The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2008, 181, 2831 -2845
Copyright © 2008 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Akool, E.-S.
Right arrow Articles by Eberhardt, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Akool, E.-S.
Right arrow Articles by Eberhardt, W.

Molecular Mechanisms of TGFβ Receptor-Triggered Signaling Cascades Rapidly Induced by the Calcineurin Inhibitors Cyclosporin A and FK5061

El-Sayed Akool, Anke Doller, Andrea Babelova, Wasiliki Tsalastra, Kristin Moreth, Liliana Schaefer, Josef Pfeilschifter and Wolfgang Eberhardt2

Pharmazentrum Frankfurt/ZAFES, Klinikum der Johann Wolfgang Goethe-Universität, Frankfurt am Main, Germany


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The calcineurin inhibitor (CNI)-induced renal fibrosis is attributed to an exaggerated deposition of extracellular matrix, which is mainly due to an increased expression of TGFβ. Herein we demonstrate that the CNI cyclosporin A and tacrolimus (FK506), independent of TGFβ synthesis, rapidly activate TGFβ/Smad signaling in cultured mesangial cells and in whole kidney samples from CNI-treated rats. By EMSA, we demonstrate increased DNA binding of Smad-2, -3, and -4 to a cognate Smad-binding promoter element (SBE) accompanied by CNI-triggered activation of Smad-dependent expression of tissue inhibitor of metalloprotease-1 (TIMP-1) and connective tissue growth factor. Using an activin receptor-like kinase-5 (ALK-5) inhibitor and by small interfering RNA we depict a critical involvement of both types of TGFβ receptors in CNI-triggered Smad signaling and fibrogenic gene expression, respectively. Mechanistically, CNI cause a rapid activation of latent TGFβ, which is prevented in the presence of the antioxidant N-acetyl cysteine. A convergent activation of p38 MAPK is indicated by the partial blockade of CNI-induced Smad-2 activation by SB203580; conversely, both TGFβ-RII and TGFβ are critically involved in p38 MAPK activation by CNI. Activation of both signaling pathways is similarly triggered by reactive oxygen species. Finally, we show that neutralization of TGFβ markedly reduced the CNI-dependent Smad activation in vitro and in vivo. Collectively, this study demonstrates that CNI via reactive oxygen species generation activate latent TGFβ and thereby initiate the canonical Smad pathway by simultaneously activating p38 MAPK, which both synergistically induce Smad-driven gene expression.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The calcineurin inhibitors cyclosporin A (CsA)3 and tacrolimus (FK506) are structurally dissimilar immunosuppressive drugs that both block T cell activation by inhibiting the protein phosphatase calcineurin (1). Despite their beneficial actions in transplantation and in the treatment of many autoimmune disorders, the clinical use of both drugs is limited by their nephrotoxic potential. Calcineurin inhibitor (CNI) nephrotoxicity may cause glomerulosclerosis and tubulointerstitial as well as mesangial fibrosis associated with excessive deposition of extracellular matrix (ECM) (2, 3). Despite the drug-induced increase in collagen production, the accumulation of ECM is thought to be primarily due to an insufficient matrix degradation (4). Physiologically, the degradation of renal matrix components is mainly regulated by the action of two matrix-degrading enzyme systems, the matrix metalloproteinases and the plasminogen activators and their intrinsic inhibitors, the tissue inhibitors of metalloproteinases (TIMPs) and the plasminogen activator inhibitors, respectively (for review, see Refs. 5 , 6). The constitutive expression of these proteinase inhibitors underlies transcriptional regulation by various stimuli including proinflammatory cytokines and fibrogenic growth factors such as TGFβ. Additionally, TGFβ does potently induce the expression of several ECM components, for example, collagens, laminin, fibronectin, proteoglycans, and the connective tissue growth factor (CTGF). The latter one by itself is a strong stimulator of ECM synthesis (7, 8). The critical role of TGFβ in renal fibrosis has been convincingly demonstrated by different TGFβ antagonizing strategies (9, 10, 11, 12, 13). In some of these studies it was shown that TGFβ also plays also a pivotal role in the pathogenesis of chronic kidney disease induced by CNI (9, 10, 11, 12). Mechanistically, TGFβ signaling is propagated through interaction of type I and type II TGFβ receptors (two cell surface receptors with intrinsic serine threonine kinase activity) exerting rapid phosphorylation of receptor-bound Smad proteins (R-Smads), namely Smad-2 and Smad-3 (for review, see Refs. 14, 15, 16). Upon activation, the R-Smad proteins together with Smad-4, a member of the "Co-Smad" subfamily, form transcriptionally active complexes that translocate into the nucleus. Subsequently, these complexes bind with a high affinity to specific promoter elements, the Smad-binding elements (SBEs), and thereby can activate the transcription of many TGFβ-induced target genes, including CTGF (17) and TIMP-1 (18, 19). There is increasing evidence that profibrotic actions of CNI may not only be due to enhanced TGFβ expression but, additionally, may result from a direct interference of CNI with specific modules of the TGFβ/Smad signaling cascade. We therefore aimed to elucidate a direct modulation of profibrotic signaling pathways by CNI in rat renal mesangial cells (MC) and in rats undergoing short-term CNI treatment.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Materials

