点击显示 收起
【摘要】 It has been shown that store-operated Ca 2+ influx (SOC) plays critical roles in the activation of endothelial nitric oxide (NO) synthase (eNOS) and generation of NO in endothelial cells. Recent studies indicate stromal interaction molecule 1 (STIM1) is the molecule responsible for SOC activation following Ca 2+ depletion in the ER. Retinoic acids (RA) have beneficial effects in the treatment of renal diseases. The mechanism of the RA action is still largely unknown. In the current study, we used primary cultured rat mesangial cells to examine the effect of RA on SOC and STIM1. In these cells, BK caused concentration-dependent [Ca 2+ ] i mobilization. Treatment of the cells with RA, while it had no effect on the initial peak, reduced the plateau phase of BK-mediated [Ca 2+ ] i response, indicating the inhibition of SOC by RA. The level of STIM1 protein but not mRNA in RA-treated cells was significantly reduced. RA treatment did not affect TGF- -mediated gradual Ca 2+ influx which occurred by superoxide anion-mediated mechanism, indicating RA treatment specifically inhibited SOC in mesangial cells. RT-PCR and Western blot analysis demonstrated that eNOS was expressed in rat mesangial cells grown in media containing 11 and 30 but not 5.5 mM glucose. Downregulation of STIM1 protein and BK-induced SOC by RA treatment or STIM1 dsRNA were associated with abolished NO production. The 26S proteasome inhibitor lactacystin blocked the RA-mediated downregulation of BK-induced SOC, suggesting that ubiquitin-proteasome pathway may be involved in RA-mediated STIM1 protein downregulation in rat mesangial cells. Our data suggest that glucose-induced eNOS expression and NO production in mesangial cells may contribute to hyperfiltration in diabetes and RA may exert beneficial effects by downregulation of STIM1 and SOC in mesangial cells.
【关键词】 diabetes protein downregulation glomerular hyperfiltration proteasome lactacystin eNOS
DIABETES IS THE LEADING CAUSE of renal disease in humans with 30% of type 1 and 2 diabetic patients developing diabetic nephropathy and end-stage renal disease ( 2 ). The initial phase of diabetic nephropathy is characterized by an increase of glomerular filtration rate (GFR) or glomerular hyperfiltration, which may gradually progress to end-stage renal failure ( 13, 25, 42 ). In the diabetic patients who are destined to develop clinically significant diabetic nephropathy, the GFR is usually elevated by 20-40% with no renal histological alterations at the time of initial diagnosis ( 32 ). Glomerular hyperfiltration occurring in early diabetes in the absence of histological changes in the kidney indicates that abnormality of carbohydrate/lipid metabolism and/or altered renal hemodynamics in the diabetic state may be the factor(s) responsible for diabetic hyperfiltration ( 32 ). Animal studies demonstrate that elevated NO production by endothelial nitric oxide (NO) synthases (eNOS) or neuronal NOS (nNOS) is causally linked with diabetic renal hyperfiltration ( 15, 45 - 47 ). Shear stress and agonist-induced intracellular Ca 2+ ([Ca 2+ ] i ) mobilization plays an important role in the regulation of eNOS and nNOS activity. It has been shown that [Ca 2+ ] i mobilization, especially store-operated Ca 2+ influx (SOC), is required for agonist-mediated eNOS activation and NO production in vascular endothelial cells ( 16, 19 ). Although multiple molecules, particularly the transient receptor potential (TRP) family of ion channels, are implicated in SOC, the molecular identity of SOC remains a mystery. Recently, by screening of hundreds to thousands of genes using RNA interference (RNAi), stromal interaction molecule 1 (STIM1) and 2 (STIM2) have been identified by two independent groups to play critical roles in the regulation of SOC in a number of cell types ( 20, 33 ). Evidence indicates that STIM1 may function as a Ca 2+ sensor in the ER, i.e., STIM1 is initially located in the ER membrane while ER Ca 2+ store is full and translocated to the plasma membrane to activate SOC when ER Ca 2+ store is depleted ( 53 ).
Retinoic acid (RA) and other retinoids have beneficial effects in the treatment of diabetic and nondiabetic nephropathy ( 5, 11, 35, 49 ). In the anti-Thy1.1 nephropathy rat model, RA treatment has been shown to limit glomerular cell proliferation and renal damage by reduction of renal TGF- 1 and TGF receptor II expression ( 26 ). In a diabetic rat model, treatment with RA caused a drop in the urinary excretion of protein and albumin, although the effect did not reach significance ( 11 ). RA regulates the expression of multiple genes by binding and subsequent activation of RA receptors (RAR),, and and/or retinoid X receptors (RXR),, and ( 39 ). The binding of RA or other retinoids causes the dissociation or release of corepressors and recruitment of coactivators to prompt and facilitate gene transcription ( 39 ). It is speculated that the therapeutic effect of RA in nondiabetic renal disease animal models may be linked to downregulation of genes related to inflammation, cell proliferation, and fibrosis ( 28 ). The mechanism of RA protective effect on diabetic nephropathy is currently not known.
Mesangial cell tone plays an important role in the regulation of renal hemodynamics. Mesangial cells are also shown to be major contributors to glomerular mesangial matrix expansion and capillary basement membrane thickening with increased expression of the extracellular matrix (EM) components collagen IV, fibronectin, and laminin observed in advanced diabetic nephropathy ( 50 ). We previously showed that the effect of TGF- on EM is mediated by a sustained Ca 2+ influx and activation of calcineurin ( 10 ) in rat mesangial cells. Bradykinin (BK) has been shown to stimulate mesangial cell proliferation and EM accumulation ( 7, 43 ). Diabetes and high glucose in vitro diminish BK-stimulated [Ca 2+ ] i mobilization and contraction in mesangial cells ( 3, 29 ). In the current study, we examined the effect of RA on STIM1 expression, SOC, and NO production in rat mesangial cells. Our results suggest that eNOS expression induced by high glucose may be responsible for glomerular hyperfiltration and RA may exert beneficial effect on hyperfiltration in diabetic nephropathy by regulation of SOC.
EXPERIMENTAL PROCEDURES
Materials. Fura-2 AM, 4-amino-5-methylamino-2',7'-difluorofluorescein diacetate (DAF-FM DA), mouse anti- -tubulin monoclonal antibody, DMEM, RPMI 1640, and EDTA were purchased from Invitrogen (Carlsbad, CA). Mouse anti-STIM1 monoclonal antibody was from BD Bioscience. All other chemicals were from Sigma.
