Autosomal-dominant polycystic kidney disease (ADPKD) is characterized by formation of cysts from tubular epithelial cells. Previous studies indicate that secretion of prostaglandin E2 (PGE2) into cyst fluid and production of cAMP underlie cyst expansion. However, the mechanism by which PGE2 directly stimulates cAMP formation and modulates cystogenesis is still unclear, because the particular E-prostanoid (EP) receptor mediating the PGE2 effect has not been characterized. Our goal is to define the PGE2 receptor subtype involved in ADPKD. We used a three-dimensional cell-culture system of human epithelial cells from normal and ADPKD kidneys in primary cultures to demonstrate that PGE2 induces cyst formation. Biochemical evidence gathered by using real-time RT-PCR mRNA analysis and immunodetection indicate the presence of EP2 receptor in cystic epithelial cells in ADPKD kidney. Pharmacological evidence obtained by using PGE2-selective analogs further demonstrates that EP2 mediates cAMP formation and cystogenesis. Functional evidence for a role of EP2 receptor in mediating cAMP signaling was also provided by inhibiting EP2 receptor expression with transfection of small interfering RNA in cystic epithelial cells. Our results indicate that PGE2 produced in cyst fluid binds to adjacent EP2 receptors located on the apical side of cysts and stimulates EP2 receptor expression. PGE2 binding to EP2 receptor leads to cAMP signaling and cystogenesis by a mechanism that involves protection of cystic epithelial cells from apoptosis. The role of EP2 receptor in mediating the PGE2 effect on stimulating cyst formation may have direct pharmacological implications for the treatment of polycystic kidney disease.
- G-protein coupled receptor
- polycystic kidney disease, three-dimensional cell culture system
autosomal-dominant polycystic kidney disease (ADPKD) is a genetic disorder derived from mutations mostly of the PKD1 and PKD2 genes, which encode for polycystin-1 and -2 (PC-1 and -2), respectively (47). ADPKD is characterized by abnormal growth of tubular epithelial cells that form multiple fluid-filled cysts (43). Various studies in vitro using renal epithelial tubular cells (RTC) and cystic epithelial cells (CEC) derived from normal and polycystic human kidneys indicate that the level of cAMP plays an important role in the regulation of cyst formation (20, 32, 55, 56). Therefore, various natural ligands of G protein-coupled receptors that are known to induce adenylyl cyclase and cAMP formation in the kidney possibly stimulate cyst enlargement and ADPKD progression.
Prostaglandin E2 (PGE2) is a major lipid mediator in the kidney that exerts both autocrine and paracrine actions (39, 51). A potential role of PGE2 in ADPKD is suggested by higher levels of PGE2 during the progression of the disease (41), secretion of PGE2 into the cyst fluid of human ADPKD kidneys (14), and inhibition of cyst fluid-induced cAMP formation with L-161982, a PGE2 receptor 4 antagonist (2). PGE2 activity is mediated by binding to four different G protein-coupled receptors named E-prostanoid (EP) receptors 1–4. Among them, EP2 and EP4 mediate their effect through coupling to G stimulatory proteins, stimulating the induction of cAMP formation and activation of protein kinase A (7, 13). Stimulation of EP3 receptors also antagonizes vasopressin action, known to stimulate cAMP, through the induction of Rho activation and actin polymerization in collecting-duct principal cells (44). However, EP3 receptor stimulation typically inhibits the formation of cAMP as well as Ca2+ release through G inhibitory pertussis toxin-sensitive protein (7). Interestingly, EP3 mRNA has been shown to be upregulated in human ADPKD kidneys (38). The EP1 receptor induces inositol 3-phosphate formation and consequently the release of Ca2+.
The cellular actions of PGE2 are possibly regulated via the expression of the different EP receptors located in different tubular segments and cell types in the kidney. The EP1 receptor is expressed in the collecting ducts (7, 8, 18, 31), and EP3 receptor is mainly located in distal tubules, collecting ducts, and glomeruli (8, 31). Whereas the EP4 receptor is located in the collecting ducts and glomeruli in normal kidney, the presence of EP2 receptor in the kidney is still elusive (7, 8, 31). The absence of EP2 receptor in RTC remains an enigma, because experiments using EP2 receptor knockout mice (24), in which the animals develop salt-sensitive hypertension, suggest a critical role for EP2 receptor in sodium regulation in the kidney. Both the EP2 and EP4 receptors are identified by their vasodepressor effect, in which EP2, but not EP4, is sensitive to butaprost, a PGE2 agonist (7). PGE2 induces cAMP formation in CEC from ADPKD kidneys, which suggests a role for EP2 or EP4 receptor in mediating this effect (3). Also, it has been reported that L-161982, an EP4 antagonist, partially inhibited PGE2-induced cAMP formation (2). Because the particular EP receptor mediating PGE2-induced cystogenesis has not been extensively characterized, our study aims to define PGE2 receptor subtype and signaling involved in ADPKD.