Human recombinant TGFβ1 was purchased from Cell Concepts. A rabbit pan-specific TGFβ Ab, a neutralizing monoclonal TGFβ1–3 Ab, mouse IgG1, and the Quantikine rat TGFβ1 immunoassay kit were purchased from R&D Systems. The CNI CsA and FK506 were from Axxora. The TGFβ-RI kinase inhibitor [3-(pyridin-2-yl)-4-(4-quinonyl)]-1H-pyrazole (Alk-5 inhibitor), U0126, SP600125, SB203580, and ebselen were obtained from Calbiochem. Diphenylene iodonium (DPI), N-acetyl cysteine (NAC), catalase, polyethylene glycol-superoxide dismutase (PEG-SOD), hypoxanthine, and xanthine oxidase were derived from Sigma-Aldrich. Abs specifically raised against Ser465/467 Smad-2 (no. 3101), total Smad-2 (no. 3102), Smad-2/Smad-3 (no. 3102), phospho-p38 MAPK (no. 9211), and p38 MAPK (no. 9212) were derived from Cell Signaling Technology. Abs against Smad-3, Smad-4, CTGF, histone deacetlyase-1 (HDAC-1), NFATc1, TIMP-1, TGFβ-RII, β-actin, anti-rabbit and anti-mouse HRP-linked IgGs, as well as control IgG were obtained from Santa Cruz Biotechnology. A small interfering RNA (siRNA) against rat cyclophilin A (no. 16708) was from Ambion, and an Ab specifically raised against cyclophilin A was from Upstate Biotechnology. Customer siRNA against the TGFβ type II receptor was synthesized from Eurogentech, and siRNA against rat Smad-2 was from Quiagen. 2',7'-Dichlorofluorescein diacetate, a cell-permeable indicator for reactive oxygen species (ROS), was from Molecular Probes.

Cell culture

Rat glomerular MC were characterized as described (20) and grown in RPMI 1640 supplemented with 10% FCS, 2 mM glutamine, 5 ng/ml insulin, 100 U/ml penicillin, and 100 µg/ml streptomycin. Serum-free preincubations were performed in DMEM supplemented with 0.1 mg/ml of fatty acid-free BSA for 24 h before treatment with CNI. All cell culture media and supplements were purchased from Invitrogen.

Determination of TGFβ1 Ag levels

The amount of TGFβ1 in cell culture supernatants, plasma, and urine was quantified by the Quantikine immunoassay kits from R&D Systems raised against rat TGFβ1. Confluent MC (1.0 to 1.5 x 106 cells) in 6-well plates were preincubated in DMEM without FCS for 24 h before stimulation with or without the CNI. Fifty microliters of conditioned media was directly transferred to the microtest strip wells of the ELISA plate. All further procedures were performed following the manufacturer’s instructions (R&D Systems). The absorbances at 450 nm were measured in a microtest plate spectrophotometer, and Ag levels were determined by appropriate calibration curves using human TGFβ1 as a standard. To ensure that the supernatants were derived from equal cell numbers, for each experimental condition, cell numbers were determined separately by use of a Neubauer chamber.

Reporter plasmids and transient transfection of MC

The pSBE4-Luc plasmid, a cis-reporting pGL3 vector containing four copies of the octameric SBE (GTCTAGAC), has been described elsewhere (21). pMBE6-Luc was a corresponding luciferase vector that instead of the wild-type SBE contains three copies of a mutated SBE (GTTTATAC). Both vectors were kindly provided by Dr. Vogelstein (Johns Hopkins Oncology Center; Baltimore, MD). A 0.6-kb promoter fragment from the rat TIMP-1 gene was cloned by PCR from rat genomic DNA as described previously (19).

Introduction of a quadruple point mutation into a putative SBE-like binding site (GTCATAGAC to CTTATGGGC) to generate pGL-TIMP-1{Delta}SBE was performed as described previously (19). pGL-CTGF, which contains a 3.0-kb upstream promoter fragment of CTGF including a functional SBE in a region between –173 and –166 (17), was kindly provided by R. Goldschmeding (Utrecht, The Netherlands).

Transient transfections of MC were performed using the Effectene reagent (Qiagen) following the manufacturer’s instructions. The transfections were done as triplicates and repeated at least three times to ensure reproducibility of the results. Transfection with pRL-CMV coding for Renilla luciferase was used to control for transfection efficiencies. Luciferase activities were measured with the dual reporter gene system (Promega) using an automated chemoluminescence detector (Berthold Technologies).

EMSA

Preparation of nuclear extracts from rat MC and EMSA were performed as described previously (22). Nuclear extracts from whole kidneys were prepared in a similar way with slight modifications. Briefly, a piece of the renal cortex was suspendend in 400 µl buffer A containing 10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, and protease inhibitors and homogenized by 20 strokes in a glass homogenizer. Homogenates were chilled on ice for 20 min before 25 µl of 10% Nonidet P-40 was added and mixed by vortexing. Cytoplasmic fractions were separated by a centrifugation step at 14,000 x g for 15 min and nuclear precipitates suspended in 100 µl high salt buffer (20 mM HEPES (pH 7.9), 0.4 M NaCl, 25% glycerol, 1 mM EDTA, 1 mM EGTA, and protease inhibitors) and incubated on ice for 30 min with frequent agitation. The supernatants from a final centrifugation step at 14,000 x g for 30 min were collected and further used as nuclear extracts. For EMSA analysis, Smad (SBE) consensus oligonucleotides encompassing either wild-type (sc-2603) or, alternatively, a mutant SBE (sc-2604) were used. DNA-protein complexes were separated from unbound oligonucleotides by electrophoresis through native 4.5% polyacrylamide gels and run in 0.5x Tris-borate EDTA.

