The PPARδ Ligand GW501516 Reduces Growth but Not Apoptosis in Mouse Inner Medullary Collecting Duct Cells

The collecting duct (CD) expresses considerable amounts of PPARδ. While its role is unknown in the CD, in other renal cells it has been shown to regulate both growth and apoptosis. We thus hypothesized that PPARδ reduces apoptotic responses and stimulates cell growth in the mouse CD, and examined the effect of GW501516, a synthetic PPARδ ligand, on these responses in mouse IMCD-K2 cells. High doses of GW501516 decreased both DNA and protein synthesis in these cells by 80%, but had no overall effect on cell viability. Although anisomycin treatment resulted in an increase of caspase-3 levels of about 2.59-fold of control, GW501516 did not affect anisomycin-induced changes in active caspase-3 levels. These results show that a PPARδ ligand inhibits growth but does not affect anisomycin-apoptosis in a mouse IMCD cell line. This could have therapeutic implications for renal diseases associated with increased CD growth responses.


Introduction
Traditional prostaglandin (PG) signaling in renal epithelial cells occurs through activation of prostanoid receptors [1][2][3]. In recent years, however, a novel signaling pathway for PGs has been found in which ligand-activated transcription factors called peroxisome proliferator-activated receptors (PPARs) [4] are activated at the nuclear membrane. One PPAR type, PPARδ, also known as PPARβ, is ubiquitously expressed in almost every tissue examined [5]. Both prostanoids PGI 2 and PGE 2 have been known to activate PPARδ [6,7]. The main role of PPARδ is in cell survival and this has been well characterized in colon cancer [8][9][10].
In the kidney, PPARδ is found in every part of the nephron with high levels in the glomerulus, cortical collecting duct, and inner medullary collecting duct (IMCD) and low expression in the outer medullary collecting duct. This PPAR has been implicated in the renal complications of metabolic syndrome [11], but most studies describe a link between PPARδ and survival of renal cells. In a study by Hao et al. [6], PPARδ was shown to increase renal medullary interstitial cell survival during high levels of hypertonic stress, in response to PGI 2 . PPARδ activation also protected kidneys from renal failure after renal ischemia/reperfusion experiments via the antiapoptotic Akt pathway [12]. Although, most of the studies have suggested an antiapoptotic role of PPARδ, a few papers have shown that PPARδ can increase apoptosis [13]. However, PPARδ is recognized as an important protein in the survival of many renal cell types.
PPARδ has also been shown to promote growth responses in a variety of cells. PGE 2 , working through PPARδ, can stimulate proliferation of stem cells [14] and overexpression of PPARδ has been shown to reverse inhibitory growth signals in a prostate epithelial cell line [15]. Furthermore, upregulation of PPARδ in smooth muscle cells promotes cell cycle progression [16]. The effect of PPARδ activation on cell growth in the kidney, however, is unknown.
The collecting duct (CD), specifically the IMCD, survives in a harsh environment due to the hypertonic conditions in the interstitium needed to concentrate urine [17,18]. In most cells, apoptosis would occur at the levels of stress at which they reside, but the CD manages to survive and resist any apoptotic activity [19]. It is known that cells react to high levels of tonicity by increasing the transport of several osmolytes into the cells aiding in their survival [20]. However, in cells where there is chronic hypertonicity, like the IMCD, the accumulation of these osmolytes may not be sufficient for their continued survival [21]. Cultured IMCD cells have been known to increase levels of cyclooxygenase-2 (COX-2) and stimulate PGE 2 release up to 33-fold of control in response to hypertonic stress [22]. It is likely then that PGs may contribute to the inherent CD resistance to stress by targeting PPARδ to promote cell survival.
It would appear in most cell lines that PPARδ has opposing growth and apoptotic effects; however, its role in the CD has not been determined. Our study examines the role of PPARδ activation in the survival and growth of IMCD-K2 cells. The mouse IMCD-K2 cell line is derived from the initial portion of the IMCD from a mouse that is transgenic for Simian Virus 40 (SV40) and was shown to retain many characteristics of the intact IMCD, including amiloride-sensitive sodium absorption stimulated by aldosterone [23]. Therefore, we propose that PPARδ activation will regulate growth and apoptotic responses in the mouse IMCD-K2 cells.  from the GeneAmp RNA PCR Core Kit (Applied Biosystems, Foster City, Calif, USA) as per the manufacturer's instructions. Next, the RNA was amplified by polymerase chain reaction (PCR), using AmpliTaq DNA polymerase and specific primers for the prostanoid receptors (i.e., EP [1][2][3][4] and IP) and PPARδ. The primers were obtained from the University of Ottawa Biotechnology Research Institute (Ottawa, ON, Canada). The PCR parameters used were 40 cycles of: 95 • C for 30 seconds, 59 • C for 45 seconds, and 72 • C for 60 seconds, followed by 600 seconds at 72 • C.
The amplification was carried out using the GeneAMP PCR System 2400 (Applied Biosystems). The PCR products were run on an agarose gel and the bands were visualized using ethidium bromide under UV light. Fluorescence intensity (fold control)

