Peroxisome Proliferator-activated Receptor (cid:1) Activation Promotes Cell Survival following Hypertonic Stress*

COX2-selective non-steroidal anti-inflammatory drugs (NSAIDs) cause selective apoptosis of renal medullary interstitial cells (RMIC) in vivo and reduce their ability to tolerate hypertonic stress in vitro . To determine the mechanism by which COX2 activity promotes RMIC viability, we examined the capacity of COX2-derived prostanoids to promote RMIC survival. Although RMICs syn-thesize prostaglandin E2 (PGE2) >> PGI2 > PGF2a > TxA2, only PGI2 enhanced RMIC viability following hypertonic stress. RMICs do not express the prostacyclin receptor, but they do express the prostacyclin responsive nuclear transcription factor peroxisome proliferator-acti-vated receptor (cid:1) (PPAR (cid:1) ). Hypertonic stress increased PGI2 synthesis 330% above base line and also activated a PPAR (cid:1) specific reporter (delta response element (DRE)) by 90% above base line. Conversely DRE activity was only inhibited by the COX2-selective inhibitor SC236 but not by a COX1-selective NSAID (SC560). Overexpression of PPAR (cid:1) using an adenovirus not only drove DRE activity but also prevented RMIC death due to COX2 inhibition. These studies are consistent with a model whereby hypertonicity activates COX2-derived prostaglandin production, which promotes RMIC viability through PPAR (cid:1) . Inhibition of

Cells residing in the renal medulla are uniquely subjected to recurrent hypertonic stress with ambient tonicity rapidly shifting between 300 and 3000 mosmol. Renal medullary interstitial cells (RMICs) 1 are a distinct cell type, physically spanning the gap between the vasa recta and medullary tubules. These cells are distinguished by the presence of abundant lipid droplets comprised of long chain unsaturated fatty acids including arachidonate (1) and have also recently been found to be the major site of renal COX2 expression (2). COX2 is an important factor promoting survival of RMICs following hypertonic stress (3,4). COX2-selective inhibitors cause RMICs to die when exposed to otherwise non-lethal hypertonic stress (4). These results may account for the finding that RMICs are an early site of injury caused by COX-inhibiting analgesics and nonsteroidal anti-inflammatory drugs (NSAIDs) (5,6).
The downstream mechanisms through which COX2 activity promotes RMIC survival remain uncharacterized. COX2 mediates the conversion of arachidonate to five major bioactive prostanoids: PGE2, PGD2, PGF2␣, PGI2, and TxA2. These compounds exert a variety of cellular effects via a family of G-protein-coupled receptors that not only exhibit ligand selectivity but also differential signaling (7,8). For example, PGI2 selectively activates a G s -coupled cAMP-stimulating IP receptor, whereas TxA2 activates a G q -coupled thromboxane receptor (TP receptor) (7). PGE2 interacts with at least four receptor subtypes designated EP1, EP2, EP3, and EP4 that couple through G q , G s , G i , and G s , respectively (7,8). Other studies have suggested that biological effects of prostanoids could be mediated via nuclear receptors including the peroxisome proliferator-activated receptors ␣, ␦(␤), or ␥. The goal of the present studies was to identify the COX2-derived products and their downstream targets promoting cell survival following hypertonic stress.

MATERIALS AND METHODS
Cell Culture-Rabbit medullary interstitial cells were cultured as described previously. Briefly, female New Zealand White rabbits were anesthetized (44 mg/kg ketamine and 10 mg/kg xylazine, intramuscularly). The left kidney was removed, and the medulla was dissected and minced with a razor blade under sterile conditions in 5 ml of sterile RPMI 1640 plus 10% (v/v) fetal bovine serum (Hyclone, Logan, UT). This homogenate was injected subcutaneously in the abdominal wall using a 14-gauge needle. Twenty days postsurgery, subcutaneous nodules appeared. The rabbits were re-anesthetized and sacrificed by decapitation, and the nodules were removed under sterile conditions. Nodules were minced into 1-mm fragments and explanted in 75-cm 2 tissue culture plates. Cells were cultured in RPMI 1640 tissue culture medium supplemented with 10% (v/v) fetal bovine serum and streptomycin and penicillin. Cultures were incubated at 37°C in 95% O 2 /5% CO 2 . Tissue culture medium was changed every 48 -72 h. Cells were studied in their third to fourth passages. These cells exhibit characteristic abundant oil red O positive lipid droplets, which typifies type I RMICs (9).
