点击显示 收起
【摘要】 Epithelial-to-mesenchymal transition (EMT) has emerged as a critical event in the pathogenesis of tubulointerstitial fibrosis. EMT is typically induced by transforming growth factor- 1 (TGF- 1 ) and inhibited by hepatocyte growth factor (HGF). The present study was undertaken to evaluate the potential role of cyclooxygenase (COX)-2-derived PGE 2 in regulation of EMT in cultured Madin-Darby canine kidney (MDCK) cells, in the setting of HGF treatment. Exposure to 50 ng/ml HGF significantly induced COX-2 protein expression and PGE 2 release, whereas other growth factors, including epidermal growth factor, the insulin-like growth factor I protein, platelet-derived growth factor-BB, and TGF- 1, had no effects on COX-2 expression or PGE 2 release. COX-2 induction by HGF was preceded by activation of ERK1/2, and an ERK1/2-specific inhibitor, U-0126 (10 µM), completely abolished HGF-induced COX-2 expression. Exposure of MDCK cells to 10 ng/ml TGF- 1 for 72 h induced EMT as evidenced by conversion to the spindle-like morphology, loss of E-cadherin, and activation of -smooth muscle actin. In contrast, treatment with 1 µM PGE 2 completely blocked EMT, associated with a significant elevation of intracellular cAMP and complete blockade of TGF- 1 -induced oxidant production. cAMP-elevating agents, including 8-Br-cAMP, forskolin, and IBMX, inhibited EMT and associated oxidative stress induced by TGF- 1, but inhibition of cAMP pathway with Rp-cAMP, the cAMP analog, and H89, the protein kinase A (PKA) inhibitor, did not block the effect of PGE 2. The effect of HGF on EMT was inhibited by 50% in the presence of a COX-2 inhibitor SC-58635 (10 µM). Therefore, our data suggest that PGE 2 inhibits EMT via inhibition of oxidant production and COX-2-derived PGE 2 partially accounts for the antifibrotic effect of HGF.
【关键词】 tubulointerstitial fibrosis endstage renal disease
RENAL TUBULOINTERSTITIAL FIBROSIS (TIF), which is caused by diverse clinical entities, is the common pathological pathway leading to end-stage renal disease (ESRD) ( 9, 10, 35, 39 ). TIF is considered to be an irreversible process characterized by the progressive loss of renal tubules and stimulation of myofibroblasts which produce extracellular matrix (ECM) proteins, such as collagens types I, III, fibronectin, and laminin, leading to renal fibrosis ( 10, 18 ). Emerging evidence suggests that epithelial-to-mesenchymal transition (EMT) is a major event in the pathogenesis of TIF. In response to transforming growth factor- 1 (TGF- 1 ), tubular epithelial cells transdifferentiate to myofibrobalcts ( 4, 5, 9, 32 ). The phenotypic conversion involves loss of epithelial polarity and E-cadherein, disruption of tubular basement membrane, acquisition of spindle-like morphology, de novo synthesis of -smooth muscle actin ( -SMA), and production of matrix proteins ( 22 ).
Hepatocyte growth factor (HGF) has multiple biological activities in a wide variety of cell types. In carcinoma cells, HGF induces EMT, producing the invasive and metastatic phenotype ( 3, 11, 37 ). In contrast, HGF inhibits EMT in renal epithelial cells, exhibiting beneficial effects on chronic renal disease ( 22 ). In cultured renal proximal tubular cells, HGF blocks TGF- 1 -induced EMT ( 43 ). In several animal models of chronic renal disease, including rats with chronic allograft nephropathy ( 1 ) and mice with the ICR-derived glomerulonephritis (ICGN) ( 25 ), administration with the recombinant HGF exhibits remarkable beneficial effects on the kidney. On the contrary, neutralization of endogenous HGF with anti-HGF antibody aggravates renal injury induced by ICGN ( 26 ) and obstructive nephropathy ( 27 ). Administration of HGF clearly holds a promise for treatment of patients with ESRD. However, the molecular mechanism underlying the antifibrotic effects of HGF is still not fully characterized.
Prostaglandins (PGs) are a group of compounds derived from arachidonic acid through the cyclooxygenase (COX) pathway by constitutive COX-1-inducible COX-2 ( 38 ). PGs play an important role in regulation of a wide spectrum of cellular functions in various cell types. A large body of experimental work documents that PGE 2 functions as an antifibrotic factor in the lung and reduction of PGE 2 levels contributes to the pathogenesis of pulmonary fibrosis ( 16, 23, 24 ). In cultured lung fibroblasts, PGE 2 suppresses cell proliferation ( 13 ), collagen production ( 14, 24 ), and fibroblast-to-myofibroblast transition ( 20 ). In contrast to the detailed knowledge about the antifibrotic function of PGE 2 in the lung, the role of PGE 2 in renal fibrosis is less understood. In addition, there is no information concerning the role of PGE 2 in regulation of EMT in renal tubular epithelial cells or other types of epithelial cells. The present study provides evidence that PGE 2 is a potent inhibitor of EMT in renal epithelial cells and COX-2-derived PGE 2 partially mediates the antifibrotic effect of HGF.