MATERIALS AND METHODS
Cell cultures and culture conditions.
RTC were isolated from human kidney cortex and were maintained as previously described by our laboratory (48, 49). CEC from cyst walls of human kidneys from individuals with ADPKD were isolated essentially as described (53, 55). Cells derived from different kidneys were identified by using different cell numbers. RTC and CEC were propagated in MEMα (Invitrogen, Carlsbad, CA) supplemented with 10 ng/ml EGF, 5 μg/ml hydrocortisone (Sigma, St. Louis, MO), 6.25 μg/ml insulin, 6.25 μg/ml transferin, 6.25 ng/ml selenous acid (BioWhittaker, Walkersville, MD), 10% (vol/vol) heat-inactivated FBS (Atlanta Biologicals, Atlanta, GA), and 1% antibiotic/antimycotic solution (Invitrogen). Chinese hamster ovary (CHO)-AA8 cells were purchased from Clontech (Mountain View, CA) and were cultured in MEMα supplemented with 10% FBS and 1% antibiotic/antimycotic solution. CEC treatments with PGE2 (Sigma), butaprost free acid (Cayman, Ann Arbor, MI), or forskolin (Sigma) were performed as indicated. Each experiment was performed at least two times with cells derived from different kidneys. Our study protocols were approved by the Institutional Review Board of the University of Oklahoma Health Sciences Center.
Three-dimensional gel system.
RTC and CEC (0.15 × 105–0.5 × 105 cells/well) were suspended in a 4°C solution containing complete MEMα and a 1:1 (vol/vol) mixture of liquid rat-tail collagen I and Matrigel (Becton Dickinson, Franklin Lakes, NJ) and were poured into cell-culture inserts as previously described (19). Cells were treated as indicated in figure legends. Cysts were observed by phase-contrast microcopy and were counted before fixation by incubating gels in 10% formalin (Labsco, Louisville, KY). Permeabilization of cells was performed in PBS supplemented with 0.2% saponin and 1% BSA (Sigma), and nuclei were stained with 2 μg/ml 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma) in PBS. Cysts were viewed with an IX50 inverted fluorescent microscope (Olympus, Melville, NY), and size was determined by using the SPOT program (Diagnostic Instruments, Sterling Heights, MI).
Quantitative real-time RT-PCR.
The method for mRNA quantitation by real time RT-PCR was previously published (12). mRNA was treated with RNase-free DNase (Invitrogen), or extraction was performed by using an RNeasy Plus mini kit (Qiagen, Valencia, CA) that removes contaminating genomic DNA from the sample. RNA extraction from three-dimensional (3-D) gels was performed by using the TRlzol method (Invitrogen). mRNA was measured by using SYBR Green technology (Applied Biosystems, Foster City, CA) on an ABI PRISM 7000 sequence detection system (Applied Biosystems) and was calculated by using the formula 2(−ΔΔCT) (29). The sequences of primers not published previously are shown in Table 1.
Transfection of EP receptors.
cDNA for human EP1, EP2, EP3, and EP4 receptors were provided by the cDNA Resource Center (University of Missouri-Rolla). DNA plasmids were purified by using a maxiprep kit (Qiagen) according to the manufacturer's instructions. pcDNA3.1 vector (1 μg/well) containing EP1, EP2, EP3, and EP4 cDNAs or empty vector (used as a negative control) were transiently transfected by using FuGENE 6 (Roche, Indianapolis, IN) in a CHO-AA8 cell suspension that was plated in six-well plates at ∼50% confluence. After 2 days, cells were washed with PBS and stored at −70°C until they were used for Western blot analysis.
Western blot analysis.