Prostaglandin Measurement-PG production by cultured RMICs was quantitated utilizing gas chromatography/negative ion chemical ionization mass spectrometric assays as described previously (10). Cultured RMICs were treated with phorbol ester (10 nM) for 6 h, culture medium was replaced with fresh medium, and 2 h later culture medium was collected for prostanoid determination.
Immunoblotting-PPAR␦ immunoblots were performed on whole cell lysates from cultured RMICs. Protein concentration was determined using bicinchoninic acid protein assay (Sigma). Twenty micrograms of protein extract was loaded in each lane of a 10% SDS-PAGE mini-gel and run at 120 V. Protein was transferred to a nitrocellulose membrane at 22 V overnight at 4°C. The membrane was washed three times with § Recipient of a Veterans Affairs career development award. 1 The abbreviations used are: RMIC, renal medullary interstitial cell; NSAID, non-steroidal anti-inflammatory drug; PG, prostaglandin; IP receptor, prostacyclin receptor; PPAR, peroxisome proliferator-activated receptor; RT, reverse transcription; Ad, adenovirus; GFP, green fluorescent protein; DRE, delta response element; TK, thymidine kinase; cPGI2, carbaprostacyclin.
TBST (50 mM Tris, pH 7.5, 150 mM NaCl, 0.05% Tween 20) and then incubated in blocking buffer (150 mM NaCl, 50 mM Tris, 0.05% Tween 20, and 5% Carnation nonfat dry milk, pH 7.5) for 1 h at room temperature. The membrane was then incubated with an anti-human PPAR␦ antibody (1:300, sc7197, Santa Cruz Biotechnology Inc., Santa Cruz, CA) in blocking buffer overnight at 4°C. Following washing (ϫ3), the membrane was incubated with a horseradish peroxidase-conjugated secondary antibody (1:20,000, Jackson Immuno-Research Lab) for 1 h at room temperature followed by three 15-min washings. Antibody labeling was visualized by the addition of chemiluminescence reagent (Renaissance, PerkinElmer Life Sciences), and the membrane was exposed to Kodak XAR-5 film.
RT-PCR Detection of IP Receptor Expression-Total RNA was extracted from cultured RMICs or rabbit aorta using Tri-zol reagent (Invitrogen). RT was performed on 1 g of total RNA (PerkinElmer Life Sciences). A pair of degenerate primers was designed according to human and rat IP receptor sequences: CTGGTGCTTCMTCCGCAT (forward) and GGAAGACAGCCTTKCGGAAA (reverse). A 398-bp PCR product was amplified from rabbit aorta RNA under the following conditions: 95°C for 5 min; 35 cycles of 95°C for 45 s, 58°C for 45 s, and 72°C for 45 s; 72°C for 7 min; 4°C. A 700-bp rabbit glyceraldehyde-3phosphate dehydrogenase PCR product was obtained from the same sample used as RNA loading control.
Ad-PPAR␦ and Ad-GL-Adenoviral vectors encoding a human PPAR␦ (Ad-PPAR␦) were a gift of Dr. Kenneth Kinzler (11). GFP adenovirus was made as previously described (4). A full-length green fluorescent protein cDNA (Green Lantern or GL, Invitrogen) was subcloned into pACCMV. The pACCMV shuttle plasmid contains the cytomegalovirus immediate early enhancer and promoter and the SV40 polyadenylation sequence. The resulting shuttle plasmids were cotransfected into HEK 293 cells along with the pJM17 vector by Super-Fect (Qiagen). Ad-GFP was generated by a homologous recombination event, resulting in plaque formation in the HEK 293 cells. The resulting infectious adenovirus was plaque-purified. For infection of RMICs, 200 l of virus (multiplicity of infection, 100) was added to each culture dish. After a 2-h incubation, the virus was removed and fresh Dulbecco's modified Eagle's medium with 10% fetal bovine serum was added. Experiments were carried out 48 -72 h after infection.