MATERIALS AND METHODS
Reagents and antibodies. Recombinant human HGF, TGF- 1, epidermal growth factor (EGF), PDGF-BB, insulin-like growth factor (IGF)-1, and extracellular signal-regulated kinases 1/2 (ERK1/2) inhibitor, U-0126, were purchased from Calbiochem (Cambridge, MA). 8-Br-cAMP, forskolin, IBMX, Rp-cAMP, H89, antibodies against -SMA, E-cadherin, and -actin were purchased from Sigma (St. Louis, MO). Rabbit anti-murine COX-2 polyclonal antiserum was from Cayman (Ann Arbor, MI). Fluorescein isothiocyanate (FITC)-conjugated goat anti-rat IgG, horseradish peroxidase (HRP)-conjugated goat anti-mouse and anti-rabbit IgG were from Santa Cruz Biotechnology (Santa Cruz, CA).
Cell culture. Madin-Darby canine kidney (MDCK) cells were cultured in DMEM, containing 10% FBS, 1% penicillin/streptomycin (GIBCO, Burlington, ON) at 37°C under 5% CO 2 in a humidified incubator. Cells were grown to 80% confluence, serum starved for 24 h, and then treated with vehicle or 50 ng/ml human recombinant HGF, TGF- 1 with/without PGE 2 for the indicated times. Microscopic examination was performed during each experiment to assess the morphological changes before sample analysis.
Western blotting. MDCK cells were lysed and subsequently sonicated in PBS containing 1% Triton X-100, 250 µM PMSF, 2 mM EDTA, and 5 mM DTT (pH 7.5). Protein concentration was determined by Coomassie reagent. Thirty micrograms of protein from whole cell lysates were denatured in boiling water for 10 min, separated by SDS-PAGE, and transferred onto nitrocellulose membranes. The blots were blocked overnight with 5% nonfat dry milk in Tris-buffered saline (TBS), followed by incubation for 1 h with rabbit anti-murine polyclonal antiserum to COX-2 or mouse anti- -SMA monoclonal antibody. After being washed with TBS, blots were incubated with a goat anti-HRP-conjugated secondary antibody and visualized with ECL kits (Amersham). Immunoblotting of -actin served as a loading control. Quantitative data were generated as arbitrary values relative to control and normalized by -actin.
Immunofluorescence microscopy for E-cadherin. Cells were grown on coverslips and stimulated for 72 h with TGF- 1 in the presence or absence of HGF or PGE 2. The medium was removed, and the cell layer was rinsed with PBS. Cells were fixed and permeabilized with acetone-methanol for 10 min at -20°C and then were rehydrated with PBS and blocked with 5% BSA in PBS for 1 h. Coverslips were sequentially incubated with rat monoclonal anti-E-cadherin and FITC-labeled goat anti-rat antibody each for 60 min at room temperature. Cells were then visualized and photographed by fluorescence microscopy at x 40 magnification. Negative controls were performed using nonimmune serum or IgG instead of first antibodies.
PGE 2 enzyme immunoassay. PGE 2 in the culture media was measured with an enzyme immunoassay kit. The assay was performed according to the manufacturer?s instruction. Briefly, 25 or 50 µl of the medium, along with a serial dilution of PGE 2 standard samples, were mixed with appropriate amounts of acetylcholinesterase-labeled tracer and PGE 2 antiserum and incubated at room temperature for 18 h. After the wells were emptied and rinsed with wash buffer, 200 µl of Ellman?s reagent were added.
Phosphorylation of MAP kinases. MDCK cells grown in a six-well plate were lysed by sonication for 10 s in 300 µl of 1 x Laemmli sample buffer containing 10 mM Tris, 1.4% SDS, and 40 mM DTT (pH 6.8). The protein samples were heated at 60°C for 15 min and electrophoresis was performed. The blots were blocked in 5% nonfat dry milk for 1 h and incubated overnight at 4°C with the primary antibodies against phospho-ERK1/2 and phospho-p38 at a dilution of 1:1,000. The secondary antibody and ECL reaction were the same as described above.
DCFDA fluorescence measurement of reactive oxygen species. The fluorogenic substrate 2',7'-dichlorofluorescein diacetate (DCFDA) is a cell-permeable dye that is oxidized to highly fluorescent 2',7'-dichlorofluorescein (DCF) by H 2 O 2 and can therefore be used to monitor intracellular generation of reactive oxygen species (ROS). For measurement of ROS, cells were grown onto glass cover slides. When the cells reached confluence, they were washed twice with PBS and incubated for 30 min with 50 µM DCFDA diluted in Opti-MEM with 10% FCS. Then hypertonic medium was added. At the end of the incubation period, the cells were again washed twice with PBS and imaged by confocal laser microscopy. To quantitate ROS levels, cells were seeded to 96-well plates and were treated as described above. Relative fluorescence was measured using a fluorescence plate reader (FLUOstar OPTIMA) at excitation and emission wavelengths of 485 and 528 nm, respectively, three times at 90-s intervals.
cAMP assay. Serum-starved MDCK cells grown in six-well plates were pretreated with a phosphodiesterase inhibitor [3-isobutyl-1-methylxanthine (IBMX)] at 10 µM for 30 min and then treated with 1 µM PGE 2. After treatment, medium was removed and the cells were washed with phosphate-buffered saline (PBS). Immediately after being washed, 0.5 ml of 0.1 M HCl was added. After 20 min, cells were scraped into a centrifuge tube and spun for 10 min at 1,000 g to pellet the cell debris. The cAMP enzyme-linked immunosorbent assay was performed according to manufacturer?s instructions (Cayman Chemicals).