Membranes from ADPKD kidney were prepared by tissue homogenization with a Polytron homogenizer in ice-cold 20 mM HEPES (pH 7.6), supplemented with 1 mM EDTA, 2 mM tetrasodium pyrophosphate, 50 μg/ml bacitracin, 10 μg/ml soybean trypsin inhibitor, 10 μg/ml leupeptin, and 1 mM PMSF (Sigma). The homogenates were centrifuged on a cushion of 43% (wt/vol) sucrose in the homogenization buffer. Membranes at the sucrose interface were collected and were stored at −70°C. Protein concentration was assessed by using a Bio-Rad protein assay (Bio-Rad, Hercules, CA). Membranes from tissue and CHO-AA8 cells were solubilized in sample buffer (62.5 mM Tris·HCl, pH 6.8, 2% SDS, 1% 2-mercaptoethanol, 3M urea, 20% glycerol; Bio-Rad) at 65°C for 20 min. Proteins were subjected to SDS-10% PAGE and immunoblot analysis with an enhanced chemiluminescence detection system (Amersham, Little Chalfont, UK).
Immunohistochemical staining analysis.
Formalin-fixed paraffin-embedded ADPKD kidney sections were deparaffinized. To block endogenous peroxidase activity, slides were incubated with the Peroxidase Block Envision system (Dako, Carpinteria, CA) for 15 min at room temperature. Sections were then incubated in PBS supplemented with 0.2% saponin and 1% BSA. Slides were incubated at 4°C overnight with mouse IgG (mIgG, 1 μg/ml; Jackson, West Grove, PA) or mouse polyclonal anti-EP2 receptor antibody (PE22; obtained from the University of Texas, Southwestern Medical Center; currently available from Arizona State University; http://www.biosignatures.org/antibody/) diluted 1:100. Sections from formalin-fixed paraffin-embedded gels from 3-D cultures were processed similarly. Proliferation of cells was assessed by using a proliferating cell nuclear antigen (PCNA) antibody (Dako) diluted 1:500. Apoptosis was analyzed by using the Cell Death In Situ detection system that used a terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assay, including detection of fluorescein dUTP by an anti-fluorescein antibody according to the manufacter's instructions (Roche). The sections were processed with DakoCytoformation Envision kit (Dako) or SuperpicTure Polymer detection kit (Zymed, South San Francisco, CA) as described by the manufacturers, in which peroxidase-conjugated secondary antibody was reacted with diaminobenzidine. Slides were counterstained with hematoxylin and then were cemented with Crystal Mount (Biomedia, Foster City, CA).
Measurement of cAMP levels.
Cells at confluence were incubated for 2 days in serum-free, supplement-free, antibiotic-free MEMα. Following pretreatment with 0.25 mM 3-isobutyl-1-methylxanthine (Sigma) in MEMα containing 5 mg/ml lactalbumin hydrolysate (Sigma) for 30 min to inhibit cAMP degradation, cells were treated with or without PGE2, butaprost free acid, or forskolin for 30 min. Then the cells were washed with PBS and were scraped into 0.1 M HCl with 0.1% Triton X-100. After centrifugation, the supernatant was submitted to a cAMP immunoassay kit (Assay Design, Ann Arbor, MI) and Bio-Rad protein assay. Results were analyzed by using curve fitting, and statistics were analyzed with GraphPad PRISM software (GraphPad, San Diego, CA).
RNA interference (RNAi) analysis of human EP2 receptor was performed by using small interfering RNA (siRNA) duplexes from Qiagen. The control siRNA sense sequence was UUCUCCGAACGUGUCACGUdTdT, and two different siRNA EP2 receptor sense sequences were GAGUGGACUCAGUGGGUUAdTdT and CUACCGUUAUACACAUAUAdTdT. CEC (∼2 × 105 cells) were transfected with 3 μg siRNA by nucleofection using a Nucleofector basic kit with “primary mammalian epithelial cell solution” and program T-23 (Amaxa, Gaithersburg, MD) and were plated at 60–80% confluence in 6- or 24-well plates for measurement of mRNA or cAMP, respectively. Cells at confluence were incubated for 1–2 days as indicated in serum-free, supplement-free, antibiotic-free MEMα and were treated with or without 77 nM PGE2 for mRNA analysis.
Effect of PGE2 on cyst formation in 3-D matrix cultures.