GW/ A1000
Cis * * * § * § Figure 8: GW501516 has no effect on anisomycin-induced cell death. Serum-starved cells were stimulated with serum-free media (control), 10 −6 M GW501516 (GW), 100 and 1000 ng/mL anisomycin (A100 and A1000, resp.), or 50 ng/mL cisplatin (Cis) for 18 hours. Some of the cells treated with anisomycin were pretreated with 10 −6 GW501516 for 24 hours and cotreated with the agonist and anisomycin for 18 hours (GW/A100 and GW/A1000). The cells were then incubated for one hour with the DNA dyebinding solution from the CyQUANT NF assay kit (Invitrogen). Fluorescence intensity, representing number of cells, was then measured as per the manufacturer's instructions. Values are means ± S.E.M.; n = 3. * P < .05 compared to control; §P < .05 compared to GW.
Canada) and the percent stimulation of cAMP levels over control levels of each sample was calculated. 8 μM EDTA, 0.2 mM sodium phosphate (pH 7.2), and fresh 0.5 mM PMSF, 1 : 100 protease inhibitor cocktail (Sigma), 1 mM sodium pyrophosphate, 10 mM sodium fluoride, and 100 μM sodium orthovanadate. The protein was quantified using the Bradford reagent method (Bio-Rad, Hercules, Calif, USA). The lysates were then resolved by SDS-PAGE on a polyacrylamide gel using a Mini-PROTEAN II apparatus (Bio-Rad) and transferred onto nitrocellulose membranes (Amersham, Amersham, UK). After blocking for 90 minutes in 10% milk/TBS-T (137 mM NaCl, 20 mM Tris base, 0.1% Tween20), the membranes were incubated overnight with the corresponding primary antibody. The membranes were then incubated with their respective secondary antibodies (i.e., anti-rabbit, anti-mouse, and anti-donkey) for 90 minutes and then washed for 60-90 minutes in TBS-T. The bands were then visualized by ECL. Detection of β-actin was used to normalize samples and densitometric analysis of the bands was performed using the Kodak 1D Image Analysis software (Eastman Kodak Company, Rochester, NY, USA).

Real-Time PCR.
To determine the effect of PGE 2 on PPARδ levels in IMCD-K2 cells, RNA was isolated and DNase-treated as described above. The cells were treated for 2, 8, or 24 hours with serum-free media as control, 10 −5 M indomethacin, or 10 −6 M PGE 2 . A fifteen-minute pretreatment with indomethacin was performed prior to stimulations with PGE 2 and forskolin. The mRNA levels of PPARδ were ascertained by real-time PCR using TaqMan One-Step RT-PCR master mix reagents (Applied Biosystems) and an ABI Prism 7000 sequence detection system. Reactions were carried out by using 50 ng of total IMCD-K2 RNA under the following conditions: 48 • C for 30 minutes, and 95 • C for 10 minutes, and 40 cycles of 95 • C for 15 seconds and 60 • C for 1 minute. The probe and primers used for mouse PPARδ were forward primer, 5 -GAGCCCAAGTTCGAGTTTGC-3 , reverse primer, 5 -TGAAGAGCGCCAGGTCACT-3 , and probe, FAM-AGTTCAATGCGCTGGAGCTCGATGA-TAM (Sigma Genosys, Oakville, ON, Canada). All values were normalized to GAPDH mRNA levels in the same sample, which was determined by the TaqMan Rodent GAPDH control reagent kit (AppliedBiosystems).