Statistical analysis. Values shown represent means ± SE. Statistical analysis was performed by ANOVA and Bonferroni tests or independent sample t -test with a P value of <0.05 being considered statistically significant.
RESULTS
HGF specifically induced COX-2 expression and PGE 2 production. Treatment with HGF at 50 ng/ml elevated COX-2 protein expression by 230 ± 13% ( P < 0.01) at 3 h, and by 283 ± 23% ( P < 0.01) at 6 h, and the expression returned to basal levels at 12 h ( Fig. 1 A ). The induction of COX-2 expression was paralleled by a significant stimulation of PGE 2 release as determined by EIA. The stimulation of PGE 2 release following HGF treatment was detected at 3 h, gradually increasing with time with an maximal effect at 12 h ( Fig. 1 B ).
Fig. 1. Effects of hepatocyte growth factor (HGF) on cyclooxygenase (COX)-2 expression and PGE 2 production. Madin-Darby canine kidney (MDCK) cells were treated with vehicle or 50 ng/ml HGF for the indicated periods of time. COX-2 protein levels were determined by immunoblotting and PGE 2 concentrations by enzyme immunoassay. A : time course of COX-2 induction in response to HGF treatment. Immunoblotting of -actin served as a loading control. Top : densitometric analysis. Bottom : representative immunoblots. Quantitative data were generated as arbitrary values relative to control and normalized by -actin. Unless otherwise specified, relative COX-2 and -smooth muscle actin (SMA) protein levels shown in rest of the figures were all normalized by -actin. B : time course of PGE 2 stimulation. * P < 0.01. ** P < 0.05 vs. controls.
We performed parallel experiments to examine effects of other growth factors, including EGF, IGF-1, TGF- 1, and PDGF-BB, on the COX-2 expression as well as PGE 2 production. In contrast to HGF, none of the other growth factors tested had any significant effects on COX-2 expression or PGE 2 production ( Fig. 2 ).
Fig. 2. Effects of various growth factors on COX-2 expression and PGE 2 production. MDCK cells were treated by EGF (20 ng/ml), IGF-1 (10 ng/ml), TGF- 1 (10 ng/ml), and PDGF-BB (10 ng/ml) for 6 or 12 h, and COX-2 expression and PGE 2 concentration were determined as described above. A : densitometric analysis of COX-2 protein expression. B : PGE 2 concentration. Values are the mean intensity ± SE of 3 independent experiments.
Induction of COX-2 expression by HGF was mediated by ERK1/2. MAP kinase has emerged as a common downstream mediator in the stimulation of COX-2 expression in response to divergent stimuli. We examined involvement of ERK1/2 in HGF-induced COX-2 expression in MDCK cells. HGF at 50 ng/ml rapidly activated ERK1/2 as assessed by immunoblotting detection of phospho-ERK1/2 ( Fig. 3 A ). Pretreatment with an ERK1/2 inhibitor, U-0126 (10 µM), completely blocked the COX-2 stimulation induced by HGF. U-0126 even reduced the basal levels of COX-2 expression ( Fig. 3 B ). No obvious cytotoxicities were observed with the dose and duration of the U-0126 treatment.
Fig. 3. ERK1/2 mediation of HGF-induced COX-2 expression. A : HGF activation of ERK1/2. MDCK cells were treated with vehicle or 50 ng/ml HGF for the indicated periods of times. Phospho-ERK1/2 and total ERK1/2 were determined by immunoblotting. B : effect of a ERK1/2 inhibitor, U-0126, on HGF-induced COX-2 expression. The graph summarized densitometric analysis of 3 independent experiments, and representative immunoblots for COX-2 were displayed below. * P < 0.01 vs. controls. P < 0.01 vs. HGF-stimulated cells.
PGE 2 reversed EMT induced by TGF- 1. TGF- 1 is a well-characterized inducer of EMT in renal tubular epithelial cells. To test the role of PGE 2 in TGF- 1 -induced EMT, we first monitored the morphologic changes in MDCK cells treated by TGF- 1 alone or in combination with PGE 2. MDCK cells exhibit typical epithelial-like morphology ( Fig. 4 A ), whereas TGF- 1 treatment for 3 days induced a complete conversion to spindle-like morphology ( Fig. 4 B ). The TGF- 1 -induced morphologic changes were completely prevented by treatment with 1 µM PGE 2 ( Fig. 4 C ).
Fig. 4. Morphologic changes in MDCK cells. The cells were grown in 6-well plates until 80% confluence and then treated with vehicle ( A ), TGF- 1 ( B ), TGF- 1 plus PGE 2 ( C ), TGF- 1 plus HGF ( D ), or TGF- 1 plus HGF and SC-58635 ( E ) for 3 days. Photographs were taken using a Nikon microscope (phase contrast).
E-cadherin, a classic epithelial cell marker, is a membrane-bound protein involved in cell-cell interactions in intact renal tubular epithelial cells. We then performed immunocytochemical analyses to monitor changes in E-cadherin protein expression. As expected, E-cadherin was expressed exclusively in the basolateral membrane of MDCK cells in basal state ( Fig. 5 A ). In contrast, incubation with 10 ng/ml TGF- 1 for 3 days dramatically reduced E-cadherin expression ( Fig. 5 B ). The reduction of E-cadherin was completely prevented in the presence of 1 µM PGE 2 ( Fig. 5 C ).
