INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE MEDULLARY THICK ascending limb (MTAL) is central to the regulation of extracellular fluid volume and is responsible for establishing the osmotic gradient in the medulla, the critical event for concentrating urine (15, 16). This nephron segment metabolizes arachidonic acid (AA) via a cytochrome P-450-dependent pathway (CyP450-AA) to several products, including 20-hydroxyeicosatetraenoic acid (20-HETE), which contribute importantly to MTAL function by regulating ion transport mechanisms (5). Despite immunohistochemical evidence that cyclooxygenase (COX) levels in the MTAL are low (38), several reports have demonstrated substantial levels (10⁸-10⁶ M) of PGE₂ by MTAL preparations in vitro (11, 17, 24). Indeed, PGE₂ also may be an important regulator of ion transport in the MTAL, as this prostanoid has been shown to inhibit basolateral Na⁺-K⁺-ATPase (Na⁺ pump) and the apical Na⁺-K⁺-2Cl cotransporter, two important transepithelial ion transport mechanisms present in MTAL epithelial cells (18, 40, 44).

The presence of two distinct COX isozymes, COX-1 and COX-2, has been well-documented. COX-1 is constitutively expressed in many cell types, whereas COX-2 gene transcription is induced by mitogens, growth factors, cytokines, and tumor promoters (4, 7, 8, 12, 33). Expression of COX-2 in the kidney has been reported (14), and recent immunohistochemical localization studies have shown that COX-2 is present in a subset of tubular epithelial cells located in the cortex and outer medulla (43). These cells contained both Na⁺-K⁺-ATPase and Tamm-Horsfall protein, indicating that they are TAL cells.

We previously demonstrated that the MTAL produces tumor necrosis factor- (TNF) when stimulated with lipopolysaccharide (LPS) or angiotensin II (ANG II) (9, 26). This cytokine may be an important mediator/modulator of ion transport in the MTAL (6), as exogenous TNF, or TNF produced by the MTAL in response to LPS or ANG II, inhibited ouabain-sensitive ⁸⁶Rb uptake by this nephron segment via a prostanoid-dependent mechanism, an effect consistent with the reported natriuretic action of TNF (6, 42). The prostanoid-dependent component of this mechanism required a latency period of more than 4 h. As TNF increased the expression of COX-2 in several cell types, the present study was designed to determine whether MTAL cells express COX-2 after challenge with this cytokine and to test the hypothesis that the latent period for inhibition of ⁸⁶Rb uptake by TNF was related to increases of COX-2 gene transcription and protein expression.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Male Sprague-Dawley rats (Charles River Lab, Wilmington, MA) weighing 100-115 g were maintained on standard rat chow (Ralston-Purina, Chicago, IL) and given tap water ad libitum.

Reagents. Tissue culture media and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers were obtained from Life Technologies (Grand Island, NY). Reagent-grade chemicals and collagenase (type 1A) were from Sigma (St. Louis, MO). COX-1 and COX-2 antisera and primers were obtained from Cayman (Ann Arbor, MI). In addition, another COX-2 primer set, obtained from Life Technologies, was used in some experiments. TNF was purchased from Genzyme (Boston, MA), and NS-398 was from Biomol (Ann Arbor, MI). Polyvinylidene difluoride (PVDF) membranes were obtained from Amersham (Arlington Heights, IL).

Isolation of MTAL cells. MTAL cells were isolated and characterized as previously described (6, 26). Briefly, male Sprague-Dawley rats were anesthetized with an intraperitoneal injection of pentobarbital (0.65 mg/100 g body wt). The kidneys were perfused with sterile 0.9% saline, via retrograde perfusion of the aorta, and cut along the corticopapillary axis. The inner stripe of the outer medulla was excised, minced with a sterile blade, and incubated for 10 min at 37°C in a 0.75% collagenase solution gassed with 95% oxygen. The suspension was sedimented on ice, mixed with Hanks' balanced salt solution (HBSS) containing 2% BSA, and the supernatant containing the crude suspension of tubules was collected. The collagenase digestion was repeated four times with the remaining undigested tissue. The combined supernatants were spun, resuspended in HBSS, and filtered through a 52-µm nylon mesh (Fisher Scientific, Springfield, NJ). The filtrate was discarded, and the tubules retained on the mesh were resuspended in DMEM (Life Technologies). Combination of the perfusion and size exclusion steps was done to eliminate blood elements and yielded a tubule suspension that was ~95% MTAL, as previously described by our laboratory (6, 26) as well as by other investigators (19, 41). The cells were cultured in DMEM-Ham's F-12 medium (1:1) (GIBCO), 5% fetal calf serum (FCS), epidermal growth factor (20 ng/ml; Life Technologies), streptomycin-penicillin (100 U/ml), and Fungizone (1 µg/ml; Life Technologies). After ~5-7 days, monolayers of cells were 80-90% confluent, and dome formation, indicative of vectorial transport, was exhibited. The cells were quiesced in DMEM containing 0.5% FCS for 18-24 h prior to their use.