3 H-Thymidine Incorporation.
To study the effect of the PPAR activation on cell proliferation, DNA synthesis was measured using the incorporation of 3 H-thymidine. IMCD-K2 cells were cultured in 24-well plates, grown to ∼50% confluence, and then starved with serum-free media. Afterwards the cells were treated for 24 hours with vehicletreated control (DMSO), or GW501516 (10 −8 , 10 −7 , 10 −6 , 10 −5 M). GW501516 is among the more specific and most commonly used synthetic PPARδ ligands [24]. 3  (Amersham, 0.5 μCi/mL) was added during the final four hours of stimulation. The plates were then washed four times in ice-cold PBS. Next, the cells were permeabilized in 500 μL of 1N NaOH at 37 • C for ∼30 minutes, and the amount of 3 H-thymidine in counts per minute (cpm) was measured using a scintillation counter. Samples were done in triplicate and thymidine incorporation is expressed as fold control.

3 H-Leucine Incorporation.
To study the effect of the PPAR ligands on cell growth, protein synthesis was measured using the incorporation of 3 H-leucine. IMCD-K2 cells were cultured in 24-well plates, grown to ∼50% confluence, and then starved with serum-free media for 24 hours. Next the cells were stimulated for 24 hours with vehicle-treated control (DMSO) and GW501516 (10 −8 , 10 −7 , 10 −6 , and 10 −5 M). 3 H-leucine (Amersham, Perkin Elmer, Waltham, Mass, USA, 0.5 μCi/mL) was added to each well during the 24-hour stimulation. The cells were then prepared as described above and the amount of 3 H-leucine in counts per minute was measured using a scintillation counter. Samples were done in triplicate and leucine incorporation is expressed as fold control.

Cell Viability.
To evaluate the effect of GW501516 on the viability of IMCD-K2 cells in both the presence and absence of a death response, a cell viability assay was performed. IMCD-K2 cells were grown to 70% confluence in 96-well plates and stimulated for 18 hours with control (1 μL of ethanol and 1 μL DMSO in DMEM/F-12), GW501516 (10 −8 , 10 −7 , 5 × 10 −7 , 10 −6 , 5 × 10 −6 , and 10 −5 M), anisomycin (Sigma; 100, 1000 ng/mL), or cisplatin (Sigma; 50 ng/mL). Some cells were treated with both anisomycin and GW501516 (10 −6 M) following a 24-hour pretreatment with GW501516 (10 −6 M). Anisomycin and cisplatin were used to induce cell death. Following the stimulation, the cells were incubated in 100 μL of a 1x DNA dye-binding solution from the CyQAUNT NF kit (Invitrogen). The fluorescence intensity, representing the number of cells, was measured in a FLUOstar Galaxy plate reader (BMG Labtechnologies, Durham, NC, USA). Samples were done in triplicate and fluorescence intensity is expressed as fold control.