Competition experiments were done by preincubating the DNA-binding reaction for 30 min with different dilutions of a primer stock solution corresponding to 50- (1/50), 100- (1/100), and 500- (1/500) fold dilutions of the unlabeled double-stranded wild-type or mutant oligonucleotide, respectively. Supershift analysis was done by preincubation of 1 µg supershift Ab to the binding reaction 30 min before the addition of the radioactively labeled oligonucleotides.

TGFβ neutralization experiments

The impact of TGFβ on the CsA-mediated cell responses in vitro was tested by the addition of a neutralizing rabbit pan-specific TGFβ Ab (R&D Systems). MC were pretreated with 20 µg of the neutralizing Ab for 60 min before stimulation. To exclude any unspecific inhibitory effects by immunoglobulins, cells were treated with the same amount of rabbit IgG instead of the anti-TGFβ antiserum.

Western blot analysis

For detection of CTGF, TIMP-1, TGFβ-RII, cyclophilin A, phosphorylated MAPK, and total MAPK, whole-cell lysates were prepared as described previously (22). Total cell extracts containing 50 µg of protein were prepared in SDS sample buffer and subjected to SDS-PAGE, and Western blot analysis was performed by standard procedures. Proteins were transferred to polyvinylidene difluoride membranes before immunodetection. Nuclear extracts (20–50 µg) from MC were used for assessment of the nuclear import of phoshorylated R-Smad proteins (Smad-1, Smad-2) and the Co-Smad (Smad-4), respectively. For ensuring an equal sample loading of nuclear proteins, the blots were reprobed with an anti-HDAC-1-specific Ab. After 1 h blocking in 2% BSA in Tris-buffered saline containing 0.05% Tween, Western blots were probed with the primary Ab overnight at 4°C. Following incubation with a HRP-conjugated secondary Ab, signals were detected with an ECL system.

siRNA

Gene silencing was performed using siRNAs for rat cyclophilin A (no.16708) from Ambion and small interfering siRNAs against the TGFβ type II receptor (23) and against rat Smad-2, which both had been synthesized according to rat-specific sequences. Subconfluent MC were transfected for 48 h with 25–50 nM of siRNA by use of the Oligofectamine reagent (Invitrogen) according to the manufacturer’s instructions.

ROS measurement

Measurement of ROS was performed as described previously (24).

Animals

All animal experiments were conducted in accordance with the German Animal Protection Act and were approved by the Ethics Review Committee for laboratory animals of the District Government of Darmstadt, Germany. Male Wistar rats weighting 180–200 g (Charles River Laboratories) received a single dose of FK506, dissolved in PBS by i.p. injections either at 4 h or at 24 h at a dose of 1 mg/kg body weight. Control animals received i.p. injections of PBS only. In experiments investigating the role of TGFβ in the FK506-mediated Smad-2 activation, a neutralizing mouse monoclonal anti-TGFβ1–3 Ab or control mouse IgG1 was administered i.p. at a dose of 0.5 mg/kg body weight to either controls or FK506-treated rats (n = 4 animals/group). Four hours after injection of FK506 or PBS the animals were anesthetized with pentobarbital (150 mg/kg body weight) and kidneys from these animals were removed and snap frozen in liquid nitrogen.

Blood samples of 2 ml were taken with a heparin-coated syringe from the aorta immediately after sacrifice and collected in EDTA-containing tubes. Plasma was obtained by centrifugation at 3000 rpm for 20 min and supernatants were immediately snap frozen in liquid nitrogen.

Statistical analysis

Results are expressed as means ± SD. Statistical analysis was performed using Student’s t test and for multiple comparisons the ANOVA test for significance. The data are presented as relative induction compared with control conditions or compared with CNI stimulated values. p values ≤0.05, ≤0.01, and ≤0.001 were considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Calcineurin inhibitors activate TGFβ/Smad signaling in MC

Initially, we evaluated possible modulatory effects of the two CNI, CsA and FK506, on the Smad signaling pathway by monitoring the nuclear content of phosphorylated Smad-2 indicative for the TGFβ-triggered Smad signaling cascade. As a positive control we employed 10 ng/ml TGFβ. Time-course experiments revealed that stimulation of MC with either CsA (1 µM) or FK506 (0.01 µM) induced a rapid accumulation of phosphorylated Smad-2 within nuclear fractions that was comparable to that caused by TGFβ1 (Fig. 1A). Furthermore, the CNI-induced increase in phospho-Smad-2 level peaked at 1 h and declined afterwards constantly, with a basal level in Smad-2 phosphorylation being reached after 24 h of stimulation (Fig. 1B).