PPAR␦ Reporter Studies-A PPAR␦ reporter (delta response element, DRE) was generously provided by Dr. Kenneth Kinzler (11). The reporter contains PPAR␦ and retinoid X receptor recognition motifs that were identified by an in vitro site selection approach (11). Firefly luciferase PPAR␦ reporter plasmid and a plasmid containing Renilla luciferase driven by TK promoter were transfected into cells with using SuperFect (Qiagen). Cells were lysed 48 h after transfection for luciferase activity measurement using the dual luciferase assay system (Promega Corp., Madison, WI).
Cell Viability Analysis-Cell viability was assessed using crystal violet (12). Following treatment with NSAIDs or hypertonicity for predetermined time periods, culture medium was removed and plates were washed with phosphate-buffered saline. The remaining viable attached cells were stained with 0.5% crystal violet in 50% methanol for 15 min. The plates were then gently rinsed with water and dried. A solution containing 0.1 M citrate sodium, pH 5.4, and 20% methanol was added, and 30 min later, the absorbance at 570 nm was read using a spectrophotometer. The percentage of cell survival was defined as the relative absorbance of treated versus untreated cells. The unattached cells in the medium were confirmed to be dead by trypan blue exclusion assay.
Carbaprostacyclin (cPGI2) Increased RMIC Survival following Hypertonic Stress-NaCl was added to abruptly increase tonicity from 300 to 500, 550, or 600 mosmol. RMICs tolerated a shift of 250 mosmol or less, but when osmolality was increased to 600 mosmol, 80 Ϯ 12% of RMICs died (Fig. 2A). The ability of exogenous PGs to promote RMIC survival following exposure to lethal hypertonic stress was examined. Pretreatment of RMICs with PGE2 or the thromboxane A2 agonist U46619 did not affect cell viability following exposure to 600 mosmol. In contrast, PGD2 and PGA2 decreased cell viability following hypertonic stress compared with vehicle-treated cells. Importantly the prostacyclin analog cPGI2 (20 M) significantly increased RMIC survival almost 3-fold following hypertonic stress (20.0 Ϯ 3.4% versus 55.3 Ϯ 7.8%; n ϭ 3; p Ͻ 0.01) (Fig. 3).  RMICs were cultured to confluence and shifted to media of the indicated osmolalities (produced by adding NaCl). 48 h later, cell viability (A) was assessed using a crystal violet assay as described under "Materials and Methods." Medium PGI2 metabolite 6-keto PGF1␣ was measured via gas chromatography/mass spectrometry (B). **, p Ͻ 0.01 versus cells exposed to 300 mosmol (n ϭ 4).

Expression of Prostacyclin Receptors in RIMCs-Potential
targets of prostacyclin action on RMIC were identified by RT-PCR using primers specific for the cell surface G-protein-coupled IP receptor and nuclear transcription factor PPAR␦, and by immunoblot using antibody specific for the PPAR␦ (13,14). Although RT-PCR demonstrated abundant expression of IP receptor in aorta, IP receptor mRNA was absent from cultured RMICs (Fig. 4). In contrast, PPAR␦ mRNA and protein were detected in RMIC by RT-PCR and immunoblot. Immunoreactive protein migrated at the same speed on SDS gel as recombinant PPAR␦ protein transduced into RMICs via adenoviral vector (Ad-PPAR␦) (Fig. 4).