Figure 1 shows the effect of PGE2 on cyst formation in a 3-D gel matrix using CEC derived from human ADPKD kidneys. CEC primarily formed cysts in medium supplemented with FBS; cyst formation was augmented by addition of PGE2 in medium supplemented with or without serum. The size of cysts was maximal when medium was supplemented with both FBS and PGE2. CEC cysts formed not only inside the gels, but some macroscopic cysts expanded out from the surface and were easily visible without magnification after prolonged culture. Cystic structures were also well documented by appearance of nuclei on staining with DAPI. The 3-D structure of cysts is clearly visible, with nuclei in the cyst walls lining a central cavity, which can be appreciated at different levels of focus.
Figure 2 shows that in the absence of PGE2, RTC formed tubular structures that can be visualized by phase-contrast microscopy and also by the linear appearance of nuclei on staining with DAPI. Under these same conditions, CEC formed cysts (Fig. 1). PGE2 treatment inhibited tubule formation and induced formation of small cysts. The effect of PGE2 on cyst size and number was determined in three different CEC and RTC cultures derived from different kidneys. Table 2 shows that RTC formed few cysts with a small average size compared with CEC in the absence of PGE2. PGE2 treatment increased cyst number by 9- to 52-fold in RTC and 2- to 3-fold in CEC. Note that the size and number of cysts formed by CEC in the absence of PGE2 was greater than for RTC even in the presence of PGE2. Therefore, PGE2 stimulates cyst formation in both CEC and RTC, but CEC have a greater propensity for cystogenesis, consistent with their derivation from ADPKD kidneys.
Because kidney formation is mostly normal at birth during ADPKD (33) and cysts form from tubular epithelial cells after birth, it was of interest to determine whether PGE2 affects well-organized RTC tubules in 3-D matrix. Figure 3 shows that addition of PGE2 to already formed tubules results in tubule disintegration into patches that expand to form cysts. Staining of nuclei demonstrates that the tubules expand and cells become organized in a circular manner, forming cysts. Therefore, our results indicate that PGE2 induces cyst formation in RTC independent of mutations in both PC-1 or PC-2, but CEC have a greater propensity toward forming more and larger cysts.
Characterization of the EP receptor subtypes present in CEC.
To determine the different EP receptor subtypes present in CEC, we initially analyzed the mRNA expression of four different EP receptors. Figure 4 shows that CEC express high levels of EP2 receptor mRNA and low levels of EP1 and EP4, whereas EP3 was not detected. The expression of EP2 mRNA is particularly surprising because previous studies suggested the absence of EP2 receptor in RTC of human tissue. To further confirm the presence of EP2 receptor protein in human kidneys, we performed Western blot analysis using an EP2 antibody generated by genetic immunization. We did not find other commercially available antibodies to be specific or sensitive enough to be used in Western blot and immunohistochemical analysis. The specificity of this antibody to EP2 receptor, but not to other PGE2 receptor subtypes, was demonstrated by analyzing extracts of CHO cells transfected with the different receptor subtypes. Figure 5 shows that cells transfected with EP2 receptor exhibit a 65-kDa protein band that does not appear in other CHO extracts. We also observed a protein band of similar molecular weight detected by EP2 antibody in membranes from ADPKD kidney.
We next performed immunohistochemical-staining analysis with the same antibody for EP2 receptor that was used for immunoblot analysis. mIgG was used as a negative control. Figure 6 shows that EP2 receptors are present on the CEC-lining cysts in polycystic kidney tissue from patients with ADPKD. Staining was not evident in blood vessels (Fig. 6, middle). EP2 receptors are mostly expressed on the apical side of the cysts, which implies potential direct contact with PGE2 secreted into the cyst fluid during progression of ADPKD. Therefore, our study indicates that EP2 receptors are expressed in CEC and may play an important role in the cystogenic potential of PGE2 activity.
Pharmacological and functional characterization of EP2 receptors in CEC.