Figure 1
View larger version (43K):
[in this window]
[in a new window]

  		FIGURE 1. Time-dependent activation of Smads by CsA and FK506. Quiescent MC were stimulated with either vehicle (–) or with TGFβ (10 ng/ml) for 60 min or, alternatively, with CsA (1 µM) or FK506 (0.01 µM) for the indicated short (A) or later (B) time points. For Western blot analysis, 30 µg of nuclear extracts was subjected to SDS-PAGE and successively probed with anti-phospho-Smad-2 Ab used at a dilution of 1/1000 or with an antiserum raised against Smad-4 (A). Loading of equal amounts of nuclear extracts was ascertained by incubating the blots finally with an anti-HDAC-1-specific Ab (1/1000). C, Dose-dependent activation of Smad-2 by CsA and FK506. Quiescent MC were stimulated for 60 min with either vehicle (–) or the indicated concentrations of CsA or FK506. Thirty micrograms of nuclear extracts was used for Western blot analysis using a specific anti-phospho-Smad-2 Ab (1/1000), and loading of equal amounts of nuclear extracts was ascertained by reprobing the blots with an anti-HDAC-1 Ab. Data are representative of two independent experiments giving similar results.

 
Furthermore, the increase in nuclear phospho-Smad-2 by CNI or TGFβ is accompanied by a marked increase in the nuclear Smad-4 content (Fig. 1A). In contrast, the levels of HDAC-1, a constitutive protein of the nuclear compartment, remained unchanged, which demonstrates that the effects on Smad levels are not due to an overall increase in the nuclear protein content (Fig. 1, A and B). Since activation of Smad-2 is representative of an activation of the TGFβ-triggered signaling cascade, we focused in the following experiments on this Smad isoform.

Dose-response experiments revealed that in a full accordance with its higher potency to inhibit calcineurin (25), FK506 was able to induce Smad-2 activation at concentrations that were 100-fold lower than that needed for CsA (Fig. 1C).

Smad-2 activation by CsA is independent of cyclophilin A

Next, we explored a possible involvement of immunophilin expression, and for this purpose cyclophilin A (CyPA), the intrinsic binding protein for CsA, was depleted by siRNA. MC transfected with an siRNA against cyclophilin A (si-CyPA) showed an almost complete reduction in CyPA expression when compared with those cells treated with a control siRNA (Fig. 2). As expected, CyPA depletion caused an inhibition in the downstream signaling of CsA as demonstrated by the failure of CsA to inhibit the ionomycin-triggered nuclear translocation of the transcription factor NFATc1 (Fig. 2). In contrast, Smad-2 phosphorylation by CsA was not affected by CyPA depletion, thus indicating that Smad activation by CNI is independent of CyPA and therefore seems independent of the immunosuppressive activity of CsA.


Figure 2
View larger version (52K):
[in this window]
[in a new window]

  		FIGURE 2. CNI-induced Smad activation is independent of cyclophilin A. MC were transfected with siRNA duplexes of rat cyclophilin A (si-CyPA) or with a non-gene-related control siRNA (control-siRNA). After transfection, MC were treated with ionomycin (1 µM) for 2 h in the absence (–) or presence of 1 µM CsA (+) before cells were harvested for nuclear and cytoplasmic fractions, respectively. To prove the knockdown of CyPA, 30 µg of cytoplasmic extracts was probed with anti-CyPA Ab (CyPA) by Western blot analysis. In parallel, 30 µg of nuclear extracts was successively probed with an anti-NFATc1 and a phospho-Smad-2-specific Ab (both used at a dilution of 1/1000), and loading of equal amounts of protein was ascertained by reprobing with anti-HDAC-1 antiserum. Data are representative for two independent experiments giving similar results (n. indicates nuclear; c., cytoplasmic).

 
Activation of Smad-2 by CsA and FK506 depends on extracellular TGFβ and involves TGFβ-RI kinase

Next, we tested for a functional involvement of TGFβ in the CNI-dependent activation of Smad-2 by assessment of a neutralizing TGFβ Ab. MC were preincubated without (+ vehicle) or with either 20 µg/ml of pan-specific TGFβ Ab (+ a.-TGFβ a.b.) or, alternatively, with the same amont of control IgG (+ IgG) before cells were additionally treated for 60 min with CsA (1 µM) (Fig. 3A) or with FK506 (0.01 µM) (Fig. 3B). The neutralization of TGFβ caused a significant reduction in CNI-induced Smad-2 phosphorylation, whereas addition of isotype-specific control IgG (+IgG) had no effect on nuclear phospho-Smad-2 levels (Fig. 3). Next, we tested for an involvement of the TGFβ-RI by using the specific-TGFβ-RI kinase inhibitor [3-(pyridin-2-yl)-4-(4-quinonyl)]-1H-pyrazole (26), an activin receptor-like kinase (ALK)-5 inhibitor. Preincubation of cells with 100 nM of the ALK-5 inhibitor (+TGFβ-RI-k.inhibitor) completely blocked Smad-2 activation in response to the CNI as well as to TGFβ (Fig. 3).