Isolation of total RNA/RT-PCR analysis. Total RNA was isolated by lysing the cells in Trizol reagent (Life Technologies) and precipitation with isopropyl alcohol. A 3-µg aliquot of total RNA isolated from unstimulated or stimulated MTAL cells was used for cDNA synthesis using the Superscript Preamplification system (Life Technologies) in a 20-µl reaction mixture containing Superscript II reverse transcriptase (200 U/µl) and random hexamers (50 ng/µl). The reaction was incubated at room temperature for 10 min to allow extension of the primers by reverse transcriptase, then at 42°C for 50 min, 70°C for 15 min, and 4°C for 5 min. An aliquot of the cDNA was then amplified using Taq DNA polymerase (2.5 U) in the presence of sense and antisense primers (1 µM) for murine COX-1, COX-2, or GAPDH. In control experiments, total RNA was amplified prior to cDNA synthesis to exclude the possibility of contamination with genomic DNA. The PCR primer sets for COX-1, COX-2, and GAPDH amplified specific genes of transcript sizes 756 bp, 724 bp, and 983 bp, respectively. A second primer set for COX-2 (Life Technologies) yielded a PCR product that was 304 bp. The amplification (35 cycles) was initiated by 1 min of denaturation at 94°C, 1 min of annealing at 53°C, and polymerization for 2 min at 72°C followed by autoextension at 72°C for 8 min. PCR products were quantitated by normalizing mRNA accumulation for either COX-1 or COX-2 with GAPDH.

Southern blot analysis. The amplified PCR product was analyzed by Southern blot analysis following electrophoresis through a 1.5% agarose gel, transfer to nitrocellulose, and hybridization with a ³²P-labeled cDNA probe for COX-2. Blots were analyzed using a Molecular Dynamics Storm Phosphorimager.

Western blot analysis of COX proteins. After treatment with TNF (1 nM) or phorbol myristate acetate (PMA, 1 µM), the media were removed and cells washed twice with PBS. Cells were harvested, centrifuged at 600 g for 4 min in the cold room and lysed using 10 mM Tris · HCl, pH 7.5, 1 mM EDTA, and 1% SDS for 5 min on ice. The lysate was centrifuged at 10,000 g for 20 min at 4°C. Protein concentrations were determined using a detergent-compatible Bio-Rad protein assay kit. Thirty micrograms of cell lysate were mixed with an equal volume of 2× SDS-PAGE sample buffer (100 mM Tris-Cl, pH 6.8, 200 mM dithiothreitol, 4% SDS, 0.2% bromophenol blue, and 20% glycerol) and boiled for 3 min. The proteins in the cell lysate were separated on a 10% SDS-PAGE gel and transferred to nitrocellulose or PVDF membranes. Nonspecific sites on the membrane were blocked by incubating in blocking solution containing 5% nonfat dry milk in Tris buffer saline + Tween (TBST) at room temperature for 30 min. Membranes were immunoblotted with a rabbit anti-mouse COX-2 polyclonal antibody or mouse anti-sheep COX-1 monoclonal antibody for 1 h at room temperature. Membranes were washed with TBST and incubated with horseradish peroxidase (HRP)-conjugated antisera (Santa Cruz, CA) for 30 min at room temperature. Membranes were washed, and COX proteins were detected by the enhanced chemiluminescence system (ECL; Amersham, Arlington Heights, IL). Alternatively, enhanced chemifluorescence and phosphorimaging was used for the analysis of COX-2 protein expression.

PGE₂ ELISA. Quiescent MTAL cells were incubated with TNF (1 nM) or PMA (1 µM) in media containing 0.5% serum for varying times, after which the cell-free supernatants were assayed for PGE₂ by ELISA (Oxford Biomedical Research, Oxford, MI). Briefly, 50 µl of diluted medium and 50 µl of HRP-conjugated PGE₂ were added for 1 h to wells of a 96-well plate that had previously been coated with anti-PGE₂ antibody. Following incubation, substrate for HRP was added to each well for 30 min, and the reaction was stopped by addition of 1 N HCl. Quantitation was achieved by measuring absorbance at 450 nm.