To demonstrate the functionality of the EP2 receptors, we analyzed the effect of pharmacological analogs of PGE2 to modulate cAMP levels in CEC. Figure 7 shows the effect of PGE2 and butaprost free acid, a selective agonist for EP2 receptors, on cAMP formation in CEC. The dose-response curves for PGE2 and butaprost free acid yielded ED50 of 37.4 ± 1.7 and 212.5 ± 75.8 nM (means ± SE, n = 3 and 4), respectively. This difference in effective concentration of PGE2 and butaprost free acid may correspond to differences in binding affinity observed in COS-1 cells transfected with the cloned rabbit EP2 receptor (17). Stimulation of cAMP with 77 nM PGE2, 5 μM butaprost free acid, and 10 μM forskolin was 23.8 ± 4.5-, 24.2 ± 4.8-, and 5.1 ± 1.3-fold over basal, respectively (means ± SE, n = 3). Therefore, maximal stimulation was similar for PGE2 and butaprost free acid, whereas the effect of forskolin was lower, possibly because forskolin needs time to permeate into cells, where it activates adenylyl cyclase.
We also used AH-6809, an antagonist of EP1 and EP2 receptor but not EP4 receptor (26). Because EP2 and EP4 receptors but not EP1 mediate an increase in cAMP, the effect of AH-6809 on cAMP is specific for EP2. Figure 7 shows that AH-6809 inhibits PGE2 stimulation of cAMP formation in CEC, which substantiates the conclusion that the PGE2 effect is mediated by EP2 receptor.
We further determined whether activation of EP2 receptors contributes to cystogenesis in CEC cultures. For this purpose, CEC were cultured in a 3-D matrix in the presence or absence of PGE2, butaprost free acid, and forskolin, a pharmacological adenylate cyclase activator. Figure 8 shows that PGE2, butaprost free acid, and forskolin stimulate cyst formation. This result indicates that the mechanism of cystogenesis induced by PGE2 is likely to involve EP2 receptors and formation of cAMP.
To further demonstrate that EP2 receptors mediate the PGE2 effect, we used RNAi to reduce endogenous levels of EP2 receptors. Figure 9, top shows the effect of two different EP2 siRNAs on stimulation of cAMP. EP2 RNAi greatly reduced cAMP levels in CEC treated with PGE2 and butaprost free acid. Therefore, these results demonstrate that EP2 receptor expression mediates PGE2 and butaprost free acid effect on cAMP formation. The effect of siRNA on EP2 receptor expression was also analyzed by real-time RT-PCR analysis (Fig. 9, middle). Because we demonstrated previously that various treatments may modulate the expression of reference genes (12), we used cyclophilin A and 18S as two different reference genes to confirm mRNA quantitation. Our results indicate that the two different siRNAs reduced EP2 receptor mRNA expression. In addition, we also analyzed the effect of PGE2 treatment on EP2 receptor expression. PGE2 treatment induced EP2 receptor expression by fourfold. This effect was specific because EP2 siRNAs abolished it (Fig. 9, middle). We further demonstrated that PGE2, butaprost free acid, and forskolin similarly stimulated EP2 receptor expression. Inhibition by siRNA indicates specificity of this effect (Fig. 9, bottom). Therefore, we demonstrated here that PGE2-, butaprost free acid-, and forskolin-mediated cAMP induction stimulate expression of EP2 receptor. This result implies possible regulation of EP2 receptor expression by PGE2 and cAMP-mediated signaling in the kidney.
PGE2-induced cystogenesis involves apoptosis but not proliferation.
Both proliferation and apoptosis are induced during cystogenesis (6). To determine the mechanism by which PGE2 alters cell number during cystogenesis, we analyzed the expression of PCNA and performed TUNEL assays in cysts formed in 3-D cultures of CEC treated with or without PGE2 in growth medium that supported maximal cyst formation (Fig. 1). Figure 10 shows that proliferation monitored by PCNA expression is evident in CEC-lining cysts either with or without PGE2 treatment. In contrast, apoptosis was more apparent in the absence of PGE2. To quantify these results, we determined the percentage of cells undergoing proliferation or apoptosis in three different experiments using CEC derived from different kidneys. Table 3 indicates that a similar percentage of cells, ∼70%, are proliferative in the presence or absence of PGE2. However, the number of cells undergoing apoptosis was significantly lower, by 22%, in the presence of PGE2. Therefore, our study indicates that PGE2 protects cells from apoptosis, thereby promoting cyst formation.