Cell culture. Rat mesangial cells were provided by Dr. H. E. Abboud (Department of Medicine, University of Texas Health Science Center San Antonio) and were originally cultured from glomeruli isolated by differential sieving as previously described ( 36 ). For experiments in the present studies, rat mesangial cells were cultured in RPMI 1640 medium containing 2,000 mg/l (11 mM) glucose unless otherwise specified, supplemented with 14% fetal bovine serum, and penicillin (50 U/ml)/streptomycin (50 µg/ml), at 37°C in a humidified 5% CO 2 atmosphere incubator. Mesangial cells were harvested with trypsin (0.05%)-EDTA (0.02%) when reached confluence (usually 48-h culture following passage).
Measurement of intracellular calcium. Rat mesangial cells suspended in HNG buffer containing 140 mM NaCl, 5 mM KCl, 1 mM CaCl 2, 1 mM MgCl 2, 1.2 mM KH 2 PO 4, 10 mM glucose, 10 mM HEPES, pH 7.4, were labeled with fura-2 AM (1 µM) at 37°C for 30 min with agitation at 50 rpm. Loaded cells were washed and resuspended in HNG and [Ca 2+ ] i mobilization in populations of mesangial cells was measured in a fluorometer (QM-6, Photon Technology International) using a cuvette with the temperature stabilized at 37°C through a connected water bath. The ratio of fluorescence excited at 340 and 380 nm with emission of 510 nm was recorded and used to index the change of [Ca 2+ ] i as previously reported ( 52 ).
For experiments involving measurement of [Ca 2+ ] i in Ca 2+ -free medium, following fura-2 loading the cells were washed and resuspended in HNG buffer containing 100 µM EGTA, instead of 1 mM Ca 2+, and immediately used the same procedure as described above to measure [Ca 2+ ] i.
Simultaneous measurement of [Ca 2+ ] i and NO production. Rat mesangial cells suspended in HNG buffer were double-labeled with the [Ca 2+ ] i indicator fura-2 AM (1 µM) and NO indicator DAF-FM DA (0.5 nM) at 37°C for 30 min in the dark in the presence of 100 µM L -arginine with agitation at 50 rpm. Double-labeled cells were washed and resuspended in HNG buffer containing 100 µM L -arginine. Alterations of [Ca 2+ ] i and cellular NO production were simultaneously measured using the multi-dye function of the QM-6 fluorometer. The ratio of fluorescence excited at 340 and 380 nm with emission of 510 nm was recorded and used to index the change of [Ca 2+ ] i and fluorescence excited at 488 nm with emission wavelength of 515 nm was used to detect NO production inside the cells. The validity of the use of DAF-FM fluorescence to index the production of cellular NO was confirmed with the specific NOS inhibitor N G -monomethyl- L -arginine ( L -NAME; 1 mM). The presence of L -NAME blocked increase in DAF-2 fluorescence and had no effect on fura-2 fluorescence ratio.
Measurement of STIM1 and eNOS mRNA. The cellular mRNA levels of STIM1 and eNOS genes were measured in primary cultured rat mesangial cells by RT-PCR as previously described ( 54 ). Total RNA was isolated from rat mesangial cells grown to confluence in RPMI 1640 medium containing different concentrations of glucose as specified in 100-mm dishes, using the SV total RNA isolation system by following the manufacturer?s instruction (Promega, Madison, WI). For RT-PCR, 2 µg of total RNA were digested with DNase I to eliminate any DNA contamination and then used to synthesize the first-strand cDNA using cDNA synthesis kit (Promega). PCR amplification of the cDNA of STIM1 and eNOS was used to measure STIM1 and eNOS mRNAs. Parallel amplification of 18S rRNA under the same condition was used as internal control. The PCR condition was 94°C, 5 min and 28 cycles of 94°C, 1 min; 56°C, 1 min; 72°C, 1 min, with a final extension of 10 min at 72°C. PCR products were separated on 1% agarose gel and analyzed. The accession number and amplified fragment size of each gene and the specific primer pairs used were: STIM1, XM_341896, 544 bp, 5'-AGCCTGCAGAGCAGTGTCCG-3' and 5'-GAAACTTCTTCCTGCCTG-3'; 18S rRNA, M11188, 728 bp, 5'-TAGAGGTGAAATTCTTGGAC-3' and 5'-GACTTAATCAACGCAAGCTT-3'; eNOS, NM_021838, 715 bp, 5'-TACTACTCCATCAGCTCCTC-3' and 5'-TCTGGGTGCGGATGCGGC-3'.
Additionally, the effect of RA on STIM1 mRNA levels was also measured with real-time RT-PCR using 18S rRNA as an internal standard. Real-time PCR was carried out using an Applied Biosystems 7500 Real-Time PCR System and 96-well MicroAmp optical plates. PCR conditions were 2 min 50°C, 10 min 95°C, 40 cycles consisting of 15 s 95°C, and 1 min 60°C. The Taq man primers of human STIM1 and 18S rRNA (Cat. nos. 4351372 and 4319413E, respectively), which have the same sequences as rat, were purchased from Applied Biosystem (Foster City, CA). Each sample was running in duplicate. Standard curves for 18S rRNA and STIM1 were performed in each experiment. The relative STIM1 mRNA levels of vehicle- and RA-treated samples were calculated using the 2 - C T protocol ( 21 ).
Western blot analysis of STIM1 and eNOS proteins. Rat mesangial cells grown to confluence (48 h after passage) in 100-mm plates in RPMI 1640 media with different concentrations of glucose (5.5, 11, or 30 mM) as indicated were resuspended in ice-cold RIPA buffer containing 50 mM Tris, 150 mM NaCl, 1% NP-40, 0.25% Na-deoxycholate, 1 mM EDTA, 1 mM PMSF, 1 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µg/ml pepstatin, 1 mM Na 3 VO 4, 1 mM NaF, pH 7.4, and kept on ice for 30 min. The mixture was centrifuged at 12,000 g for 15 min at 4°C and protein concentrations in the supernatant were determined by the BCA assay ( 37 ) using bovine serum albumin as standard and following the instruction of the company (Pierce, Rockford, IL). Proteins (50 µg) were separated by 8% SDS-PAGE and transferred to PVDF membranes. The membranes were incubated with blocking buffer (TBS, containing 0.1% Tween 20 and 5% nonfat dry milk) for 1 h at room temperature, followed by anti-GOK mice monoclonal antibody (1:250 dilution, BD Bioscience) or rabbit anti-eNOS polyclonal antibody (1:1,000 dilution, Cayman Chemical) overnight at 4°C. -Tubulin was used to control protein loading. The membranes were then washed and incubated with secondary antibody (1:5,000 dilution, Amersham Life Science) and STIM1 and eNOS proteins were detected by ECL following the instruction of company (Pierce).