Figure 3
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 3. CNI-induced Smad-2 activation depends on TGFβ and involves TGFβ-RI kinase activity. Quiescent MC were pretreated for 60 min with vehicle (+ vehicle) or with either 20 µg of a pan-specific TGFβ antiserum (+ a.-TGFβ a.b.), 20 µg of control IgG (+ IgG), or with 100 nM of TGFβ-RI kinase inhibitor (+ TGFβ-RI-k.inhibitor). Thereafter, MC were stimulated for 60 min with either vehicle (–), 1 µM CsA (A), 0.01 µM FK506 (B), or with 10 ng/ml TGFβ (A and B) as indicated. For Western blot analysis, 30 µg of nuclear extracts was subjected to SDS-PAGE and probed with either an anti-phospho-Smad-2 Ab or with an anti-HDAC-1-specific Ab (both at a dilution of 1/1000). The data in the lower panels of A and B show a densitometric analysis of nuclear phospho-Smad-2 relative to nuclear HDAC-1 levels and represent means ± SD (n = 3). **, p ≤ 0.01 vs vehicle; #, p ≤ 0.05, and ##, p ≤ 0.01 vs the respective CsA, FK506, or TGFβ-stimulated conditions.

 
CsA and FK506 induce release of active TGFβ in a ROS-dependent manner

Next, we assessed whether CNI would trigger an activation of latent TGFβ. To this end, we measured the levels of active TGFβ1 in the conditioned media from CNI-treated MC by ELISA. We used an ELISA specific for TGFβ1 since in most cells all TGFβ isoforms equipotently activate the Smad signaling cascade (16). In vitro the activation of latent TGFβ requires addition of acid solutions and, therefore, the omission of this acidification step allows determination of activated TGFβ. Most interestingly, CsA and FK506 already after a short exposure of 60 min caused a 2-fold increase in the extracellular content of active TGFβ1 (Fig. 4A, left panel). These TGFβ levels were in a range sufficient to activate Smad-2 phosphorylation, since MC treated with different concentrations of recombinant TGFβ1 displayed a clear increase in basal Smad-2 phosphorylation already at 25 pg TGFβ/ml (data not shown). Since CsA has previously been shown to induce generation of ROS (27), we next assessed whether the CNI-induced release of TGFβ in MC would also rely on ROS production. To this end MC were again stimulated with CsA or FK506 and preincubated for 30 min with either the antioxidant NAC (5 mM) or with DPI (10 µM), an inhibitor of NADPH oxidases. Interestingly, preincubation with NAC completely blocked the CNI-induced activation of TGFβ, whereras DPI had no effects on CNI-induced TGFβ release (Fig. 4A, left panel). As expected, the inhibition of TGFβ release by NAC is accompanied by an almost complete inhibition of CNI-induced Smad-2 activation (Fig. 4A, right panel). Next, we test whether CNI are capable of producing ROS in MC. To this end we measured cellular radical formation by using the cell-permeable ROS acceptor 2',7'-dichlorofluorescein diacetate (24). Stimulation of MC with CsA or with FK506 caused a significant increase in ROS production and, again, preincubation with NAC almost completely abrogated CNI-induced ROS generation (Fig. 4B, left panel). In contrast, coincubation with DPI did not significantly affect CNI-triggered ROS formation, thus demonstrating that NADPH oxidases do not contribute to CNI-induced ROS formation in rat MC. To further delineate which molecular source was responsible for the CNI-triggered ROS production, we additionally tested the following compounds: ebselen, which by mimicking the action of glutathione peroxidase is used as an efficient scavenger of peroxynitrite generation; PEG-SOD, a membrane permeable SOD; and finally catalase, which degrades H2O2. Importantly, besides NAC, among these compounds only PEG-SOD was able to reduce CsA-dependent Smad-2 activation (Fig. 4B, right panel). A similar inhibitory effect by PEG-SOD was found on the FK506-triggered Smad-2 activation (data not shown). Collectively, these data indicate that CNI, by promoting a rapid release of active TGFβ1 via increased ROS production, induce Smad-2 signaling in MC. Furthermore, the study indicates that superoxide, but not H2O2, is the candidate ROS involved in CNI-induced Smad signaling. To further prove an involvement of ROS in TGFβ activation, we tested the stimulatory effects of exogenously applied ROS, which was generated by the hypoxanthine/xanthine oxidase system (HXXO). Addition of HXXO was able to mimick the stimulatory effects of CNI on TGFβ activation (Fig. 4C, left panel) and on Smad-2 phosphorylation (Fig. 4C, right panel). In contrast, the total level of TGFβ remained unchanged by HXXO (data not shown), thus indicating that ROS mainly acts by an activation of latent TGFβ.