Hypertonicity Increased PPAR␦ Activity through Increased PGI2 Release-To examine whether PPAR␦ is involved in RMIC survival-promoting action of COX2/PGI2 following hy-pertonic stress, the effect of tonicity on endogenous PPAR␦ activity was examined using a PPAR␦ reporter, i.e. a DRE luciferase reporter construct. Fig. 5 shows hypertonicity nearly doubled PPAR␦ transcriptional activity when RMICs were switched to a non-lethal hypertonicity of 550 mosmol (1.9 Ϯ 0.09-versus 1.0 Ϯ 0.02-fold; n ϭ 3; p Ͻ 0.01). Hypertonicity also increased PGI2 production (Fig. 2B), and the exogenous PGI2 analog cPGI2 activated PPAR␦ reporter activity by 2.7-fold at a concentration of 20 M (Fig. 5), suggesting a link between increased PGI2 production and PPAR␦ activation during hypertonic stress. Interestingly, the concentration at which PGI2 significantly activated PPAR␦ is identical to the concentration at which cPGI2 prevented RMIC death in otherwise lethal hypertonic culture conditions (Fig. 3).
PPAR␦ Activation Restores Osmotic Tolerance of COX2-inhibited RMICs-When subjected to osmotic stress, RMICs are markedly sensitized to the lethal effects of COX2 inhibition. To further characterize the role of PPAR␦ in promoting RMIC survival, an adenoviral vector was used to overexpress PPAR␦. Following hypertonic stress, low concentrations of SC236 killed RMICs. This lethal effect of COX2 inhibitor was blocked by overexpression of PPAR␦ (Fig. 7). DISCUSSION Medullary interstitial cells are a major source of renal prostaglandin synthesis and play an important role in modulating renal medullary blood flow and salt absorption (9). COX is a key enzyme converting arachidonic acid to PGH2, which is then converted to different bioactive prostaglandins through differ-ent synthases (7,15). Although COX1 is constitutively expressed in many tissues including the collecting duct, COX2 expression is generally induced by growth factors, cytokines, or osmotic stress (4,16,17). Accumulating evidence suggests COX2 is selectively expressed in renal medullary interstitial cells where it may play a role in promoting cell survival (2)(3)(4). The present studies demonstrate that COX2-derived prostacyclin may be a key component allowing RMIC to survive hypertonic conditions present in the renal medulla.
Cultured RMICs produce several prostaglandins including PGE2 Ͼ Ͼ PGI2 Ͼ PGF2␣ Ͼ TxA2. However, only carbaprostacyclin, an analog of PGI2, has the ability to increase the tolerance of RIMC to hypertonic stress that also increases prostacyclin synthesis. This result is consistent with early reports that show the protective effect of prostacyclin in other organ injury including the heart (18) and brain (19,20). The bioactivity of PGI2 is generally via activation of a cell surface G-protein-coupled IP receptor that activates G s and increases cAMP levels or via activation of a nuclear transcription factor PPAR␦ and downstream PPAR␦-responsive gene transcription (13,21). The present study examined the involvement of these mechanisms in PGI2's protective effect against hypertonic stress. Only PPAR␦ but not the IP receptor was detected in cultured RMICs. When RMICs were switched from isotonic conditions to a hypertonic medium, PPAR␦ transcriptional activity was markedly increased. Activation of PPAR␦ by hypertonicity appears to be mediated through increased PGI2 production following hypertonic stress, based on the observations that hypertonicity also increased prostacyclin synthesis and exogenous cPGI2 directly activated PPAR␦ specific reporter in cultured RMICs.