Downregulation of STIM1 with RNAi. The annealed Stealth siRNA duplexes were purchased from Invitrogen. Rat mesangial cells were transfected with 10 nM of the equal combination of three siRNA duplexes (A-C below) using Lipofectamine RNAiMAX in 100-mm dishes according to the manufacturer?s instructions (Invitrogen). Sequences of the three sets of STIM1 siRNA duplexes used were: A: antisense 5'-UAGAGUAACGGUUCUGGAUAUAGGC-3'; sense 5'-GCCUAUAUCCAGAACCGUUACUCUA-3'; B: antisense 5'-AAUAGAGCCAUUAUC CUCCUCAGCC-3'; sense 5'-GGCUGAGGAGGAUAAUGGCUCUAUU-3'; C: antisense 5'-UUCGGUAGACUCUUCAGAAGUCGCC-3'; sense 5'-GGCGACUUCUGAAGAGUCUACCGAA-3'.
Control cells were treated in parallel with a mixture of scrambled siRNA. Cells were harvested at 48-h posttransfection. Total RNA was isolated from cell lysates by SV total RNA isolation system (Promega). The mRNA level of STIM1 was checked with RT-PCR to ensure the downregulation of STIM1 gene expression by RNAi.
Statistical analysis. Summarized results from multiple experiments were presented as means ± SE. Differences between vehicle- and RA-treated groups were tested by Student?s t -test. Significant differences were defined as P < 0.05.
RESULTS
BK is a vasoactive peptide involved in the regulation of renal hemodynamics. The effect of BK is mediated by the B 2 bradykinin receptor, which belongs to G protein-coupled receptor superfamily and plays important roles in pathogenesis of diabetic nephropathy ( 14 ). As shown in Fig. 1, in suspensions of fura-2 AM-loaded rat mesangial cells BK (10 -11 -10 -7 M) induced a concentration-dependent increase of [Ca 2+ ] i. At lower concentrations (<10 -10 M), BK caused a gradual and sustained increase in [Ca 2+ ] i, similar to the response of TGF- in these cells as we previously showed ( 10 ). At concentrations 10 -9 M, BK evoked a rapid rise of [Ca 2+ ] i followed by an elevated plateau phase, which is presumably reflected by the activation of capacitative or SOC. Maximal [Ca 2+ ] i responses were observed at 10 -7 M BK and this concentration was used in the subsequent experiments of this study.
Fig. 1. Bradykinin (BK) caused concentration-dependent [Ca 2+ ] i mobilization. BK (10 -11 -10 -7 M)-induced [Ca 2+ ] i mobilization was measured in suspensions of fura-2-loaded cells as described in EXPERIMENTAL PROCEDURES. The arrows indicated the addition of BK.
Pretreatment of the rat mesangial cells with RA (50 µM) for 48 h during culture significantly inhibited the plateau phases ( Fig. 2 D ) of the [Ca 2+ ] i signal, whereas it had no effect on the initial peak ( Fig. 2 C ) of the BK (100 nM) responses, suggesting that RA treatment selectively blocks SOC in mesangial cells. Experiments performed in Ca 2+ -free HNG further supported the block of SOC in RA-treated mesangial cells ( Fig. 3 ). In the absence of extracellular Ca 2+, BK (100 nM) caused a transient increase of [Ca 2+ ] i followed by rapid decay to baseline, which reflected IP 3 -mediated Ca 2+ release from the ER and ER Ca 2+ store depletion. Subsequent addition of Ca 2+ (2 mM) in the medium triggered SOC-mediated Ca 2+ influx ( Fig. 3 B ). Pretreatment of the cells with RA during culture had no effect on BK-induced transient increase of [Ca 2+ ] i, but largely inhibited SOC-mediated Ca 2+ influx following Ca 2+ repletion in the medium ( Fig. 3, B and C ). The remaining Ca 2+ influx in RA-treated cells observed after medium Ca 2+ replenishment is likely to be mediated by non-SOC Ca 2+ influx, which is also observed in control cells briefly exposed to Ca 2+ -free HNG without BK stimulation ( Fig. 3, A and C ). Since brief exposure to Ca 2+ free HNG does not deplete ER Ca 2+ store, and thus these Ca 2+ influx belongs to non-SOC. These experiments indicate that RA selectively inhibits SOC in rat mesangial cells.
Fig. 2. Effect of retinoic acid (RA) treatment on BK-induced [Ca 2+ ] i mobilization. BK (100 nM)-induced [Ca 2+ ] i mobilization in cells treated with vehicle (control; A ) or RA (50 µM, 48 h; B ) was measured as described in EXPERIMENTAL PROCEDURES. C and D : means ± SE of the summarized results of the effect of RA on the peak and the plateau of BK responses from 7 experiments. The methods for calculation of peak and plateau values are depicted in A. * P < 0.05 control vs. RA.
Fig. 3. Effect of RA treatment on BK-induced [Ca 2+ ] i mobilization in the absence of extracellular Ca 2+. BK (100 nM)-induced [Ca 2+ ] i mobilization was measured in cells treated with vehicle (control; A and B ) or RA (50 µM, 48 h; C ) in the absence of extracellular Ca 2+ as described in EXPERIMENTAL PROCEDURES.
Since STIM1 proteins have been shown to be a potential ER Ca 2+ sensor required for SOC, we investigated the effect of RA treatment on the protein levels of STIM1. Immunoblotting analysis of the STIM1 proteins indicated that the expression of STIM1 proteins in RA-treated (50 µM, 48 h) mesangial cells was decreased ( Fig. 4 A; * P < 0.05 control vs. RA, n = 14). However, measurement of STIM1 mRNA with RT-PCR ( Fig. 4 B ) and real-time RT-PCR ( Fig. 4 C ) demonstrated that RA treatment had no effect on STIM1 mRNA levels in mesangial cells ( Fig. 4 C, P = 0.46 control vs. RA, n = 3), indicating that reduced STIM1 proteins in RA-treated cells were not resulted from a reduction in STIM1 mRNA. Further experiments demonstrated that the effect of RA treatment on STIM1 protein and BK-induced activation of SOC was dependent on the concentrations of RA and time of incubation ( Fig. 5 ). As shown in Fig. 5, A and C, both the BK-induced SOC as indexed by the amplitude of the plateau phase of BK-mediated [Ca 2+ ] i response and STIM1 proteins were reduced by treatment with increasing concentrations of RA. The time dependence of concurrent RA effects on BK-induced SOC and STIM1 was demonstrated in Fig. 5, B and D.