Figure 4
View larger version (29K):
[in this window]
[in a new window]

  		FIGURE 4. CNI induce a rapid release of active TGFβ1 which depends on ROS generation. A, Left panel, Quiescent MC were treated for 60 min with vehicle, CsA (1 µM), or FK506 (0.01 µM) in the absence or presence of either NAC (5 mM) or DPI (10 µM) before cell supernatants were assessed for active TGFβ1. Data represent means ± SD (n = 3). *, p ≤ 0.05, and **, p ≤ 0.01 vs vehicle; ##, p ≤ 0.01 vs CSA-induced conditions or §, p ≤ 0.05 vs FK506-induced conditions. Right panel, Subsequently, nuclear extracts from the same cells were assessed for phospho-Smad-2 contents indicative for Smad-2 activity. Loading of equal amounts of nuclear extracts was ascertained by reprobing with an anti-HDAC-1 antiserum. B, Left panel, Effects of antioxidants on CNI-induced radical formation. MC were treated for 60 min with CNI with or without antioxidants as described in A before radical formation was determined by dichlorofluorescein (DCF) formation. Data are means ± SD (n = 3). *, p ≤ 0.05 and **, p ≤ 0.01 vs vehicle; #, p ≤ 0.05 vs CsA or §, p ≤ 0.05 vs FK506, with all experiments done in triplicates. Right panel, Quiescent MC were stimulated with either vehicle (–) or with the CsA (1 µM) in the absence or presence of either NAC (5 mM), ebselen (10 µM), PEG-SOD (100 U/ml), or catalase (500 U/ml) for 60 min before being lysed for Western blot analysis. Subsequently, 30 µg of nuclear extracts was assessed for phospho-Smad-2 content, and loading of equal amounts of nuclear extracts was ascertained by additionally probing the blots with an anti-HDAC-1 antiserum. Data are representative of two independent experiments with similar results. C, Quiescent MC were treated with vehicle, CsA (1 µM), FK506 (0.01 µM), or with xanthine oxidase (8 mU/ml) plus 50 µM hypoxanthine (HXXO) for 60 min, and supernatants were assessed for active TGFβ1 (left panel) before cells were extracted for protein homogenates. Protein lysates (30 µg) from the same cells were subjected to SDS-PAGE and immunoblotted using the indicated Abs (right panel). Data are means ± SD (n = 3). *, p ≤ 0.05 and **, p ≤ 0.01 vs vehicle.

 
Activation of Smad-2 by CNI partially depends on p38 activity

Since in many cell types the MAPKs have been implicated in TGFβ signaling (for review, see Refs. 28 , 29), we tested whether activation of Smad-2 by CNI also involves an activation of MAPKs. To this end, we employed different pharmacological inhibitors that were added 60 min before the CNI and that were applied as follows: SP600125 (10 µM), a specific inhibitor of the JNK; SB203580 (10 µM), a specific inhibitor of the p38 MAPK; and U0126 (20 µM), an inhibitor of MEK1/2. Whereas SP600125 and U0126 exerted no effect on CsA-induced (Fig. 5A) and on FK506-induced Smad-2 activation (Fig. 5B), preincubation with SB203580 caused a strong inhibition on the nuclear phospho-Smad-2 content, indicating that activation of Smad-2 by CNI partially depends on p38 MAPK activity. Most interestingly, none of the applied MAPK inhibitors had any affect on TGFβ-triggered Smad-2 activation, which indicates that the signaling cascades triggered by CNI are at least partially different from those activated by TGFβ (Fig. 5, A and B). To confirm an effective inhibition of different MAPKs by the tested pharamcological inhibitors, we assessed their specific inhibitory effect on the FK506-triggered activation of MAPKs since we had previously shown that in MC, FK506 can exert a moderate stimulatory effect on JNK and ERK, respectively (30). Interestingly, a shorter stimulation of MC with FK506 for only 10 min caused a much higher increase in phosphorylation of all three MAPKs (30), with the different MAPKs being specifically blocked by their corresponding inhibitor (Fig. 5C). In contrast, a negative interference with FK506-induced Smad-2 phosphorylation was only achieved with SB203580. Furthermore, results with TGFβ-ELISA demonstrated that a coincubation with SB203580, in contrast to its strong inhibitory effect on CNI-induced Smad-2 phosphorylation, had no effect on the CNI-induced TGFβ1 release, thus indicating that p38 activation by CNI is downstream from TGFβ activation (Fig. 5D).


Figure 5
View larger version (46K):
[in this window]
[in a new window]

  		FIGURE 5. CNI-triggered Smad-2 activity and TGFβ1 activation is abrogated in the presence of the p38 inhibitor SB203580. Quiescent MC were pretreated for 60 min with vehicle (+ vehicle), SP600125 (10 µM), SB203580 (10 µM), or U0126 (20 µM). Thereafter, cells were left untreated or stimulated for a further 60 min with 1 µM CsA (A), 0.01 µM FK506 (B), or with 10 ng/ml TGFβ as indicated. The data in the lower panels show a densitometric analysis of the nuclear phospho-Smad-2 content relative to nuclear HDAC-1 levels and represent means ± SD (n = 3). *, p ≤ 0.05 and **, p ≤ 0.01 vs vehicle; #, p ≤ 0.05 vs the respective CsA-, FK506-, or TGFβ-stimulated conditions. C, Effects of different MAPK inhibitors on FK506-induced MAPK activation. Quiescent MC were pretreated for 60 min with vehicle (+ vehicle) or with different MAPK inhibitors, and thereafter cells were stimulated for a further 10 min with 0.01 µM FK506. For Western blot analysis, 30 µg of total extracts was subjected to SDS-PAGE and successively probed with anti-phospho-Smad-2, with the indicated phosphorylation state-specific Abs against the indicated MAPK (ppMAPK) and, finally, with an anti-β-actin-specific Ab (all used at a dilution of 1/1000). D, MC were stimulated for 60 min with 1 µM CsA or with 0.01 µM FK506 in the absence or presence of 10 µM SB203580, which was preincubated for 60 min. After stimulation with CNI, cell supernatants were collected and assessed for active TGFβ1 by Quantikine ELISA. Data show a triplicate measurement of one experiment representative for three independent experiments with similar results.