PPAR␦ is a nuclear receptor, functioning as ligand-dependent sequence-specific activator of transcription (22). In the presence of ligand, PPAR␦ heterodimerizes with a retinoid X receptor. This heterodimer then modulates transcription by binding to a PPAR␦ response element or DRE in target genes. The array of specific target genes that are transcriptionally regulated by PPAR␦ have not yet been identified. There are several reports suggesting roles for PPAR␦ in adipocyte differentiation, brain function, epidermal differentiation, embryo implantation, colon cancer, etc. (11,13,(23)(24)(25)(26)(27)(28). The present FIG. 6. Effect of selective COX2 inhibition on PPAR␦ reporter activity and cell viability of cultured RMICs. A, DRE activity. RMICs were cultured to confluence and co-transfected with DRE-driven firefly luciferase vector and TK-driven Renilla luciferase vector. 12 h later the cells were treated with COX2-selective (SC236) or COX1selective (SC560) inhibitor for 36 h, followed by determination of luciferase activity as described under "Materials and Methods." To examine the effect of PPAR␦ overexpression on PPAR␦ reporter activity, cultured RMICs were transduced with Ad-PPAR␦. 24 h later the cells were co-transfected with DRE-driven firefly luciferase vector and TK-driven Renilla luciferase vector; 12 h later the cells were exposed to COX2 inhibitor (SC236) for 36 h as in other groups. B, cell viability. RMICs were cultured to confluence and treated with the COX2 (SC236) or COX1 (SC560) selective inhibitor. Cell viability was determined 48 h later using crystal violet assay. To examine the effect of PPAR␦ overexpression on cell viability, cultured RMICs were transduced with Ad-PPAR␦. The cells were then treated with COX1 and COX2 inhibitors for 48 h and cell viability was assessed. **, p Ͻ 0.01 versus vehicletreated cells; ##, p Ͻ 0.01 versus 20 M SC236-treated cells, n ϭ 6. studies suggest that PPAR␦ activation is also involved in the physiological response of RMICs to environmental osmotic stress.
That COX2-mediated prostacyclin synthesis activates PPAR␦ in RMICs is further supported by studies showing COX-inhibiting NSAIDs blocked PPAR␦ reporter activity. Importantly, the inhibitory effect of NSAIDs on PPAR␦ activity in RMICs is limited to the COX2-selective inhibitor, while a structurally similar COX1-selective inhibitor increased PPAR␦ activity rather than inhibiting its activity. This is consistent with the fact that COX2 is the predominant isoform in RMICs (3), supporting the hypothesis that the effect of NSAIDs on PPAR␦ is mediated through COX inhibition. Additionally, because some NSAIDs can bind to PPARs (11), it is also conceivable that the COX2 inhibitors directly prevent PPAR␦ activation by competing with PGI2 activation of PPAR␦. Importantly, the ability of COX inhibitor to block PPAR␦ activity correlated with its ability to kill RMICs; the COX2 selective inhibitor, which blocked PPAR␦ activity, killed RMICs, whereas the COX1 inhibitor, which did not inhibit PPAR␦, did not kill RMICs. PPAR␦ overexpression prevented RMIC from COX2 inhibitorinduced cell death and was accompanied by restoration of PPAR␦ activity. These studies suggest that COX2-mediated PPAR␦ activity is a survival factor in RMICs. During hypertonic stress, the importance of COX2-derived products in maintaining RMIC survival is supported by studies that show RMICs treated with a sublethal dose of COX2 inhibitor will die following a mild hypertonic stress that will not otherwise kill RMICs with intact COX2 activity (4). Importantly, this reduced tolerance of hypertonic stress in RMICs treated with a COX2 inhibitor can be restored by activating PPAR␦ via PPAR␦ adenovirus. These studies further support the hypothesis that the survival-promoting action of COX2 activity during hypertonic stress relies on PPAR␦ activity. The present studies are consistent with an earlier report using colon cancer epithelial cells (11) and suggest that the survival-promoting ability of PPAR␦ is not only involved in the pathogenesis of tumor formation but also contributes to the physiological adaptation of RMICs to the hyperosmotic environment.
In summary, hypertonicity can induce COX2 expression in renal medullary interstitial cells. Increased COX2 expression is associated with increased PGI2 production and subsequent PPAR␦ activation, transcriptional activity, and increased cell survival. NSAIDs, which inhibit COX2-mediated prostacyclin release, may damage renal interstitial cells by inhibiting PPAR␦ activity. This could contribute to the papillotoxic effect of these drugs.