Fig. 4. Effect of RA on STIM1 expression. A : Western blot analysis of STIM1 protein in vehicle (control)- or RA (50 µM, 48 h)-treated cells. Left : representative experiment. Right : value is means ± SE of the intensity ratio of STIM1 to -tubulin bands from 14 experiments. * P < 0.05 control vs. RA. B and C : STIM1 mRNA was determined by RT-PCR ( B ) and real-time RT-PCR ( C ) as described in EXPERIMENTAL PROCEDURES.
Fig. 5. Concentration and time dependence of RA effect on the plateau of BK-induced [Ca 2+ ] i mobilization and STIM1 protein. Mesangial cells were treated with increasing concentrations (0-50 µM as indicated) of RA for 48 h ( A and C ) or treated with 50 µM RA for different times ( B and D ) during culture. A and B : changes in the plateau of BK-induced [Ca 2+ ] i mobilization relative to control (100%). The results were average of 2 experiments. C and D : STIM1 protein in samples treated with different concentrations of RA for 48 h or with 50 µM RA for different times as indicated.
It has been previously shown that TGF- caused a gradual and sustained Ca 2+ influx, which leads to activation of calcineuron and subsequent downstream signaling pathways in rat mesangial cells ( 10, 23 ). Both TGF- and BK have been demonstrated to play critical roles in the pathogenesis and progression of diabetic nephropathy ( 14, 22, 43, 55 ). It is interesting to compare the effect of RA on the [Ca 2+ ] i responses to BK and TGF-, which may provide insight into the mechanisms of pathogenesis of the disease. As shown in Fig. 6 ( B compared with A ), RA (50 µM, 48 h) treatment led to reduced plateau phase of BK-mediated [Ca 2+ ] i response and inhibition of SOC. In contrast, the same treatment had no effect on TGF- -mediated [Ca 2+ ] i response ( Fig. 6, C and D ). In vehicle (control)- or RA-treated (50 µM, 48 h) mesangial cells, TGF- (10 ng/ml) induced similar gradual and sustained [Ca 2+ ] i responses while the plateau phase of BK-mediated [Ca 2+ ] i responses is abolished. These results indicate that RA specifically affects BK-induced SOC but not TGF- -mediated Ca 2+ influx in rat mesangial cells, and thus inhibition of BK-induced SOC by RA is pathway specific and is not a result of a global downregulation of Ca 2+ transport pathways in rat mesangial cells.
Fig. 6. Effect of RA treatment on TGF- -induced [Ca 2+ ] i mobilization. BK (100 nM; A and B )- and TGF- (10 ng/ml; C and D )-induced [Ca 2+ ] i mobilization in cells treated with vehicle (control; A and C ) or RA (50 µM, 48 h; B and D ) were measured as described in EXPERIMENTAL PROCEDURES. The arrows indicated the addition of BK ( A and B ) or TGF- ( C and D ).
SOC-mediated Ca 2+ influx may regulate NO, which is a potential contributor to diabetic nephropathy. In endothelial cells, Ca 2+ influx via SOC has been shown to be required for Ca 2+ -dependent eNOS activation ( 16, 19 ). We next examined whether SOC activation also led to NO generations in mesnagial cells. We took the advantage that RA treatment selectively downregulated STIM1 and SOC in rat mesangial cells ( Figs. 2 - 5 ), which should result in the block of eNOS activation and NO production by BK. As shown in Fig. 7, in fura-2 and DAF-FM doublelabeled rat mesangial cells, BK (100 nM) induced simultaneous [Ca 2+ ] i mobilization (solid traces) and NO generation (dashed traces; Fig. 7 A ). Treatment of the cells with RA (50 µM, 48 h) reduced the plateau phase of the BK-mediated [Ca 2+ ] i response and blocked NO production ( Fig. 7 B ). Inclusion of N G -monomethyl- L -arginine (1 mM), a general NOS inhibitor, completely blocked BK-induced increase in DAF-FM fluorescence, e.g., NO generation, while it had no effect on BK-mediated [Ca 2+ ] i response ( Fig. 7 C ), indicating the specificity of DAF-FM to NO. The dependence of BK-induced NO production on SOC was further demonstrated in the experiments shown in Fig. 7 D. In the absence of extracellular Ca 2+, BK caused a transient increase of [Ca 2+ ] i without a sustained plateau phase and NO generation. Subsequent addition of Ca 2+ (2 mM) in the medium triggered SOC, which was accompanied by NO production ( Fig. 7 D ). Taken together, these experiments suggested that NO production in mesangial cells was activated by BK-mediated [Ca 2+ ] i mobilization and downregulation of STIM1 protein expression and SOC by RA was associated with eliminated NO generation in mesangial cells.
Fig. 7. BK-induced [Ca 2+ ] i mobilization and nitric oxide (NO) production. Mesangial cells were double labeled with fura-2 AM (1 µM) and DAF-2 DA (0.5 nM) and BK (100 nM)-induced [Ca 2+ ] i mobilization (solid traces) and NO production (dashed traces) were measured as described in EXPERIMENTAL PROCEDURES. A : BK (100 nM) responses in control cells. B : BK (100 nM) responses in the presence of L -NAME (1 mM). C : BK (100 nM) responses in cells treated with RA (50 µM, 48 h). D : BK (100nM)-induced [Ca 2+ ] i mobilization and NO production in control cells in the absence of extracellular Ca 2+ and following Ca 2+ repletion in the media as indicated. The arrows in A - C indicated the addition of BK. In D, the arrows indicated sequential additions of BK and 2 mM Ca 2+.
The role of STIM1 in BK-induced SOC and NO production was further supported by the data from STIM1 RNAi experiments ( Fig. 8 ). As indicated by the results of RT-PCR ( Fig. 8 A ), treatment of mesangial cells with STIM1 dsRNA largely reduced the STIM1 mRNA levels. Simultaneous measurement of BK-induced [Ca 2+ ] i mobilization (solid traces) and NO production (dashed traces) demonstrated that STIM1 RNAi abolished the elevated plateau of [Ca 2+ ] i response and NO production ( Fig. 8, C compared with B ). To ensure the inhibition of NO production by STIM1 dsRNA did not result from an off-target inhibition of eNOS expression, we measured and found no difference in the eNOS mRNA levels in control and STIM1 dsRNA-treated mesangial cells (data not shown).
Fig. 8. Effect of STIM1 gene silence with siRNA on the mRNA level of STIM1 and BK-induced [Ca 2+ ] i mobilization and NO generation. A : STIM1 mRNA levels were measured with RT-PCR in cells treated with scramble (control) or STIM1 dsRNA (RNAi) as described in EXPERIMENTAL PROCEDURES. B and C : mesangial cells treated with scramble ( B ) or STIM1 dsRNA ( C ) were double labeled with fura-2 AM (1 µM) and DAF-FM DA (0.5 nM) and BK (100 nM)-induced [Ca 2+ ] i mobilization (solid traces) and NO production (dashed traces) were measured as described in EXPERIMENTAL PROCEDURES. The arrows indicated sequential additions of arginine (100 µM) and BK (100 nM).
