31). But even activation of identical PKC isoforms did not have comparable effects on iNOS expression in different cell lines (9, 29, 31). At present, we cannot ultimately determine which PKC isoform is involved in ATP-mediated suppression of iNOS induction in MCs until selective inhibitors of PKC isoforms are available (17).

Very recently, we found that cultured rat MCs express G protein-coupled P2Y6 receptors, which are stimulated by extracellular UDP (38). Although stimulation of the P2Y6 receptor by extracellular UDP results in signaling events similar to P2Y2 receptor-mediated signaling, extracellular UDP did not inhibit LPS/IFN--induced iNOS activation as observed with ATP and UTP. Activation of these two different nucleotide receptors expressed by cultured MCs resulted in different phenotypic changes, although the primary signal transduction pathways appear to be similar. The molecular mechanisms involved in these divergent effects have not yet been determined.

In contrast to our finding in cultured MCs, Denlinger et al. (8) reported a P2Y1 signaling-mediating iNOS suppression induced by ATP in macrophages (ex vivo after LPS stimulation and in vitro after stimulation with LPS/IFN-). Tonetti et al. (46) described a stimulation of iNOS activity via not further specified P2Y receptors in the macrophage cell line RAW 264.7. Greenberg et al. (14) reported that intratracheal administration of the P2Y1 receptor agonist 2-MeS-ATP into rats increased iNOS expression in alveolar macrophages obtained by bronchoalveolar lavage. Concluding from these results, which may be due to the differing cell types and the stimuli involved, we would like to stress the importance of a molecular characterization of the assumed purinergic receptor. Lustig et al. (26) and Tokuyama et al. (45) provided first evidence for the expression of mRNA for P2Y1 and P2Y2 receptors in the kidney. We extended these observations by showing that quiescent cultured rat MCs express mRNA encoding the functionally postulated P2Y2 receptor.

A cooperative effect of the short-term mediators ATP and NO has been suggested to regulate local blood flow (4). Several investigators have shown that extracellular ATP, e.g., released on shear stress, stimulates NO release from endothelial cells via activation of endothelial NO synthase (2, 4, 15). On the other hand, at subendothelial vascular smooth muscle cells, ATP has been shown to cooperate with norepinephrine in inducing vasoconstriction (4). Since MCs are smooth muscle-like cells of the glomerulus, it is conceivable that the suppressive effect of ATP on iNOS induction serves to preserve the contractile response of glomerular MCs in inflammatory conditions.

Another interrelationship of ATP and NO must be considered in conditions of severe tissue injury. NO generated by the action of iNOS has been shown to cause MC injury in some models of experimental glomerulonephritis (30). Therefore, it is very likely that this detrimental action of NO will result in nucleotide release in this experimental model. Considering the results of the present study, extracellular ATP released from injured tissue may serve to limit further activation of iNOS to prevent excessive tissue damage. As soon as selective P2 receptor blockers are available, these hypotheses on the effects of extracellular ATP and P2Y2 receptor activation on iNOS expression need to be examined in animal models of glomerulonephritis.