Colorimetric Caspase-3 Activity Assay.
To determine if GW501516 is involved in apoptosis, a caspase-3 activity assay was carried out using the colorimetric CaspACE Assay System kit (Promega, Madison, Wis, USA), as per manufacturer's instructions. IMCD-K2 cells were grown to ∼50% confluence in 100 mm plates and starved for 24 hours. Next the cells were treated with control (1 μL of ethanol and 1 μL of DMSO in 1 mL of DMEM/F-12), 10% FBS, GW501516 (10 −6 M), anisomycin (250 ng/mL), or Z-VAD-FMK (2 × 10 −5 M), a pan-caspase inhibitor supplied in the kit. Some cells were treated with both anisomycin and GW501516 following a 24-hour pretreatment with GW501516. Also, some plates were treated with anisomycin and Z-VAD-FMK following a 24-hour pretreatment with the inhibitor.
Following stimulations, the cells were resuspended in 100-200 μL of the caspase lysis buffer subjected to two freezethaw cycles (at −20 • C), and a 15-minute incubation on ice. The lysates were then centrifuged for 20 minutes at 13 100 revolutions per minute and protein was quantified using the Bradford method. 50 μg of each sample was incubated with the colorimetric caspase-3 substrate Ac-DEVD-pNA (2 × 10 −4 M) in a 96-well plate at room temperature overnight. The endogenous active caspase-3 bound to the substrate was determined by the absorbance at 405 nm (FLUOstar Galaxy). Samples were done in duplicate and the caspase activity is expressed as relative absorbance.

M). Either an unpaired t-test or a one-sample t-test
(with a hypothetical value of 1) was used to evaluate the statistical significance between data points. A P-value < .05 was considered statistically significant.

EP Receptor Subtypes and PPARδ
Expression. PGE 2 is known to activate prostanoid receptors, EP 1 , EP 2 , EP 3 , and EP 4 as well as PPARδ. Thus, the presence of these receptors in IMCD-K2 cells was determined. PCR was performed and bands corresponding to the prostanoid and PPARδ receptors were visualized as shown in Figures 1(a) and 1(b), respectively. Previous studies in our laboratory demonstrated that at higher stringency the band in the EP 2 lane disappears and that the product detected with the IP primers, present at approximately 200 base pair (bp) greater than the predicted size of 407 bp, does not correspond to the IP receptor, as determined by cloning and sequencing. The band seen between 700 and 800 bp in the EP 1 lane represents protein kinase N (PKN), as the genes for PKN and EP 1 overlap [25]. The multiple bands seen in the EP 3 are most likely due to alternative splicing of the EP 3 gene [26]. The products were not present when the reverse transcriptase was omitted (negative control; data not shown). These results indicate that IMCD-K2 cells express EP 1 , EP 3 , EP 4 , and PPARδ.

Effect of PGE 2 on cAMP
Stimulation. Both the EP 3 and EP 4 receptors elicit changes in intracellular cAMP; therefore, the effect of different prostanoids on cAMP stimulation was examined. Forskolin was used as a positive control. As shown in Figure 2(a), prostacyclin (PGI 2 ) analogs cicaprost (CCP) and iloprost (ILP) did not alter cAMP levels. PGE 2 , however, did cause a significant increase in percent stimulation, at about 70% above control, comparable to forskolin. As shown in Figure 2(b), treatment with increasing amounts of PGE 2 resulted in a concentration-dependent increase in cAMP, from 14.0 to 85.5%.

The Effect of PGE 2 on PPARδ Expression.
Since PGE 2 has been known to activate PPARδ to regulate several growth processes [7,14], it was ascertained that if PGE 2 stimulation could affect PPARδ expression in IMCD-K2 cells. Indomethacin, a COX inhibitor, was used to inhibit endogenous PG synthesis. As shown in Figures 3(a) and 3(b), treatment with PGE 2 , in the presence of indomethacin, for 8 hours resulted in a significant increase, of ∼2-fold, in PPARδ protein expression compared to control or indomethacin alone, as measured by Western Blotting. Stimulation for 24 hours also resulted in increased PPARδ protein expression, but this increase was not significant when compared with indomethacin alone. A representative blot for 24 hours is shown in Figure 3(a). However, two-hour stimulation did not alter the protein levels of PPARδ. We also examined the effect of PGE 2 on PPARδ mRNA expression by realtime PCR. As shown in Figure 4, PPARδ mRNA remained unchanged with any treatment group or exposure time. Forskolin had no effect on expression either (data not shown) indicating that cAMP does not alter PPARδ protein or RNA expression.