Among the three isoforms of NOS, eNOS and nNOS are known to be regulated by Ca 2+. It is currently unclear which isoforms of the NOS are expressed in rat mesangial cells. We next determined the expression of eNOS and nNOS in mesangial cells. RT-PCR and Western blot analysis indicated that eNOS mRNA and protein were expressed in rat mesangial cells, and the levels of eNOS mRNA and protein were dependent on the concentrations of glucose in the RPMI 1640 media during culture ( Fig. 9, A and B ). When the cells were cultured in RPMI 1640 media containing 11 or 30 mM glucose, the expression of eNOS mRNAs and proteins was detected by RT-PCR and Western blot analysis, respectively ( Fig. 9 ). In cells cultured in RPMI1640 media containing 5.5 mM (1,000 mg/l) glucose, eNOS mRNA and protein were barely detectable with RT-PCR and Western blot analysis ( Fig. 9 ). Regardless of the glucose levels during culture, nNOS mRNA was not detected with RT-PCR ( Fig. 9 A ). Additionally, treatment of mesangial cells grown in 11 mM glucose media with RA (50 µM, 48 h) had no effect on the eNOS protein levels ( Fig. 9 C ).
Fig. 9. Detection of eNOS expression. A : eNOS mRNA levels in rat mesangial cells grown in RPMI 1640 containing 5.5, 11, and 30 mM glucose, respectively, were analyzed with RT-PCR as described in EXPERIMENTAL PROCEDURES. B and C : Western blot analysis of eNOS protein in cells grown in RPMI 1640 containing different concentrations of glucose as indicated ( B ), and the effect of RA on eNOS protein ( C ).
It is well established that ubiquitin-proteasome pathway plays important roles in posttranslational protein modification and degradation, which contributes to the accurate maintenance of dynamic protein levels responding to altered physiological conditions ( 30 ). Containing of the PEST region ( Fig. 10, top ) may target STIM1 to the degradation by proteasomes. As shown in Fig. 10, A and B, treatment of rat mesangial cells with RA (50 µM, 48 h) reduced the plateau phase, i.e., the SOC of BK-induced [Ca 2+ ] i responses. However, in the presence of lactacystin (10 µM) the same RA treatment showed no effect on the plateau of BK-mediated [Ca 2+ ] i responses ( Fig. 10, A - D ), indicating that proteasome pathways were involved in the RA inhibition of BK-mediated SOC activation. On the other hand, treatment of the cells with lactacystin, RA, or combination of the two reagents had no effect on the gradual [Ca 2+ ] i responses activated by TGF- ( Fig. 10, E-H ).
Fig. 10. Effects of lactacystin on BK-induced SOC in control and RA-treated cells. Rat mesangial cells were either treated with vehicle ( A, C, E, and G ) or RA (50 µM; B, D, F, and H ) for 48 h in the absence ( A, B, E, and F ) or presence ( C, D, G, and H ) of lactacystin (10 µM) and [Ca 2+ ] i responses to BK (100 nM; A - D ) and TGF- (10 ng/ml; E - F ) were measured as described in EXPERIMENTAL PROCEDURES. The arrows under each trace indicated the addition of BK ( A - D ) or TGF- ( E - F ).
DISCUSSION
Elevated NO production by neuronal and/or endothelial nitric oxide synthase (nNOS/eNOS) but not inducible NOS (iNOS) has been shown to be a major factor responsible for the pathogenesis of glomerular hyperfiltration in early diabetic nephropathy ( 15, 45 - 47 ). Increased evidence from animal models indicates that the BK B 2 receptor is critically involved in diabetic nephropathy ( 14 ). B 2 receptors are detected in multiple cell types in the kidney, including mesangial and endothelial cells ( 38 ). Disruption of B 2 receptor expression markedly enhances nephropathy in diabetic mice ( 14 ). Although it has been the focus of intensive investigations, the effect of hyperglycemia on eNOS expression and NO production in renal cells remains inconclusive ( 15 ). In vitro experiments using high glucose ( 30 mM) in cultured cells to assess the effect of diabetes on NO production generally suggested a reduction of renal NO production ( 15 ). Most in vivo evidence suggests increased eNOS expression and/or NO production in the renal cortex in diabetes, which may be responsible for the glomerular hyperfiltration and hypertrophy observed in early diabetes ( 15 ). Under normal physiological conditions, eNOS is detected in the afferent/efferent vascular and glomerular endothelial and epithelial cells but not mesangial cells in human and animal kidneys ( 1, 9 ). Based on in vitro experiments with cultured endothelial cells, it is unlikely that the eNOS in vascular or glomerular endothelial cells is responsible for the excess NO production and glomerular hyperfiltration in early diabetes since hyperglycemia reduced the production of NO in these cells ( 12, 41, 44 ). Even though previous evidence supported that NO production by eNOS/nNOS may cause hyperfiltration in early diabetes, the exact cell types in the kidney from which the elevated NO generation lead to hyperfiltration are still not defined. Our present study demonstrates that rat mesangial cells cultured under moderate (11 mM) and high (30 mM) but not normal (5.5 mM) glucose expressed eNOS mRNA and protein ( Fig. 9 ). Furthermore, the expressed eNOS in mesangial cells is activated by BK-mediated [Ca 2+ ] i mobilization, e.g., SOC ( Fig. 7 ). Specific downregulation of STIM1 with RNAi, while it had no effect on the eNOS protein levels, eliminated SOC and NO generation following BK stimulation ( Fig. 8 ). These results suggest that in response to increased glucose levels, eNOS expression may be induced and NO production may be activated by Ca 2+ -dependent mechanisms in rat mesangial cells, which could become a potential NO generation site under hyperglycemic conditions. Our results therefore provide novel evidence that diabetic glomerular hyperfiltration may be caused by eNOS-mediated NO production in mesangial cells.