Fig. 5. Immunofluorescence of E-cadherin in MDCK cells. The cells were grown on coverslips until 80% confluence and then treated with vehicle ( A ), TGF- 1 ( B ), TGF- 1 plus PGE 2 ( C ), TGF- 1 plus HGF ( D ), or TGF- 1 plus HGF and SC-58635 ( E ) for 3 days. Immunofluorescence was performed using rat monoclonal anti-E-cadherin and FITC-labeled goat anti-rat antibody. Cells were visualized and photographed by fluorescence microscopy at x 400 magnification.
-SMA is an actin isoform specific to myofibroblasts and its expression undergoes characteristic changes during EMT. We monitored the changes in -SMA expression using immunoblotting. -SMA in MDCK cells was detected at a low level in basal state and was remarkably induced (3.6-fold) following a 3-day treatment with TGF- 1. PGE 2 in the dose range of 10-1,000 nM inhibited TGF- 1 -induced -SMA expression in a dose-dependent manner. The -SMA induction was significantly inhibited by 100 nM PGE 2 and was completely abolished by 1,000 nM PGE 2 ( Fig. 6 ).
Fig. 6. Effect of PGE 2 on TGF- 1 -induced -SMA protein expression in MDCK cells. The cells were grown on 6-well plates until 80% confluence and then treated with vehicle or TGF- 1 in the presence or absence of various concentrations of PGE 2. Top : densitometric analysis of 3 independent experiments. Bottom : representative immunoblots for -SMA. * P < 0.01 vs. TGF- 1 -stimulated cells.
Effects of SC-58635 on HGF-induced inhibition of EMT. Given the observations that HGF stimulated COX-2 expression and PGE 2 release, and exogenous PGE 2 blocked EMT induced by TGF- 1, we speculate that COX-2 may mediate the action of HGF in MDCK cells. Thus we examined whether a COX-2-selective inhibitor, SC-58635, would prevent the antifibrotic effect of HGF. We found that exposure of MDCK cells to 50 ng/ml recombinant human HGF for 3 days reversed TGF- 1 -induced EMT as evidenced by restoration of epithelial phenotype ( Fig. 4 D ) and E-cadherin expression ( Fig. 5 D ) and inhibition of -SMA expression ( Fig. 7 ). However, in the presence of a COX-2-specific inhibitor, SC-58635, HGF-induced inhibition of EMT was partially but significantly blocked ( Fig. 4 E for morphology, Fig. 5 E for E-cadherin immunostaining, and Fig. 7 for -SMA expression).
Fig. 7. SC-58635 attenuated HGF-induced inhibition of -SMA expression. MDCK cells were grown in 6-well plates until 80% confluence and then pretreated with 10 µM SC-58635 for 30 min, followed by treatment with 50 ng/ml HGF for 12 h. The cells were treated with 10 ng/ml TGF- 1 for an additional 3 days. -SMA expression was determined by immunoblotting. Top : densitometric analysis of -SMA expression. Bottom : representative immunoblots. * P < 0.01 vs. controls. ** P < 0.01 vs. TGF- 1 -stimulated cells. P < 0.01 vs. TGF- 1 plus HGF-stimulated cells.
Mechanisms of the inhibitory effect of PGE 2 on EMT. ROS have been shown to mediate TGF- 1 -induced EMT, thus we tested whether PGE 2 inhibited TGF- 1 -induced ROS production. Cellular ROS increased threefold 30 min following TGF- 1 treatment. This increase in ROS levels was completely blocked by 1 µM PGE 2 ( Fig. 8 ).
Fig. 8. Effect of PGE 2 on TGF- 1 -induced reactive oxygen species (ROS) generation. Confluent MDCK cells in chamber slides were pretreated with 1 µM PGE 2 for 30 min and then exposed to TGF- 1 for further 30 min in the presence of DCFDA. A : control. B : 10 ng/ml TGF- 1. C : 10 ng/ml TGF- 1 plus 1 µM PGE 2. For quantification of ROS production, confluent MDCK cells in 96-well plates were pretreated with 1 µM PGE 2 for 30 min and then stimulated by 10 ng/ml TGF- 1 for further 30 min in the presence of DCFDA. Fluorescence was quantified using FLUOstar OPTIMA ( D ). Values represent means ± SE, n = 8. * P < 0.01 vs. control. P < 0.01 vs. TGF- 1 group.
cAMP processes antioxidant properties and is a major signaling mediator of PGE 2. We hypothesize that PGE 2 may inhibit TGF- 1 -induced ROS production via cAMP. As shown in Fig. 9, PGE 2 significantly increased intracellular cAMP levels. To examine whether increases of cAMP levels will inhibit ROS generation, we examined effects of a cell membrane-permeable cAMP analog, 8-Br-cAMP (50 µM), an adenylate cyclase activator, forskolin (50 µM), and a phosphodiesterases inhibitor, IBMX (10 µM), on TGF- 1 -induced oxidant production. As shown in Fig. 10, all three cAMP-elevating agents were able to block the rise of ROS in response to TGF- 1.
Fig. 9. Effect of PGE 2 on cAMP production. MDCK cells were grown to confluence and serum-starved for 24 h and then pretreated with 10 µM IBMX for 30 min, followed by incubation with serum-free media alone or 1 µM PGE 2. Cells were harvested in 0.1 M HCl and supernatants were assayed for cAMP levels by ELISA. Values represent means ± SE, n = 6. * P < 0.01 vs. control.