Rubidium (⁸⁶Rb) uptake. MTAL cells grown in 24-well tissue culture plates were incubated in buffer containing (in mM) 140 NaCl, 1.0 CaCl₂, 1.0 MgCl₂, 4 KCl, 20 HEPES, and 5.0 glucose. Uptake was initiated by adding ⁸⁶Rb (0.5-1.0 µg, specific activity 500 mg/mCi; Amersham) for 10 min at 37°C. Isotope uptake was terminated by addition of a stop solution (10 mM HEPES and 100 mM MgCl₂), and cells were then washed twice with stop solution. The cells were lysed with 1% SDS and 4 mM EDTA, and the radioactivity associated with the cell pellet was determined in a gamma counter. The ouabain-sensitive component of total ⁸⁶Rb uptake was calculated by subtracting ⁸⁶Rb uptake in the presence of ouabain (1 mM) from uptake in the absence of ouabain, as previously described (6).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effects of COX-2-selective inhibition on PGE₂ production. MTAL cells were preincubated for 15 min in the absence or presence of the COX-2-selective inhibitor, NS-398 (13), and then challenged for various times with TNF (1 nM) or PMA (1 µM, positive control). Production of PGE₂ increased slightly in unstimulated (control) cells over the 20-h incubation period (Fig. 1, A and B). PMA significantly increased PGE₂ production at each of the time points tested; a twofold increase was observed after 2 h, whereas a four- to fivefold increase was seen at 6-20 h (Fig. 1A). Incubation of MTAL cells with TNF for 20 h increased PGE₂ production approximately threefold (Fig. 1B). However, PGE₂ production did not increase significantly when cells were incubated with TNF for either 2 or 6 h. NS-398 abolished PMA- and TNF-mediated increases in PGE₂ production at 20 h and PMA-induced increases at 6 h (Fig. 1, A and B). NS-398 did not inhibit basal PGE₂ production or PMA-mediated increases in PGE₂ production at 2 h, possibly reflecting production of this prostanoid by a COX-2-independent mechanism. Indeed, pretreatment of cells with indomethacin (1 µM) attenuated basal PGE₂ production by ~50% at 3 and 20 h (Fig. 2). Indomethacin also completely inhibited PMA-mediated PGE₂ production and reduced PGE₂ to below basal levels (Fig. 2). These data suggest that induction of COX-2 protein by TNF and PMA may be linked to PGE₂ production via this COX isoform and that both COX isoforms are active in the MTAL.

View larger version (37K): [in this window] [in a new window]   Fig. 1.   Effects of NS-398 on phorbol myristate acetate (PMA)- and tumor necrosis factor- (TNF)-mediated PGE2 production. Medullary thick ascending limb (MTAL) cells were quiesced in DMEM containing 0.5% FCS for 18 h then preincubated for 15 min in absence or presence of NS-398 (0.1 µM). Cells were challenged with either PMA (1 µM, A) or TNF (1 nM, B) for the indicated times, and the concentration of PGE2 was determined by ELISA. Data are means ± SD of triplicate determinations (n = 3, P  0.01).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Differential effects of indomethacin and NS-398 on MTAL PGE2 production. MTAL cells were quiesced in DMEM containing 0.5% FCS for 18 h, then preincubated for 15 min in absence or presence of NS-398 (0.1 µM) or indomethacin (Indo, 1 µM). Cells were challenged with PMA for the indicated times, and the concentration of PGE2 was determined by ELISA. Data are means ± SD of triplicate determinations (n = 3). * P  0.05. ** P  0.01.

View larger version (71K): [in this window] [in a new window]   Fig. 3.   RT-PCR of cyclooxygenase-1 (COX-1) and COX-2 mRNA accumulation in MTAL cells challenged with PMA. A: MTAL cells were quiesced for 18 h and then challenged with PMA (1 µM) for the indicated times. Total RNA was extracted and mRNA accumulation was assessed by RT-PCR using primer sets specific for COX-2, COX-1, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). B: relative intensities of bands was determined by scanning densitometry, and data were normalized based on the amounts of GAPDH mRNA detected in each sample. * P  0.05 and ** P  0.01; n = 3.