Protein level and activity of eNOS are two factors among others to affect NO production in mesangial cells. The Ca 2+ -calmodulin-dependent activation of eNOS is mediated by SOC in endothelial cells ( 16, 19 ). As we demonstrated in the current study, SOC also plays a critical role in the activation of BK-induced NO production in rat mesangial cells ( Figs. 7 and 8 ) and is an essential component of BK-mediated [Ca 2+ ] i mobilization ( Figs. 1 and 2 ). STIM1 and STIM2 have been identified to play critical roles in the regulation of SOC in a number of cell types ( 20, 33 ). Evidence indicates that STIM1 may function as a Ca 2+ sensor in the ER, i.e., STIM1 is initially located in the ER membrane as monomers while the ER Ca 2+ store is full and translocated to the plasma membrane to activate SOC when the ER Ca 2+ store is depleted or physically coupled to the SOC channel on the plasma membrane ( 40, 53 ). We demonstrated that STIM1 mRNAs and proteins were expressed in rat mesangial cells ( Fig. 4 ). RA as well as STIM1 RNAi treatment concurrently inhibited the expression of STIM1 and BK-induced SOC ( Figs. 2, 3, 4, and 8 ), suggesting that STIM1 is involved in agonist-mediated activation of SOC in mesangial cells. These data suggest that STIM1 may be a mediator for regulation of eNOS in mesangial cells and can serve as a potential therapeutic target for diabetic nephropathy.
In mesangial cells, [Ca 2+ ] i response may be induced by BK and other G protein-coupled receptor agonists, such as endothelin-1 (ET-1), as well as TGF-, PDGF, and other growth factors in mesangial cells ( 8, 10, 27 ). High glucose ( 30 mM) has been shown to inhibit ET-1- but not PDGF-mediated [Ca 2+ ] i response ( 8, 27 ). The reduced [Ca 2+ ] i response of ET-1 in high glucose was linked to PKC-dependent inactivation of PLC 3 ( 8 ). The effects of high glucose on the ER Ca 2+ stores and SOC are still inconclusive ( 8, 27 ). Whether high glucose affects BK-induced [Ca 2+ ] i signaling and how the effects contribute to RA regulation of STIM1 expression, SOC, and NO generation is currently unclear and requires further studies.
The therapeutic effect of RA in anti-glomerular basement membrane-induced nephritis seems linked to the downregulation of several genes related to inflammation, cell proliferation, and fibrosis ( 28 ). In a streptozotocin-induced diabetic rat nephropathy model, treatment with RA decreased proteinuria and urinary albumin/creatinine ratio, although the effect of RA did not reach significance ( 11 ). In the db/db type 2 diabetes mouse model, several synthetic RXR agonists, such as LG100268, AGN194204, LG100754, LGD1069, have been shown to reduce blood glucose ( 4, 17, 18, 24, 48 ). In cultured rat mesangial cells, RA treatment concurrently inhibited BK-induced activation of SOC and NO production ( Fig. 7 ). It is unlikely that the observed inhibition of SOC and NO production by RA treatment was resulted from a direct effect on the glucose concentrations in culture media. However, it is possible that RA treatment may alter glucose metabolism of mesangial cells, and which may lead to downregulation of STIM1 protein, inhibition of SOC, and reduced NO production. It is also possible that RA downregulates STIM1 proteins without alteration in glucose/lipid metabolism in mesangial cells. Together with previous studies ( 4, 17, 18, 24, 48 ), our present observations suggest that RA may exert beneficial effects in diabetic renal disease treatment by lowering blood glucose and inhibiting NO production through downregulation of STIM1 and inhibition of SOC in mesangial cells.
Analysis of the proposed promoter region of STIM1 gene [the 1.8-kb fragment upstream the 5'-end of the gene ( 34 )] indicates that the STIM1 promoter does not contain the consensus motif for potential RAR/RXR or RXR/RXR binding. Further search of a 2-kb region upstream the promoter, which could be the site for a potential enhancer, also does not show the presence of any potential RAR/RXR binding sites. Thus our data that STIM1 mRNA level was not altered by RA treatment was consistent with the lack of RA-regulatory elements in STIM1 gene.
RA and other retionoids exert prodifferentiation effects during embryonic development and antiproliferation effect in various cancer cells by selective degradation of cell cycle-related pathways ( 6, 51 ). RA-mediated degradation of cell cycle-specific proteins has been shown to occur via the ubiquitin-proteasome process ( 51 ). The proteasome activity as measured by the proteasome substrate sLLVY-MCA is significantly decreased in the liver and kidney homogenates of diabetic rat compared with control ( 31 ). Preincubation of mesangial cells with lactacystin prevented the inhibitory effect of RA on BK-induced SOC but had no effect on TGF- -mediated Ca 2+ influx ( Fig. 9 ), suggesting that the 26S proteasome pathways may be involved in the downregulation of STIM1 protein and SOC. Although the composition, function, and the regulation of 26S proteasome in the kidney are virtually unknown, the results of our experiments provide evidence that RA downregulation of STIM1 protein and SOC is abrogated by blockage of proteasome in rat mesangial cells.
In summary, our data suggest that inhibition of BK-induced SOC and NO production by RA treatment is associated with downregulation of STIM1 expression, the recently identified ER Ca 2+ sensor molecule. Since eNOS expression is not affected by RA, it is likely that elimination of BK-induced NO production in RA-treated cells is resulted from the inhibition of SOC, which may be the underlying mechanism for RA beneficial effect on hyperfiltration in diabetic nephropathy.
GRANTS
This work was partially supported by grants from the American Heart Association (0235065N), National Institutes of Health (R01-HL-75011), and the Department of Veterans Affairs.
Permanent address of H. Meng: The Fourth Military Medical University, Xian, China.
ACKNOWLEDGMENTS
The authors thank Dr. H. E. Abboud for providing rat mesangial cells and invaluable comments. Dr. W. Wang participated in the initial studies of this work. The discussions and encouragement from Drs. M. S. Katz and C.-K. Yeh were a great help to the work.
【参考文献】
Bachmann S, Bosse HM, Mundel P. Topography of nitric oxide synthesis by localizing constitutive NO synthases in mammalian kidney. Am J Physiol Renal Fluid Electrolyte Physiol 268: F885-F898, 1995.
Baylis C, Atzpodien EA, Freshour G, Engels K. Peroxisome proliferator-activated receptor agonist provides superior renal protection versus angiotensin-converting enzyme inhibition in a rat model of type 2 diabetes with obesity. J Pharmacol Exp Ther 307: 854-860, 2003.
Campos AH, Calixto JB, Schor N. Bradykinin induces a calcium-store-dependent calcium influx in mouse mesangial cells. Nephron 91: 308-315, 2002.
Cesario RM, Klausing K, Razzaghi H, Crombie D, Rungta D, Heyman RA, Lala DS. The rexinoid LG100754 is a novel RXR:PPAR agonist and decreases glucose levels in vivo. Mol Endocrinol 15: 1360-1369, 2001.