Fig. 10. Effects of cAMP-elevating agents on ROS production. Confluent MDCK cells in chamber slides were pretreated for 30 min with 8-Br-cAMP, forskolin, and IBMX, incubated for a further 30 min with TGF- 1 in the presence of DCFDA. A : control. B : 10 ng/ml TGF- 1. C : 10 ng/ml TGF- 1 plus 50 µM 8-Br-cAMP. D : 10 ng/ml TGF- 1 plus 50 µM forskolin. E : 10 ng/ml TGF- 1 plus 10 µM IBMX. For quantification of ROS production, confluent MDCK cells in 96-well plates were pretreated for 30 min with 8-Br-cAMP, forskolin, and IBMX, incubated for a further 30 min with TGF- 1 in the presence of DCFDA. Fluorescence was quantified using FLUOstar OPTIMA ( F ). Values represent means ± SE, n = 8. * P < 0.01 vs. control. P < 0.01 vs. TGF- 1 group.
We next tested whether PGE 2 blocked EMT via cAMP. MDCK cells were pretreated by Rp-cAMP, an cAMP analog, or H89, a protein kinase A (PKA) inhitibor, for 30 min, then incubated with TGF- 1 and PGE 2. Unexpectedly, neither Rp-cAMP nor H89 blocked the effect of PGE 2 ( Fig. 11 ).
Fig. 11. Effect of Rp-cAMP and H89 on the epithelial-to-mesenchymal transition (EMT) induced by TGF- 1. A - E : morphologic changes in MDCK cells. The cells were grown in 6-well plates until 80% confluence and then treated with vehicle ( A ), TGF- 1 alone ( B ), TGF- 1 plus 1 µM PGE 2 ( C ), TGF- 1 plus 1 µM PGE 2 and 200 µM Rp-cAMP ( D ), or TGF- 1 plus 1 µM PGE 2 and 10 µM H89 ( E ) for 3 days. Photographs were taken using a Nikon microscope (phase contrast). F : -SMA expression. MDCK cells were treated as abovementioned, and -SMA expression was determined by immunoblotting. Top : representative immunoblots. Bottom : densitometric analysis of -SMA expression. * P < 0.01 vs. TGF- 1 -stimulated cells.
DISCUSSION
EMT has emerged as a critical event in the pathogenesis of TIF. Knowledge concerning the regulatory mechanism of EMT may lead to the development of effective therapies to halt the progression of ESRD. The identification of a limited number of negative regulators of EMT, namely HGF and bone morphogenic protein 7 ( 47 ), has already shed light on the therapeutic interventions of the disease process. The present study contributes to the identification of a novel class of EMT inhibitors that are a lipid mediator, PGE 2.
MDCK cells are one of the best-characterized renal epithelial cells and have been used by a number of studies to investigate various aspects of EMT ( 6, 15, 34 ). Exposure of MDCK cells to TGF- 1 for 3 days induced a complete conversion of the epithelial cells to myofibroblasts as evidenced by acquisition of spindle-like morphology, loss of E-cadherin, and activation of -SMA. The time frame required for this transition was similar to previous reports ( 44 ). PGE 2 exhibited a remarkable inhibitory effect on TGF- 1 -induced EMT. This inhibition was complete as evidenced by a full restoration of epithelial morphology and E-cadherin expression and a complete abolishment of -SMA stimulation. These effects were observed with PGE 2 at the range of 10 to 1,000 nM which is likely to be within the range of physiological concentrations of PGE 2 in the kidney. These observations strongly suggest that PGE 2 is a potent inhibitor of TGF- 1 -induced EMT.
What would be the signaling mechanism responsible for PGE 2 -elicited inhibitory effects on EMT? ROS have an established role in the pathogenesis of chronic renal disease. Two recent studies have made a landmark observation that ROS mediates EMT in both cancer cells and renal proximal tubule cells ( 33, 36 ). Therefore, we examined the possibility that PGE 2 may inhibit EMT via reduction of ROS production. We found that TGF- 1 at 10 ng/ml induced a threefold increase in DCF-sensitive cellular ROS, confirming the report by Rhyu et al. ( 36 ). To our surprise, 1 µM PGE 2 completely blocked TGF- 1 -induced ROS production, thus establishing ROS as a target of PGE 2 during EMT.
cAMP processes antioxidant properties and is a major signaling mediator of PGE 2. We examined the potential role of cAMP as a proximal mediator of the signaling cascade elicited by PGE 2. Indeed, all cAMP-elevating agents tested (8-Br-cAMP, forskolin, and IBMX) mimicked the effect of PGE 2 in inhibiting oxidative stress and EMT. Furthermore, PGE 2 treatment was able to elevate intracellular cAMP level. These findings favor the notion that PGE 2 may act via cAMP. However, inhibition of the activity of endogenous cAMP with either Rp-cAMP or H89 failed to block the effect of PGE 2, almost ruling out involvement of cAMP. One explanation is that the production of endogenous cAMP induced by PGE 2 may not reach the level achieved with cAMP-elevating agents for effective inhibition of EMT. It is possible that cAMP pathway may be utilized by other antifibrotic factors with more effective cAMP-elevating capability.