View larger version (58K): [in this window] [in a new window]   Fig. 4.   RT-PCR of COX-1 and COX-2 mRNA accumulation in MTAL cells challenged with TNF. A: MTAL cells were quiesced for 18 h and then challenged with TNF (1 nM) for the indicated times. Total RNA was extracted, and mRNA accumulation was assessed by RT-PCR using primer sets specific for COX-2, COX-1, and GAPDH. B: relative intensities of bands were determined by phosphorimaging and analysis with Imagequant software; data were normalized based on the amounts of GAPDH mRNA detected in each sample. * P  0.05; n = 4.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Southern blot analysis. MTAL cells were quiesced for 18 h and then challenged with PMA (1 µM) for 2 h. Total RNA was extracted, and mRNA accumulation was assessed by Southern blot analysis using a 32P-labeled cDNA probe specific for COX-2. This blot is representative of 3 experiments.

View larger version (44K): [in this window] [in a new window]   Fig. 6.   Effects of actinomycin D (AcD) on COX-2 mRNA accumulation. A: MTAL cells were quiesced for 18 h and then challenged with either PMA (1 µM) or TNF (1 nM) for 2 h following pretreatment for 30 min with AcD (1 µM). Total RNA was extracted, and mRNA accumulation was assessed by RT-PCR using primer sets specific for COX-2 or GAPDH. B: relative intensities of bands were determined by phosphorimaging and analysis with Imagequant software; data were normalized based on the amounts of GAPDH mRNA detected in each sample. * P  0.01 and ** P  0.001; n = 3.

View larger version (41K): [in this window] [in a new window]   Fig. 7.   Effects of cycloheximide (CHX) on COX-2 mRNA accumulation. A: MTAL cells were quiesced for 18 h and then challenged with either PMA (1 µM) or TNF (1 nM) for 2 h following pretreatment for 15 min with CHX (1 µM). Total RNA was extracted, and mRNA accumulation was assessed by RT-PCR using primer sets specific for COX-2 or GAPDH. B: relative intensities of bands were determined by phosphorimaging and analysis with Imagequant software; data were normalized based on the amounts of GAPDH mRNA detected in each sample.  P  0.05 vs. control; * P  0.01 vs. control; ** P  0.01 vs. PMA or TNF; n = 3.

View larger version (39K): [in this window] [in a new window]   Fig. 8.   Kinetics of PMA- and TNF-mediated COX-2 protein expression. A: MTAL cells were quiesced for 18 h, then incubated for the indicated times in absence or presence of PMA (1 µM) or TNF (1 nM). Cell lysates (30 µg) were separated on a 10% SDS-PAGE gel; COX-2 was detected by Western blot analysis. Molecular weight of COX-2 (72/74 kDa) is indicated. B: relative intensities of bands were determined by phosphorimaging and analysis with Imagequant software.. * P  0.05 and ** P  0.01; n = 3.

View larger version (15K): [in this window] [in a new window]   Fig. 9.   COX-2 protein expression in MTAL tubules. Lysates from unstimulated MTAL tubules freshly isolated from four individual rats were separated on a 10% SDS-PAGE gel. Proteins were transferred to a nitrocellulose membrane and probed with anti-COX-2 antisera. Molecular mass of COX-2 (72/74 kDa) is indicated.

View larger version (32K): [in this window] [in a new window]   Fig. 10.   Effects of Dex on COX-2 protein expression. A: MTAL cells were preincubated in absence or presence of Dex (2 µM) for 2 h, then challenged with PMA (1 µM) for 6 h. Analysis of protein was by Western blot and phosphorimaging. B: relative intensities of bands were determined by phosphorimaging and analysis with Imagequant software. * P  0.05 and ** P  0.01; n = 3.

COX-2-dependent inhibition of ⁸⁶Rb uptake. Previous work from our laboratory indicated that the ability of TNF to inhibit ouabain-sensitive ⁸⁶Rb uptake was prostanoid-dependent and required a latency period of more than 4 h (6). We determined whether this effect was dependent on COX-2-mediated prostanoid formation. As shown in Fig. 8, A and B, expression of COX-2 protein was not observed until ~6 h after challenge with TNF. Moreover, significant increases in PGE₂ levels were not observed until cells had been exposed to the cytokine for ~20 h. The functional considerations of these data were assessed by inhibiting COX-2 activity with NS-398 and comparing the ability of TNF to inhibit ⁸⁶Rb uptake in the absence or presence of selective COX-2 inhibition. Ouabain-sensitive ⁸⁶Rb uptake was inhibited by ~40% after addition of TNF for 24 h (Fig. 11). In contrast, TNF had no effect when cells were pretreated with NS-398 and then challenged with the cytokine (Fig. 11). These data suggest that regulation of COX-2 by TNF may subserve a regulatory mechanism in the MTAL that is expressed after a period of several h.