The Effect of GW501516 on Cell Proliferation/Growth.
PPARδ has been known to affect cell proliferation; therefore, we studied the effect of the PPARδ agonist, GW501515, on the proliferation/growth of IMCD-K2 cells. As shown in Figure 5, 10 −5 M GW501516 reduced 3 H-thymidine incorporation to about 0.05-fold control. PPARδ also regulates cell growth and as shown in Figure 6, 10 −5 M GW501516 produced a significant reduction in 3 H-leucine incorporation to approximately 0.19 of control. At lower concentrations of GW501516, neither DNA nor protein synthesis was altered. 3.6. The Effect of GW501516 on Caspase Activity. Besides having effects in cell growth, PPARδ has also been linked to changes in apoptosis. Therefore, the antiapoptotic effect of GW501516 was evaluated in IMCD-K2 cells. Caspase-3 activity was examined because it is a downstream effector protease that is common to all apoptotic pathways. As shown in Figure 9, anisomycin treatment resulted in an increase of caspase-3 levels of about 2.59-fold of control. Pre/cotreatment with GW501516 showed a decrease that was 0.79-fold (10 −6 M) and 0.81-fold (10 −8 M) of anisomycin alone. GW501516 alone showed no considerable change in caspase activity. The caspase inhibitor, Z-VAD completely abolished caspase activity in unstimulated cells, and in anisomycin-treated cells it decreased caspase activity to 0.06fold of anisomycin alone.
Activation of caspase was also measured by Western Blotting using a cleaved caspase-3 antibody. As shown in Figure 10(a), a 17 kDa band corresponding to cleaved caspase-3 was observed for the lysates that had been treated with anisomycin. The intensity of the bands in the lanes representing control and GW501516 alone was too low to be analyzed in the majority of the experiments; therefore, densitometric analysis was not performed for those samples. The levels of active caspase-3 (see Figure 10(b)) increased to 1.48 and 1.18 of anisomycin alone with both cotreatments (10 −6 and 10 −8 M GW501516, resp.) but the difference was not significant.

The Effect of GW501516 on Apoptosis.
In addition to looking at the effect of GW501516 on caspase activity, TUNEL assays were performed to further evaluate the apoptotic response. As shown in Figure 11, the proportion of TUNEL-positive cells increased while the total number of cells decreases with the addition of 250 ng/mL anisomycin when compared to control or the PPARδ agonist (at both concentrations). Pre/cotreatment of 10 −6 M or 10 −8 M GW501516 with anisomycin was comparable to anisomycin alone. 2 and PPARδ. It has been shown previously that prostaglandins, including PGE 2 , and PPARs influence one another's activity [27][28][29][30]. Although our studies showed that the PGE 2 receptors EP 1 , EP 3 , and EP 4 are present in the IMCD-K2 cells and PGE 2 produces an increase in cAMP production, consistent with the expression of EP 4 , PGE 2 had no effect on PPARδ mRNA expression. However, treatment with PGE 2 for eight hours did elicit an increase in PPARδ protein levels. There are two likely explanations for this. The first reason is that the PPARδ protein could be stabilized. A previous study in a prostate cancer cell line has shown that PGE 2 can stabilize hypoxia-inducible factor 1 alpha (HIF-1α) protein levels, without affecting mRNA levels [31]. This would most likely be due to PGE 2 preventing proteolysis of PPARδ. Interestingly, it has been revealed that unlike most nuclear receptors, degradation of PPARδ does not occur upon ligand binding but is in fact inhibited by it [32]. This would indicate that PGE 2 could be causing an increase in PPARδ activation. The second possible role of PGE 2 is that it could be enhancing the translation of basal PPARδ mRNA levels or stabilizing it. The ability of PGE 2 to stabilize mRNA (interleukin 8) has been previously described in a paper by Yu and Chadee [33]. Therefore, from the results reported here and the literature, the effect of PGE 2 on PPARδ is most likely a posttranscriptional event. We also observed an increase in PPARδ protein in response to 24 hours indomethacin. It is not clear at this time whether this effect is due to a direct increase in expression, but it has previously been shown that PPAR expression is altered by NSAIDs [34].