Dechow C, Morath C, Peters J, Lehrke I, Waldherr R, Haxsen V, Ritz E, Wagner J. Effects of all-trans retinoic acid on renin-angiotensin system in rats with experimental nephritis. Am J Physiol Renal Physiol 281: F909-F919, 2001.
Dragnev KH, Freemantle SJ, Spinella MJ, Dmitrovsky ETHA. Cyclin proteolysis as a retinoid cancer prevention mechanism. Ann NY Acad Sci 952: 13-22, 2001.
el-Dahr SS, Dipp S, Yosipiv IV, Baricos WH. Bradykinin stimulates c-fos expression, AP-1-DNA binding activity and proliferation of rat glomerular mesangial cells. Kidney Int 50: 1850-1855, 1996.
Frecker H, Munk S, Wang H, Whiteside C. Mesangial cell-reduced Ca 2+ signaling in high glucose is due to inactivation of phospholipase C- 3 by protein kinase C. Am J Physiol Renal Physiol 289: F1078-F1087, 2005.
Furusu A, Miyazaki M, Abe K, Tsukasaki S, Shioshita K, Sasaki O, Miyazaki K, Ozono Y, Koji T, Harada T, Sakai H, Kohno S. Expression of endothelial and inducible nitric oxide synthase in human glomerulonephritis. Kidney Int 53: 1760-1768, 1998.
Gooch JL, Gorin Y, Zhang BX, Abboud HE. Involvement of calcineurin in transforming growth factor- -mediated regulation of extracellular matrix accumulation. J Biol Chem 279: 15561-15570, 2004.
Han SY, So GA, Jee YH, Han KH, Kang YS, Kim HK, Kang SW, Han DS, Han JY, Cha DR. Effect of retinoic acid in experimental diabetic nephropathy. Immunol Cell Biol 82: 568-576, 2004.
Hoshiyama M, Li B, Yao J, Harada T, Morioka T, Oite T. Effect of high glucose on nitric oxide production and endothelial nitric oxide synthase protein expression in human glomerular endothelial cells. Nephron Exp Nephrol 95: e62-e68, 2003.
Hostetter TH, Troy JL, Brenner BM. Glomerular hemodynamics in experimental diabetes mellitus. Kidney Int 19: 410-415, 1981.
Kakoki M, Takahashi N, Jennette JC, Smithies O. Diabetic nephropathy is markedly enhanced in mice lacking the bradykinin B2 receptor. Proc Natl Acad Sci USA 101: 13302-13305, 2004.
Komers R, Anderson S. Paradoxes of nitric oxide in the diabetic kidney. Am J Physiol Renal Physiol 284: F1121-F1137, 2003.
Lantoine F, Iouzalen L, Devynck MA, Millanvoye-Van Brussel E, David-Dufilho M. Nitric oxide production in human endothelial cells stimulated by histamine requires Ca 2+ influx. Biochem J 330: 695-699, 1998.
Lenhard JM, Lancaster ME, Paulik MA, Weiel JE, Binz JG, Sundseth SS, Gaskill BA, Lightfoot RM, Brown HR. The RXR agonist LG100268 causes hepatomegaly, improves glycaemic control and decreases cardiovascular risk and cachexia in diabetic mice suffering from pancreatic beta-cell dysfunction. Diabetologia 42: 545-554, 1999.
Li X, Hansen PA, Xi L, Chandraratna RAS, Burant CF. Distinct mechanisms of glucose lowering by specific agonists for peroxisomal proliferator activated receptor and retinoic acid x receptors. J Biol Chem 280: 38317-38327, 2005.
Lin S, Fagan KA, Li KX, Shaul PW, Cooper DMF, Rodman DM. Sustained endothelial nitric-oxide synthase activation requires capacitative Ca 2+ entry. J Biol Chem 275: 17979-17985, 2000.
Liou J, Kim ML, Do Heo W, Jones JT, Myers JW, Ferrell J, Meyer T. STIM is a Ca 2+ sensor essential for Ca 2+ store depletion-triggered Ca 2+ influx. Curr Biol 15: 1235-1241, 2005.
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2- CT method. Methods 25: 402-408, 2001.
McGowan TA, Zhu Y, Sharma K. Transforming growth factor-beta: a clinical target for the treatment of diabetic nephropathy. Curr Diab Rep 4: 447-454, 2004.
McGowan TA, Madesh M, Zhu Y, Wang L, Russo M, Deelman L, Henning R, Joseph S, Hajnoczky G, Sharma K. TGF- -induced Ca 2+ influx involves the type III IP3 receptor and regulates actin cytoskeleton. Am J Physiol Renal Physiol 282: F910-F920, 2002.
Michellys PY, Ardecky RJ, Chen JH, Crombie DL, Etgen GJ, Faul MF, Faulkner AL, Grese TA, Heyman RA, Karanewsky DS, Klausing K, Leibowitz MD, Liu S, Mais DA, Mapes CM, Marschke K, Reifel-Miller A, Ogilvie KM, Rungta D, Thompson AW, Tyhonas JS, Boehm MF. Novel (2E,4E,6Z)-7-(2-alkoxy-3,5-dialkylbenzene)-3-methylocta-2,4,6-trienoic acid retinoid X receptor modulators are active in models of type 2 diabetes. J Med Chem 46: 2683-2696, 2003.
Mogensen CE. Glomerular filtration rate and renal plasma flow in short term and long term juvenile diabetes mellitus. Scand J Clin Invest 28: 91-100, 1971.
Morath CHRI, Dechow CLAU, Lehrke INGO, Haxsen VOLK, Waldherr RUDI, Floege JURG, Ritz EBER, Wagner JURG. Effects of retinoids on the TGF- system and extracellular matrix in experimental glomerulonephritis. J Am Soc Nephrol 12: 2300-2309, 2001.
Nutt LK, O?Neil RG. Effect of elevated glucose on endothelin-induced store-operated and nonstore-operated calcium influx in renal mesangial cells. J Am Soc Nephrol 11: 1225-1235, 2000.
Oseto S, Moriyama T, Kawada N, Nagatoya K, Takeji M, Ando A, Yamamoto T, Imai E, Hori M. Therapeutic effect of all-trans retinoic acid on rats with anti-GBM antibody glomerulonephritis. Kidney Int 64: 1241-1252, 2003.
Ouardani M, Travo P, Rakotoarivony J, Leung-Tack J. Decrease of bradykinin-induced glomerular contraction in diabetic rat: a new cellular interpretation. Eur J Cell Biol 73: 232-239, 1997.
Pickart CM, Cohen RE. Proteasomes and their kin: proteases in the machine agE. Nat Rev Mol Cell Biol 5: 177-187, 2004.