View larger version (13K): [in this window] [in a new window]   Fig. 11.   NS-398 prevents TNF-mediated inhibition of 86Rb uptake. MTAL cells were pretreated for 15 min with NS-398 (0.1 µM) and then incubated in absence or presence of TNF (1 nM) for 20 h. Uptake was initiated by adding 0.5-1.0 µg of 86Rb for 10 min at 37°C, and was terminated by addition of a stop solution, as described in METHODS. Cells were lysed, and the radioactivity was determined in a gamma counter. Ouabain-sensitive component of total 86Rb uptake was calculated by subtracting 86Rb uptake in presence of ouabain (1 mM) from uptake in absence of ouabain (6). Data represent the means ± SD of triplicate determinations from three similar experiments. * P  0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We demonstrated that MTAL cells in primary culture express COX-2 protein after challenge with either PMA or TNF. Posttranscriptional regulatory mechanisms contributed to mRNA accumulation, and pretreatment of cells with Dex attenuated expression of the protein in unstimulated cells. Production of PGE₂, in response to either PMA or TNF, was completely prevented by pretreatment with the COX-2-selective inhibitor, NS-398, only when COX-2 protein was expressed. The inhibitory effect of TNF on ouabain-sensitive ⁸⁶Rb uptake was linked to expression and activity of COX-2 protein and was consistent with the latency period previously described for the effects of this cytokine on ⁸⁶Rb uptake (6).

The amino acid sequences for COX-1 and COX-2 are similar (~75% homology), and the residues that are important for the catalytic activities of these enzymes are highly conserved (12, 22). However, COX-1 and COX-2 differ significantly at amino acids prior to residue 30, and COX-2 has an 18-amino acid insert close to the COOH terminus that is not present in COX-1 (37). These two enzymes catalyze the same reactions; i.e., AA is converted to PGG₂ via the COX reaction, followed by a peroxidase reaction in which the 15-hydroperoxyl group of PGG₂ is reduced to the 15-hydroxyl group of PGH₂. The latter is metabolized by specific isomerases to prostanoids in a cell-specific manner. Regulation of the two COX isoforms is quite different. COX-1 is constitutively expressed in several renal cell types, including interstitial cells, renal blood vessels, and several nephron segments. Our recent finding that MTAL epithelial cells can produce TNF, a cytokine shown to increase COX-2 expression, prompted us to determine whether this cytokine increased COX-2 expression in the MTAL. We suggested that the latency period observed for TNF-mediated inhibition of ⁸⁶Rb uptake in the MTAL was related to the time required for gene transcription and subsequent protein expression of this COX isoform. The possibility that the MTAL expressed COX-2 also was addressed by stimulating cultured MTAL cells for various times (2-20 h) with PMA, a known activator of COX-2 in several cell types. Both TNF and PMA increased COX-2 protein expression and PGE₂ production. Increased expression of COX-2 protein, concomitant with an increase in NS-398-sensitive PGE₂ production, was observed after challenge with PMA for 6 and 20 h. However, although TNF-mediated COX-2 protein expression was only evident at 6 h, significant increases in NS-398-sensitive PGE₂ production were not observed until ~10-20 h of exposure to the cytokine. This apparent discrepancy probably reflected a combination of several factors including differences in signaling pathways and the interposition of additional steps such as a nitric oxide-dependent step, that has been described (34). Thus the kinetics of COX-2 expression and the effects of TNF on ⁸⁶Rb uptake suggest that the time- and prostanoid-dependent components of these effects are likely related to induction of the COX-2 gene, posttranscriptional effects, and the subsequent increase of COX-2 activity, which accounted for the eventual increase in PGE₂ production.