 
In addition to the well-documented crosstalk with the MAPK pathway, TGFβ-dependent signaling cascades can interfere with other signaling devices, including the PI3K/Akt pathway and with the Ras-Raf kinases (28, 29). However, neither the addition of Raf1 kinase inhibitor nor that of LY2940020, an inhibitor of the Akt kinase, had any suppressive effects on CN inhibitor-induced Smad-2 phosphorylation (data not shown).

Activation of p38 MAPK by CNI depends on TGFβ type II receptor expression and is abrogated by antioxidants

Furthermore, using a phospho-specific Ab we demonstrate that both CNI induce a rapid activation of p38 MAPK (Fig. 6A, upper panel), with a similar time course being observed with Smad-2 activation (Fig. 1). Additionally, preincubation with NAC (5 mM) caused an almost total block in the CNI-evoked p38 activation (Fig. 6A, lower panel). TGFβ signaling is initiated by ligand binding to the type II TGFβ receptor (31). The possible involvement of this receptor in the CNI-induced activation of Smad-2 and p38 was tested by RNA interference. Down-regulation of the TGFβ-RII resulted in a total inhibition in the basal and CNI-induced Smad phosphorylation without changing the total amount of Smad-2 (Fig. 6B). Most intriguingly, silencing of TGFβ-RII caused also a complete loss of basal and CNI-induced p38 phosphorylation, thus implicating that the TGFβ type II receptor is indispensable for p38-MAPK activation by CNI (Fig. 6B). Additionally, neutralization of extracellular TGFβ completely blocked the CNI-induced p38 phosphorylation without affecting basal p38 activity (Fig. 6C). In summary, these data indicate that an activation of both signaling pathways by CNI is mediated via the TGFβ receptor and in both cases depends on TGFβ.


Figure 6
View larger version (33K):
[in this window]
[in a new window]

  		FIGURE 6. CsA- and FK506-induced activation of the p38 MAPK depends on ROS (A), the TGFβ receptor type II (B), and TGFβ (C). A, Time-dependent stimulation of p38 MAPK by CNI. Quiescent MC were treated with vehicle (–) or with CsA (1 µM) or FK506 (0.01 µM) for the indicated time periods (upper panel) or, alternatively, for 60 min either in the absence (–) or presence of 5 mM NAC (+ NAC) before being stimulated for a further 60 min with the indicated CNI (lower panel). Total protein (50 µg) was subjected to Western blot analysis and probed with an anti-phospho-specific p38 Ab (pp38) or, alternatively, with an Ab raised against total p38 (p38). B, Quiescent MC pretreated for MC were transfected with either a non-gene-related control siRNA (control-siRNA) or with siRNA duplexes of rat TGFβ-RII (si-TGFβ-RII) as described in Materials and Methods. After transfection, MC were serum-starved for 16 h before being stimulated with either vehicle (–), CsA (1 µM), or FK506 (0.01 µM) or, alternatively, with TGFβ (10 ng/ml) for 60 min before cells were harvested for protein extraction. C, Quiescent MC were pretreated for 60 min with vehicle (+ vehicle), with either 20 µg/ml of a pan-specific TGFβ antiserum (+ a.-TGFβ a.b.), or with 20 µg/ml of control IgG (+ IgG). Thereafter, MC were stimulated for 60 min with either vehicle (–), 1 µM CsA (left panel), or with 0.01 µM FK506 (right panel). Total protein lysates (30 µg) were subjected to SDS-PAGE and successively immunoblotted with the indicated Abs (all used at a dilution of 1/1000). Data shown are representative of two independent experiments giving similar results.

 
CNI induce DNA binding activity of Smad proteins to SBE

To test whether Smad activation by CsA or FK506 correlates with an increase in the DNA binding activity of Smads, DNA binding to a consensus SBE was monitored by EMSA. As shown in Fig. 7A, treatment of MC with either CsA (1 µM) or FK506 (0.01 µM) caused a time-dependent increase in the DNA binding affinity of three constitutively bound complexes to SBE (Fig. 7A). Addition of different supershift Abs resulted either in a strong reduction of the upper complex or, with some Abs, in a clear supershift (Fig. 7B, upper panel). In contrast, the binding of two faster migrating complexes was not affected by any of the Abs applied (Fig. 7B, upper panel). Furthermore, the specificity of DNA binding was proven by competition. As shown in Fig. 7B (lower panel), the addition of unlabeled wild-type oligonucleotide encompassing a consensus SBE sequence caused a dose-dependent impairment of DNA binding, whereas the addition of a mutated oligonucleotide had no competitive effect on the DNA binding affinity of single complexes.