HGF is a novel antifibrotic cytokine that blocks TGF- 1 -induced EMT both in vitro and in vivo. We sought to examine the possibility that the antifibrotic effect of HGF might be mediated by COX-2-derived PGE 2. In the present study, we demonstrated that HGF transiently and significantly induced COX-2 expression that was paralleled by significant stimulation of PGE 2 release. In sharp contrast, exposure of MDCK cells to several other growth factors including EGF, TGF- 1, IGF-1, and PDGF-BB, had no effect on either COX-2 expression or PGE 2 release. These findings demonstrate that the induction of COX-2 in MDCK cells is specific to HGF but not other growth factors tested. In complete agreement with this finding, the inhibition of TGF- 1 -induced EMT in cultured tubular epithelial HK cells is observed only with HGF, but not EGF, PDGF, IGF-1, and IL-6 ( 44 ). These findings indicate a specific coupling of the COX-2-PGE 2 pathway with HGF in MDCK cells. We further demonstrated that in the presence of a COX-2-specific inhibitor, SC-58635, the antifibrotic effect of HGF was blocked significantly although not completely, indicating some dependence of HGF action on COX-2 products.
COX-2 is an inducible form of cyclooxygenase and has been characterized as an immediate early responsive gene that can be induced by a wide variety of growth factors, tumor promoters, and cytokines, mostly in cancer cells and inflammatory cells ( 7, 12, 38 ). In contrast, COX-2 expression in MDCK cells is induced by HGF but not other growth factors examined. In line with this observation, interleukin-1, TNF-, and phorbol 12-myristate 13-acetate (PMA) do not have any effect on COX-2 expression in MDCK cells ( 31 ). The response of COX-2 to a selective growth factor HGF in MDCK cells suggests distinct functions. As demonstrated by the present study, this phenomenon is linked to the antifibrotic function of HGF, which is not shared with other growth factors.
The signaling pathway involved in the HGF induction of COX-2 expression was examined with emphasis on MAP kinase. Exposure of MDCK cells to HGF induced a rapid increase in phosphorylation of ERK1/2 that preceded the COX-2 induction. An ERK1/2 inhibitor, U-0126 (10 µM), completely blocked the COX-2 stimulation. It is evident that ERK1/2 is required for the induction of COX-2 by HGF in cultured MDCK cells. In line with this observation, MAP kinases including ERK1/2 mediate COX-2 induction in renal epithelial cells following exposure to diverse stimuli, such as hypertonicity ( 45 ) and low chloride ( 46 ). Taken together, these observations consolidate the conclusion that MAP kinases serve as a common downstream mediator for the stimulation of COX-2 expression. However, given the quite universal phenomenon of activation of ERK1/2 by different growth factors, the specificity of signaling pathway involved in HGF activation of COX-2 is considered to be conferred by other mediators in the pathway.
Given the ability of PGE 2 to inhibit TGF- 1 -induced EMT, it is reasonable to speculate that PGE 2 will exhibit beneficial effects on renal injury associated with chronic renal disease. In a general agreement with this notion, a large number of early studies demonstrate that PGE or prostacyclin exerts beneficial actions to limit ischemic or toxic renal injury ( 19, 21, 29, 30, 40, 42 ). The beneficial actions may relate to the improved renal perfusion ( 29 ). It has also been shown that in cultured renal tubular epithelial cells PGE 2 and its analog 11-deoxy-16,16-dimethyl PGE 2 exert a direct cytoprotective action ( 17, 41 ). On the other hand, inhibition of PG synthesis with nonsteroid anti-inflammatory drugs is associated with various types of nephrotoxicities, including interstitial nephritis ( 2, 44 ). Furthermore, COX-2-deficient mice develop severe renal pathologies and progressive renal failure ( 8, 28 ).
In summary, the present study documents a counterregulatory role of the COX-2-derived PGE 2 in EMT through inhibition of ROS production ( Fig. 12 ). This pathway partially mediates the antifibrotic action of HGF. These observations are expected to shed new light on a better understanding of the lipid mediators in chronic renal disease.
Fig. 12. Schematic illustration of the counterregulatory role of PGE 2 in EMT. HGF induces COX-2 expression, leading to the release of PGE 2. PGE 2 blocks ROS generation and thereby inhibits EMT.
GRANTS
This work was supported by National Institutes of Health Grants RO-1-HL-079453, RO-1-DK-066592, and KO-1-DK-064981 (to T. Yang).
【参考文献】
Azuma H, Takahara S, Matsumoto K, Ichimaru N, Wang JD, Moriyama T, Waaga AM, Kitamura M, Otsuki Y, Okuyama A, Katsuoka Y, Chandraker A, Sayegh MH, and Nakamura T. Hepatocyte growth factor prevents the development of chronic allograft nephropathy in rats. J Am Soc Nephrol 12: 1280-1292, 2001.
Bennett WM, Henrich WL, and Stoff JS. The renal effects of nonsteroidal anti-inflammatory drugs: summary and recommendations. Am J Kidney Dis 28: S56-S62, 1996.
Birchmeier C, Birchmeier W, and Brand-Saberi B. Epithelial-mesenchymal transitions in cancer progression. Acta Anat (Basel) 156: 217-226, 1996.
Border WA and Noble NA. TGF-beta in kidney fibrosis: a target for gene therapy. Kidney Int 51: 1388-1396, 1997.
Bottinger EP and Bitzer M. TGF-beta signaling in renal disease. J Am Soc Nephrol 13: 2600-2610, 2002.