PPARδ Agonist and Cell Proliferation/Growth in IMCD-K2 Cells.
It has been previously reported that activation of PPARδ and the use of PPARδ ligands promote cell survival and proliferation. One recent study, using both PPARδ short interfering RNA (siRNA) and a PPAR antagonist, showed that downregulation or blocking the activation of PPARδ, respectively, inhibited the PGE 2 -induced proliferation of mouse embryonic stem cells [14]. Another study revealed that a PPARδ agonist can stimulate propagation in both human and mouse aortic endothelial cells [35]. However, the effects of PPARδ on renal cell growth responses have not been characterized. Here we show that a highly specific PPARδ ligand, GW501516, decreased DNA and protein synthesis in IMCD-K2 cells, but only at high concentrations. Although most studies indicate that PPARδ activation results in stimulation of cell growth, mostly in human and mouse colon cancer cells [7,36], there have been previous findings showing that PPARδ agonists, including GW501516, can hinder or have no effect on cell growth in certain human breast cancer cells and melanomas [37,38]. These differences may be due to species differences or diversity among cell lines.

PPARδ Agonist and Anisomycin-Induced Apoptosis in
IMCD-K2 Cells. Another finding in our study is that treatment with GW501516 does not reverse the reduction in cell survival due to exposure to anisomycin. These results are inconsistent with the literature that demonstrates that activation of PPARδ, or the use of a PPARδ agonist, protects many different cell types, including renal cells [12]. It is possible that the agonist may not actually be activating PPARδ in IMCD-K2 cells, as its efficacy in this cell line has never been tested. In addition, PPARδ activation may not be affecting cell survival if the death response is mostly through necrosis because studies only show a survival role for PPARδ with regard to apoptosis. To evaluate if the agonist is having an effect on PPARδ activation, experiments can be performed with cells transfected with a PPRE-driven reporter plasmid [39]. To completely assess death pathways activated by anisomycin, flow cytometry using the annexin V-FITC and propidium iodide stains should be used [40].
Even though the use of the agonist, in our experiments, elicits no response in overall cell survival in IMCD-K2 cells, the effect on apoptosis had yet to be investigated. When the CD is in a disease state due to urinary tract obstruction, ischemia-reperfusion injury or other insults apoptosis levels tend to increase [41,42]. Thus, we wished to examine if PPARδ could protect the CD from apoptosis. Using a variety of experiments assessing apoptosis (caspase activity assay, Western blotting, TUNEL), we found no significant change in the levels of apoptosis in cells treated with GW501516 compared to those stimulated with just anisomycin. The variation between our data and the majority of the literature could be due to differences in cell types and the condition of the cells, as most studies have focussed on colorectal cancer tissues [7,8,43,44]. It seems unlikely that the discrepancy in PPARδ effects is due to species differences as most of the cancer studies were done using mouse tissues. However, a few studies clearly indicate that PPARδ may not provide any prevention in certain types of apoptosis, such as in the López et al. [45] study where overexpression of PPARδ did not protect against aspirin-induced apoptosis in Jurkat cells. Thus, it is possible that PPARδ may only protect cells from apoptosis in certain conditions and cell types. From the data we collected, GW501516 has no effect on cell survival or anisomycin-induced apoptosis. The IMCD and possibly the IMCD-K2 cells are resistant to cell death because their environment in vivo is one of hypertonicity [46]. This inherent resiliency may make them resistant to different treatments, similar to that previously observed in a mouse IMCD cell line [47].

Summary.
The main objective of this study was to determine whether PPARδ regulates growth and apoptosis in IMCD-K2 cells. PPARδ is highly expressed in the CD, and thus may be responsible for the protection of the CD. We showed that growth responses, including DNA and protein synthesis, in these cells are reduced by PPARδ ligands, but only at high doses. GW501516 had no effect on anisomycininduced apoptosis in the IMCD-K2 cell line.