Portero-Otin M, Pamplona R, Ruiz MC, Cabiscol E, Prat J, Bellmunt MJ. Diabetes induces an impairment in the proteolytic activity against oxidized proteins and a heterogeneous effect in nonenzymatic protein modifications in the cytosol of rat liver and kidney. Diabetes 48: 2215-2220, 1999.
Ralph DeFronzo A. Diabetic nephropathy: etiologic and therapeutic considerations. Diabetes Rev 3: 510-564, 1995.
Roos J, DiGregorio PJ, Yeromin AV, Ohlsen K, Lioudyno M, Zhang S, Safrina O, Kozak JA, Wagner SL, Cahalan MD, Velicelebi G, Stauderman KA. STIM1, an essential and conserved component of store-operated Ca 2+ channel function. J Cell Biol 169: 435-445, 2005.
Sabbioni S, Veronese A, Trubia M, Taramelli R, Barbanti-Brodano G, Croce CM, Negrini M. Exon structure and promoter identification of STIM1 (alias GOK), a human gene causing growth arrest of the human tumor cell lines G401 and RD. Cytogenetic Genome Res 86: 214-218, 1999.
Schaier M, Liebler S, Schade K, Shimizu F, Kawachi H, Grone HJ, Chandraratna R, Ritz E, Wagner J. Retinoic acid receptor a and retinoid X receptor specific agonists reduce renal injury in established chronic glomerulonephritis of the rat. J Mol Med 82: 116-125, 2004.
Shultz PJ, DiCorleto PE, Silver BJ, Abboud HE. Mesangial cells express PDGF mRNAs and proliferate in response to PDGF. Am J Physiol Renal Fluid Electrolyte Physiol 255: F674-F684, 1988.
Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein using bicinchoninic acid. Anal Biochem 150: 76-85, 1985.
Song Q, Wang DZ, Harley RA, Chao L, Chao J. Cellular localization of low-molecular-weight kininogen and bradykinin B2 receptor mRNAs in human kidney. Am J Physiol Renal Fluid Electrolyte Physiol 270: F919-F926, 1996.
Soprano DR, Qin P, Soprano KJ. Retinoic acid receptors and cancers. Annu Rev Nutr 24: 201-221, 2004.
Spassova MA, Soboloff J, He LP, Xu W, Dziadek MA, Gill DL. STIM1 has a plasma membrane role in the activation of store-operated Ca 2+ channels. Proc Natl Acad Sci USA 103: 4040-4045, 2006.
Srinivasan S, Hatley ME, Bolick DT, Palmer LA, Edelstein D, Brownlee M, Hedrick CC. Hyperglycaemia-induced superoxide production decreases eNOS expression via AP-1 activation in aortic endothelial cells. Diabetologia 47: 1727-1734, 2004.
Sugimoto H, Shikata K, Matsuda M, Kushiro M, Hayashi Y, Hiragushi K, Wada J, Makino H. Increased expression of endothelial cell nitric oxide synthase (ecNOS) in afferent and glomerular endothelial cells is involved in glomerular hyperfiltration of diabetic nephropathy. Diabetologia 41: 1426-1434, 1998.
Tan Y, Wang B, Keum JS, Jaffa AA. Mechanisms through which bradykinin promotes glomerular injury in diabetes. Am J Physiol Renal Physiol 288: F483-F492, 2005.
Tang Y, Li GD. Chronic exposure to high glucose impairs bradykinin-stimulated nitric oxide production by interfering with the phospholipase-C-implicated signalling pathway in endothelial cells: evidence for the involvement of protein kinase C. Diabetologia 47: 2093-2104, 2004.
Thomson SC, Deng A, Komine N, Hammes JS, Blantz RC, Gabbai FB. Early diabetes as a model for testing the regulation of juxtaglomerular NOS I. Am J Physiol Renal Physiol 287: F732-F738, 2004.
Tolins JP, Shultz PJ, Raij L, Brown DM, Mauer SM. Abnormal renal hemodynamic response to reduced renal perfusion pressure in diabetic rats: role of NO. Am J Physiol Renal Fluid Electrolyte Physiol 265: F886-F895, 1993.
Veelken ROLA, Hilgers KF, Hartner ANDR, Haas ALEX, Bohmer KP, Sterzel RB. Nitric oxide synthase isoforms and glomerular hyperfiltration in early diabetic nephropathy. J Am Soc Nephrol 11: 71-79, 2000.
Vuligonda V, Thacher SM, Chandraratna RAS. Enantioselective syntheses of potent retinoid x receptor ligands: differential biological activities of individual antipodes. J Med Chem 44: 2298-2303, 2001.
Wagner JURG, Dechow CLAU, Morath CHRI, Lehrke INGO, Amann KERS, Waldherr RUDI, Floege JURG, Ritz EBER. Retinoic acid reduces glomerular injury in a rat model of glomerular damage. J Am Soc Nephrol 11: 1479-1487, 2000.
Way KJ, Katai N, King GL. Protein kinase C and the development of diabetic vascular complications. Diabet Med 18: 945-959, 2001.
Zancai P, Dal Col J, Piccinin S, Guidoboni M, Cariati R, Rizzo S, Boiocchi M, Maestro R, Dolcetti R. Retinoic acid stabilizes p27Kip1 in EBV-immortalized lymphoblastoid B cell lines through enhanced proteasome-dependent degradation of the p45Skp2 and Cks1 proteins. Oncogene 24: 2483-2494, 2005.
Zhang BX, Ma X, Yeh CK, Lifschitz MD, Zhu MX, Katz MS. Epidermal growth factor-induced depletion of the intracellular Ca 2+ store fails to activate capacitative Ca 2+ entry in a human salivary cell line. J Biol Chem 277: 48165-48171, 2002.
Zhang SL, Yu Y, Roos J, Kozak JA, Deerinck TJ, Ellisman MH, Stauderman KA, Cahalan MD. STIM1 is a Ca 2+ sensor that activates CRAC channels and migrates from the Ca 2+ store to the plasma membrane. Nature 437: 902-905, 2005.
Zhang WK, Shen YG, He XJ, Du BX, Xie ZM, Luo GZ, Zhang JS, Chen SY. Characterization of a novel cell cycle-related gene from Arabidopsis. J Exp Bot 56: 807-816, 2005.
Ziyadeh FN. Mediators of diabetic renal disease: the case for TGF- as the major mediator. J Am Soc Nephrol 15: S55-S57, 2004.
作者单位：1 Geriatric Research, Education, and Clinical Center, South Texas Veterans Health Care System, Audie L. Murphy Division, and Departments of 2 Biochemistry and 3 Medicine, University of Texas Health Science Center San Antonio, San Antonio, Texas