Carrozzino F, Soulie P, Huber D, Mensi N, Orci L, Cano A, Feraille E, and Montesano R. Inducible expression of snail selectively increases paracellular ion permeability and differentiallymodulates tight junction proteins. Am J Physiol Cell Physiol 289: C1002-C1014, 2005.
Chou TC, Fu E, and Shen EC. Chitosan inhibits prostaglandin E2 formation and cyclooxygenase-2 induction in lipopolysaccharide-treated RAW 264.7 macrophages. Biochem Biophys Res Commun 308: 403-407, 2003.
Dinchuk JE, Car BD, Focht RJ, Johnston JJ, Jaffee BD, Covington MB, Contel NR, Eng VM, Collins RJ, Czerniak PM, Gorry SA, and Trzaskos JM. Renal abnormalities and an altered inflammatory response in mice lacking cyclooxygenase II. Nature 378: 406-409, 1995.
Eddy AA. Molecular basis of renal fibrosis. Pediatr Nephrol 15: 290-301, 2000.
Eddy AA. Molecular insights into renal interstitial fibrosis. J Am Soc Nephrol 7: 2495-2508, 1996.
Elliott BE, Hung WL, Boag AH, and Tuck AB. The role of hepatocyte growth factor (scatter factor) in epithelial-mesenchymal transition and breast cancer. Can J Physiol Pharmacol 80: 91-102, 2002.
Feng L, Sun W, Xia Y, Tang WW, Chanmugam P, Soyoola E, Wilson CB, and Hwang D. Cloning two isoforms of rat cyclooxygenase: differential regulation of their expression. Arch Biochem Biophys 307: 361-368, 1993.
Fine A and Goldstein RH. The effect of PGE 2 on the activation of quiescent lung fibroblasts. Prostaglandins 33: 903-913, 1987.
Goldstein RH and Polgar P. The effect and interaction of bradykinin and prostaglandins on protein and collagen production by lung fibroblasts. J Biol Chem 257: 8630-8633, 1982.
Hay ED and Zuk A. Transformations between epithelium and mesenchyme: normal, pathological, and experimentally induced. Am J Kidney Dis 26: 678-690, 1995.
Hodges RJ, Jenkins RG, Wheeler-Jones CP, Copeman DM, Bottoms SE, Bellingan GJ, Nanthakumar CB, Laurent GJ, Hart SL, Foster ML, and McAnulty RJ. Severity of lung injury in cyclooxygenase-2-deficient mice is dependent on reduced prostaglandin E(2) production. Am J Pathol 165: 1663-1676, 2004.
Jia Z, Person MD, Dong J, Shen J, Hensley SC, Stevens JL, Monks TJ, and Lau SS. Grp78 is essential for 11-deoxy-16,16-dimethyl PGE 2 -mediated cytoprotection in renal epithelial cells. Am J Physiol Renal Physiol 287: F1113-F1122, 2004.
Kalluri R and Neilson EG. Epithelial-mesenchymal transition and its implications for fibrosis. J Clin Invest 112: 1776-1784, 2003.
Kaufman RP Jr, Anner H, Kobzik L, Valeri CR, Shepro D, and Hechtman HB. Vasodilator prostaglandins (PG) prevent renal damage after ischemia. Ann Surg 205: 195-198, 1987.
Kolodsick JE, Peters-Golden M, Larios J, Toews GB, Thannickal VJ, and Moore BB. Prostaglandin E 2 inhibits fibroblast to myofibroblast transition via E. prostanoid receptor 2 signaling and cyclic adenosine monophosphate elevation. Am J Respir Cell Mol Biol 29: 537-544, 2003.
Koyama I, Neya K, Ueda K, and Omoto R. Protective effect of lipo-prostaglandin E1 on postischemic renal failure. Transplant Proc 19: 3542-3544, 1987.
Liu Y. Epithelial to mesenchymal transition in renal fibrogenesis: pathologic significance, molecular mechanism, and therapeutic intervention. J Am Soc Nephrol 15: 1-12, 2004.
McAnulty RJ, Chambers RC, and Laurent GJ. Regulation of fibroblast procollagen production. Transforming growth factor-beta 1 induces prostaglandin E 2 but not procollagen synthesis via a pertussis toxin-sensitive G-protein. Biochem J 307: 63-68, 1995.
McAnulty RJ, Hernandez-Rodriguez NA, Mutsaers SE, Coker RK, and Laurent GJ. Indomethacin suppresses the anti-proliferative effects of transforming growth factor-beta isoforms on fibroblast cell cultures. Biochem J 321: 639-643, 1997.
Mizuno S, Kurosawa T, Matsumoto K, Mizuno-Horikawa Y, Okamoto M, and Nakamura T. Hepatocyte growth factor prevents renal fibrosis and dysfunction in a mouse model of chronic renal disease. J Clin Invest 101: 1827-1834, 1998.
Mizuno S, Matsumoto K, Kurosawa T, Mizuno-Horikawa Y, and Nakamura T. Reciprocal balance of hepatocyte growth factor and transforming growth factor-beta 1 in renal fibrosis in mice. Kidney Int 57: 937-948, 2000.
Mizuno S, Matsumoto K, and Nakamura T. Hepatocyte growth factor suppresses interstitial fibrosis in a mouse model of obstructive nephropathy. Kidney Int 59: 1304-1314, 2001.