To further analyze the effect of PGE2, butaprost free acid, and forskolin on proliferation and apoptosis, we performed real-time RT-PCR analysis for mRNA expression of marker genes in 3-D gels. Figure 11 shows that expression of the antiapoptotic factor Bcl-2 and cell cycle-regulated cyclin D3 are upregulated by treatment of PGE2, butaprost free acid, and forskolin. However, expression of proapoptotic factor Bax, cell cycle-regulated cyclin D1, cell proliferation-related PCNA and minichromosome maintenance protein 2 (MCM2), and negative control cyclophilin are not changed. These results indicate that the effect of PGE2 mediated by EP2 receptor and cAMP signaling protects CEC from apoptosis. They imply a mechanism that involves regulation of Bcl-2 expression. The lack of treatment's effect on expression of PCNA and MCM2, a marker for DNA replication, further confirms that PGE2 did not affect the rate of proliferation.
Our study addresses the role and mechanism by which PGE2 activity is involved in ADPKD. Because PGE2 stimulates different receptors mediating distinct signaling pathways, it is important to determine specific receptor subtype(s) responsible for PGE2 activity in CEC. Our study provides biochemical and functional evidence demonstrating for the first time that EP2 receptor is expressed in the human ADPKD kidney and directly mediates PGE2 activity on cyst formation.
Previous studies showed diffuse expression with no specific localization of EP2 receptor mRNA as detected by in situ hybridization in the human and rabbit kidney (8, 17). RT-PCR analysis reveals EP2 receptor mRNA expression in the glomeruli and medullary collecting ducts in rabbit kidney (17). In contrast, EP2 receptor mRNA expression and butaprost-induced cAMP are reported in the tubular cells of the rat medulla (22, 52). Also, it has been reported that micromolar concentrations of L-161982, an EP4 antagonist, partially inhibited PGE2-induced cAMP formation in CEC (2). However, the Ki for binding to EP2 and EP4 receptors is 21 μM and 24 nM, respectively (30), and an almost maximal inhibition of PGE2 effect mediated by EP4 was observed at 10 nM concentration of L-161982 (15). Therefore, it is possible that higher concentrations of L-161982 also inhibit EP2-mediated activation. EP2 receptor in human kidney was detected by immunoblot analysis and has been localized in the renal vasculature but not RTC by immunohistochemistry (31). Using a different antibody, we were able to detect EP2 receptor by immunoblot analysis but were unable to determine the specific localization for EP2 receptor in the normal human kidney (results not shown). However, we clearly observed EP2 receptor on the apical side of cysts in human ADPKD kidneys but not in the renal vasculature in ADPKD kidney (Fig. 6). It is possible that PGE2 secretion into cyst fluid induces EP2 receptor in ADPKD kidney because PGE2 induced EP2 receptor mRNA expression by three- to fourfold in CEC primary cultures (Fig. 9). Therefore, our results also imply that the effect of PGE2 in cysts is amplified by autocrine and/or paracrine stimulation of EP2 receptor expression. Previous studies indicate that PGE2 regulates cell growth and basolateral anion uptake through cAMP signaling in rabbit proximal renal tubules and human kidney epithelial cells (37). Because both cell growth and fluid secretion play a major role in cyst formation (16, 32), theoretically, EP2 and EP4 receptors may play a role in this process. It has been established that EP4 receptor mediates PGE2 effect in rabbit collecting ducts (36); however, none of these receptors has been clearly characterized in cortical tubular epithelial cells. Our data also provide functional evidence for a role of EP2 receptor in CEC by using receptor-selective agonists/antagonists and siRNA (Figs. 7–9). Therefore, our study supports a model in which PGE2 secreted into cyst fluid binds to EP2 receptor, modulates EP2 receptor expression, and stimulates cAMP formation, thus contributing to cystogenesis.
To further determine the mechanism by which PGE2 stimulates cyst formation, we determined whether PGE2 stimulates proliferation or apoptosis of epithelial cells lining cysts in 3-D matrix. Although CEC were highly proliferative (70%) and no difference was found following PGE2 treatment, reduction of apoptosis was observed. Notably, the effect of PGE2 on cyst formation was obtained in medium containing mitogens such as EGF and insulin with (Fig. 1 and Table 2) and without FBS (Figs. 1 and 8). Therefore the high rate of proliferation in CEC (measured in the presence of FBS) is consistent with the presence of mitogenic factors in 3-D culture medium. Previous studies using monolayer cultures suggest that cAMP signaling stimulates proliferation in CEC while having the opposite or no effect in RTC (20, 55). However, stimulation of cAMP-dependent signaling induces transepithelial monolayer chloride transport and cyst formation in 3-D collagen I gel cultures of both cell types (32, 53, 56). In this model, it has also been shown that cAMP promotes RTC to form cysts only in the presence of EGF (32, 56), which further indicates that the mitogenic effects of cAMP agonist observed in monolayer cultures do not necessarily reflect on cyst formation in 3-D cultures. The high percentage of apoptotic cells observed (37%) on CEC-lining cysts in 3-D cultures is consistent with data showing numerous apoptotic nuclei in human polycystic kidney (54).