In contrast to COX-1 knockout mice, COX-2 gene disruption in mice causes a severe nephropathy that results in death by the third month (29). COX-2 knockout mice present with small kidneys having few functional nephrons with immature glomeruli and marked impairment of nephrogenesis, which, in the rodent, normally continues for the first several postnatal weeks. Constitutive expression of COX-2 has been reported in the macula densa and in adjacent epithelial cells of the cortical TAL (14). The MTAL expresses low levels of COX-1 and also appears to express COX-2 protein constitutively in a subpopulation of cells (43), as well as in unstimulated freshly isolated MTAL tubules and primary cultures of MTAL cells. Thus the prostanoid-dependent effects of TNF on MTAL function may be mediated via this COX isoform. Moreover, local production of TNF by the MTAL is compatible with a role for this cytokine as an essential component of a regulatory mechanism that affects ion transport and operates in a microenvironment rather than systemically as occurs in endotoxic shock.

The MTAL metabolizes AA via a CyP450-dependent pathway to several products, including 20-HETE, which contribute importantly to MTAL function by regulating ion transport mechanisms (5, 36). Notwithstanding the relatively low levels of COX expression in the MTAL, this nephron segment also metabolizes AA via a COX-dependent pathway to PGE₂ and PGF₂ (11). Indeed, the levels of PGE₂ produced by various MTAL preparations have been reported to range from 10⁸ M (11, 35) to 10⁵ M (24). Certainly, many biochemical and functional effects mediated by PGE₂ can occur over this dose range, or less. For instance, PGE₂ inhibits oxygen consumption in rabbit MTAL at a concentration of 10⁷ M (24). It is well established that eicosanoids contribute importantly to the regulation of function along the nephron, as PGE₂, an important regulator of ion transport in the MTAL, inhibits the activity of the Na⁺-K⁺-ATPase (Na⁺ pump) and Na⁺-K⁺-2Cl cotransporter, two important transepithelial ion transport mechanisms present in this nephron segment (18, 40, 44). Moreover, the capacity of cytokines such as TNF and interleukin-1 (IL-1), to modulate function within the kidney may require interactions with AA metabolites (1, 2, 21, 46), as TNF and IL-1 stimulate prostaglandin synthesis by glomerular mesangial cells, MDCK cells, and papillary collecting duct cells (3, 20, 25). Conversely, TNF production is regulated by PGE₂ in several cell types (10, 23). Thus a regulatory feedback system involving cytokines and prostaglandins may determine the net effect exerted by these two classes of potent biological mediators.

COX-2 gene transcription is induced by mitogens, growth factors, cytokines, and tumor promoters and is glucocorticoid inhibitable, whereas expression of COX-1 mRNA is less responsive to these conditions (3, 31, 32). Accordingly, it is not surprising that COX-1 mRNA accumulation did not change in MTAL cells challenged with either TNF or PMA. Posttranscriptional mechanisms also are important in the sustained induction of COX-2 mRNA (33, 39). For instance, Srivastava et al. (39) demonstrated that IL-1 enhances the stability of COX-2 mRNA in the absence of any further transcription in rat mesangial cells. Superinduction of COX-2 mRNA in MTAL cells challenged with PMA or TNF after pretreatment with CHX is consistent with posttranscriptional regulation in these cells. Moreover, inhibition of gene transcription with AcD had little effect on mRNA accumulation, suggesting that, in the MTAL, a posttranscriptional mechanism(s) can have a major impact on mRNA accumulation even in the absence of ongoing transcription. A similar effect of AcD has recently been reported in human pulmonary A549 cells (30). As previous studies demonstrated that TNF can increase COX-2 by different signaling pathways within the same cell type (27, 28), it is likely that the signaling pathways used by TNF and PMA, although potentially convergent, are distinct. These differences in signaling pathways could contribute to molecular mechanisms that account for the differences in mRNA accumulation and PGE₂ production in response to PMA and TNF, reflecting changes in the rate of COX-2 gene transcription as well as differences in mRNA half-life.

Expression of COX-2 in the macula densa and TAL may be an essential component in renal mechanisms that affect salt and water excretion, as changes in dietary salt intake have been shown to differentially regulate renal COX-2 expression. Namely, a high-sodium diet increased COX-2 in the medulla, and a low-sodium diet increased COX-2 in the cortex (45). These findings are compatible with a COX-2-dependent mechanism that participates in the regulation of extracellular fluid volume. The findings in the present study also suggest an important link of a critical nephron segment, the MTAL, to the regulation of extracellular fluid volume via a COX-2-dependent mechanism evoked by activation of the renin-angiotensin system. Thus we have shown that ANG II stimulates TNF production by the MTAL, an effect that results in the expression of COX-2 in this nephron segment.