Morham SG, Langenbach R, Loftin CD, Tiano HF, Vouloumanos N, Jennette JC, Mahler JF, Kluckman KD, Ledford A, Lee CA, and Smithies O. Prostaglandin synthase 2 gene disruption causes severe renal pathology in the mouse. Cell 83: 473-482, 1995.
Neumayer HH, Wagner K, Groll J, Schudrowitsch L, Schultze G, and Molzahn M. Beneficial effects of long-term prostaglandin E 2 infusion on the course of postischemic acute renal failure. Long-term studies in chronically instrumented conscious dogs. Renal Physiol 8: 159-168, 1985.
Neumayer HH, Wagner K, Preuschof L, Stanke H, Schultze G, and Molzahn M. Amelioration of postischemic acute renal failure by prostacyclin analog (iloprost): long-term studies with chronically instrumented conscious dogs. J Cardiovasc Pharmacol 8: 785-790, 1986.
Ostrom RS, Gregorian C, Drenan RM, Gabot K, Rana BK, and Insel PA. Key role for constitutive cyclooxygenase-2 of MDCK cells in basal signaling and response to released ATP. Am J Physiol Cell Physiol 281: C524-C531, 2001.
Peters H, Noble NA, and Border WA. Transforming growth factor-beta in human glomerular injury. Curr Opin Nephrol Hypertens 6: 389-393, 1997.
Radisky DC, Levy DD, Littlepage LE, Liu H, Nelson CM, Fata JE, Leake D, Godden EL, Albertson DG, Nieto MA, Werb Z, and Bissell MJ. Rac1b and reactive oxygen species mediate MMP-3-induced EMT and genomic instability. Nature 436: 123-127, 2005.
Reichert M, Muller T, and Hunziker W. The PDZ domains of zonula occludens-1 induce an epithelial to mesenchymal transition of Madin-Darby canine kidney I cells. Evidence for a role of beta-catenin/Tcf/Lef signaling. J Biol Chem 275: 9492-9500, 2000.
Remuzzi G and Bertani T. Pathophysiology of progressive nephropathies. N Engl J Med 339: 1448-1456, 1998.
Rhyu DY, Yang Y, Ha H, Lee GT, Song JS, Uh ST, and Lee HB. Role of reactive oxygen species in TGF-beta1-induced mitogen-activated protein kinase activation and epithelial-mesenchymal transition in renal tubular epithelial cells. J Am Soc Nephrol 16: 667-675, 2005.
Savagner P, Yamada KM, and Thiery JP. The zinc-finger protein slug causes desmosome dissociation, an initial and necessary step for growth factor-induced epithelial-mesenchymal transition. J Cell Biol 137: 1403-1419, 1997.
Smith WL, Garavito RM, and DeWitt DL. Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. J Biol Chem 271: 33157-33160, 1996.
Stahl PJ and Felsen D. Transforming growth factor-beta, basement membrane, and epithelial-mesenchymal transdifferentiation: implications for fibrosis in kidney disease. Am J Pathol 159: 1187-1192, 2001.
Tobimatsu M, Konomi K, Saito S, and Tsumagari T. Protective effect of prostaglandin E 1 on ischemia-induced acute renal failure in dogs. Surgery 98: 45-53, 1985.
Weber TJ, Monks TJ, and Lau SS. PGE 2 -mediated cytoprotection in renal epithelial cells: evidence for a pharmacologically distinct receptor. Am J Physiol Renal Physiol 273: F507-F515, 1997.
Werb R, Clark WF, Lindsay RM, Jones EO, Turnbull DI, and Linton AL. Protective effect of prostaglandin [PGE 2 ] and in glycerol-induced acute renal failure in rats. Clin Sci Mol Med 55: 505-507, 1978.
Yang J, Dai C, and Liu Y. A novel mechanism by which hepatocyte growth factor blocks tubular epithelial to mesenchymal transition. J Am Soc Nephrol 16: 68-78, 2005.
Yang J and Liu Y. Blockage of tubular epithelial to myofibroblast transition by hepatocyte growth factor prevents renal interstitial fibrosis. J Am Soc Nephrol 13: 96-107, 2002.
Yang T, Huang Y, Heasley LE, Berl T, Schnermann JB, and Briggs JP. MAPK mediation of hypertonicity-stimulated cyclooxygenase-2 expression in renal medullary collecting duct cells. J Biol Chem 275: 23281-23286, 2000.
Yang T, Park JM, Arend L, Huang Y, Topaloglu R, Pasumarthy A, Praetorius H, Spring K, Briggs JP, and Schnermann J. Low chloride stimulation of prostaglandin E 2 release and cyclooxygenase-2 expression in a mouse macula densa cell line. J Biol Chem 275: 37922-37929, 2000.
Zeisberg M, Hanai J, Sugimoto H, Mammoto T, Charytan D, Strutz F, and Kalluri R. BMP-7 counteracts TGF-beta1-induced epithelial-to-mesenchymal transition and reverses chronic renal injury. Nat Med 9: 964-968, 2003.
作者单位：1 Division of Nephrology, University of Utah and Veterans Affairs Medical Center, Salt Lake City, Utah; and Departments of 2 Physiology and 3 Cellular Biology and Anatomy, Medical College of Georgia; and 4 Medical Research Service, Veterans Affairs Medical Center, Augusta, Georgia