Previously, it has been reported that PGE2 binding to the EP2 receptor promotes cell survival through an antiapoptotic mechanism in radiated intestinal epithelial cells (21, 46). Endothelial cells from EP2 receptor knockout mice were more susceptible to apoptosis than cells from wild-type animals (23). Two other recent studies demonstrated that PGE2 through EP2 receptor protects mouse embryonic stem cells exposed to H2O2 (28) and human lung fibroblasts incubated with cigarette smoke extract (42) from apoptosis. Although it has been proposed that PGE2 induces cell growth, our study indicates for the first time that PGE2 promotes cystogenesis of CEC through protection against apoptosis. It has been shown that both epithelial cell apoptosis and proliferation are dysregulated in ADPKD (11, 27). Our study demonstrates that PGE2 could reduce cyst-lining epithelial cell apoptosis and simultaneously increase the ratio of Bcl-2 to Bax that commonly is used to measure apoptotic susceptibility (9). PGE2 has been shown to induce Bcl-2 expression in human colon cancer cells (40). Interestingly, loss of Bcl-2 of knocked-out mice results in the development of polycystic kidney disease, provoking renal failure (50).
We also observed that PGE2, butaprost free acid, and forskolin induced mRNA expression of cyclin D3 but not cyclin D1. Functions of D-type cyclins commonly overlap in activating the cyclin-dependent kinases CDK4 and CDK6 in G1 and thereby promote the cell's entrance into S phase (1). However, lack of a treatment effect on PCNA expression that is mainly in the late G1 and S phases of the mitotic cycle suggest that induction of cyclin D3 did not significantly affect cell proliferation in the presence of highly mitogenic culture medium. Expression of MCM2, a marker for proliferation involved in DNA replication (35), further confirmed that PGE2 did not affect the rate of proliferation. Cyclin D1 and D3 are not fully redundant, and a previous study indicated a distinct role of cyclin D3 in induction and/or maintenance of terminal differentiation (1). Our result implies that PGE2 activity is dependent on high apoptotic activity in CEC. This conclusion is further supported by a study showing that inhibition of the proapoptotic caspase-3, which decreases apoptosis and proliferation of RTC, induces inhibition of cystogenesis and preserves kidney function in the Han:SPRD rat model of PKD (45). PC-1 mutations in ADPKD may be related to the large number of apoptotic cells because PC-1 expression stimulates tubulogenesis and inhibits cystogenesis by inducing growth arrest and resistance to apoptosis (4–6).
In the context of the mutation of PC-1 leading to ADPKD, conserved sequences on the cytoplasmic part of PC-1 are related to G-protein activation and mediate interaction with G inhibitory proteins and regulator of G-protein signaling 7 (10, 25, 34). Mutations of PC-1 may possibly alter G inhibitory protein-mediated inhibition of cAMP formation and be implicated in the process of cystogenesis. Our study indicates that PGE2 induces cyst formation but also inhibits tubule formation in RTC derived from the normal kidney. This result suggests that genetic conditions and environment restrict cyst formation in normal kidney but not in ADPKD kidney. Secretion of PGE2 into cyst fluid in ADPKD kidney potentially plays a major role in the progression of the disease (2, 14).
In conclusion, our results suggest that PGE2 secreted into cyst fluid binds to adjacent EP2 receptors located on the apical side of cysts and stimulates EP2 receptor expression. PGE2 binding to EP2 receptor leads to cAMP signaling and cystogenesis by a mechanism that involves protection of CEC from apoptosis. Therefore, blockage of the EP2 receptor may have significant implications for developing clinically relevant therapeutic strategies for ADPKD.
This work was supported by a grant of the Polycystic Kidney Disease Foundation, and funding of the Children's Medical Research Institute, and the Children's Miracle Network Foundation.
We thank the cDNA Resource Center (University of Missouri-Rolla, http://www.cdna.org) for providing the EP receptor expression vectors.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society