Figure 7
View larger version (43K):
[in this window]
[in a new window]

  		FIGURE 7. CsA and FK506 induce DNA binding of Smad proteins to a cognate SBE. A, Time course of induction of DNA binding of different Smad proteins was analyzed by EMSA using a consensus SBE. Serum-starved MC were stimulated with either vehicle for 30 min (–) or with CsA (1 µM) and FK506 (0.01 µM) for the indicated time periods before cells were harvested for nuclear extract preparations. DNA-protein complexes were resolved from unbound DNA by nondenaturating gel electrophoresis as described in Materials and Methods. B, Supershift (upper panel) and competition (lower panel) analysis of nuclear extracts from MC stimulated for 60 min with CsA (1 µM). For supershift analysis the indicated Abs were added 60 min before the addition of the labeled oligonucleotides. The conditions for DNA binding were as described in Materials and Methods. Lower panel, Competition capacities of different wild-type (SBEwt) and mutant (SBEmt) SBEs. Different dilutions (depicted as dilutions of an oligo stock solution) of the indicated unlabeled competitor oligonucleotide were added 30 min before the addition of the 32P-labeled SBE wild-type oligonucleotide. C, Serum-starved MC were either pretreated for 30 min with vehicle (+ vehicle) or with either 20 µg of a pan-specific TGFβ antiserum (+ a.-TGFβ a.b.) or, alternatively, with 100 nM of ALK-5 inhibitor (+ TGFβ-RI-k.inhibitor). Alternatively, MC were pretreated without (+ vehicle) or with SB203580 (10 µM). Thereafter, MC were stimulated with either vehicle (–), CsA (1 µM), or with FK506 (0.01 µM) for a further 60 min before being harvested for biochemical fractionation followed by EMSA. All EMSAs shown are representative of three independent experiments giving similar results.

 
Next, we tested whether the prevention in the nuclear entry of phosphorylated Smad-2 after TGFβ neutralization or by TGFβ receptor kinase inhibition would also result in reduced SBE binding. Pretreatment with either a neutralizing TGFβ Ab (+ a.-TGFβ a.b.) or with the ALK-5 inhibitor (+ TGFβ-RI-k.inhibitor) strongly abrogated the CNI-induced SBE binding (Fig. 7C). A similar reduction in SBE binding was observed in the extracts from MC that were stimulated in the presence of p38 MAPK inhibitor SB203580 (Fig. 7C). These data indicate that for CsA and FK506, both CNI induce nuclear translocation and subsequent DNA binding of different members of the TGFβ/Smad pathway to their cognate SBE promoter element in a manner that depends on TGFβ, TGFβ receptor-dependent signaling, and p38 activity.

CsA and FK506 induce transcriptional activity of TGFβ-inducible promoters

We evaluated whether the increase in SBE binding by CNI would correlate with an increase in Smad-controlled gene expression. First, we tested whether CsA or FK506 activate an artificial TGFβ-inducible control promoter bearing a tandem of four copies of SBE consensus motifs (pSBE4-Luc) (21). Stimulation of MC with CsA or FK506 for 16 h caused a significant increase in luciferase activity (Fig. 8A, filled bars), whereas a corresponding luciferase reporter gene, containing a tandem of mutated SBE motifs (pMBE6-Luc), was not induced by CNI (Fig. 8A, open bars), thus emphasizing the critical role of SBE.


Figure 8
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 8. CsA and FK506 induce the activity of SBE-driven gene promoters. Subconfluent MC were transfected with 0.4 µg of either pSBE4-Luc (A, filled bars) containing four tandem wild-type SBEs or, alternatively, with pMBE6 (A, open bars) containing six point-mutated SBEs (A) or with 0.4 µg of pGL-TIMP-1 (B, filled bars). Additionally, cells were cotransfected with the same promoter bearing a mutated SBE (B, open bars) or with 0.4 µg of a 3.0-kb fragment of the CTGF promoter (C, filled bars). Transfection of the plasmids was supplemented by a cotransfection with 0.1 µg of RL-CMV, coding for Renilla luciferase, as described in Materials and Methods. After the transient transfection, MC were treated for 16 h with either vehicle, CsA (1 µM), or with FK506 (0.01 µM) before being extracted for total cell lysates and assayed for luciferase activities. The values for beetle luciferase were related to the values for Renilla luciferase and are depicted as relative light units (RLU). Data represent the means ± SD (n = 6) of triplicate experiments. **, p ≤ 0.01 and ***, p ≤ 0.005 compared with vehicle.

 
As a native promoter we additionally tested a 0.6-kb fragment of the rat TIMP-1 gene promoter. This promoter fragment is able to confer TGFβ inducibility via a functional SBE-like motif (19). A 16-h stimulation of transiently transfected MC with either CsA (1 µM) or FK506 (0.01 µM) caused a strong and significant increase in basal promoter activity (Fig. 8B, filled bars). To confirm the functional role of the gene-specific SBE, MC were additionally transfected with a corresponding TIMP-1 promoter construct bearing a mutated SBE. Again, the mutation of SBE caused a complete inhibition of the CNI-induced luciferase activation (Fig. 8B, open bars), thus indicating that the SBE-like motif is critical for an activation of TIMP-1 by CNI. Finally, we tested CTGF, which contains a proximal SBE critical for TGFβ-dependent promoter activation (17). MC transiently transfected with pGL-CTGF displayed a similar response toward CNI as observed with the TIMP-1 promoter (Fig. 8C). Collectively, these data clearly indicate that both CNI can activate TGFβ-inducible promoters.

CsA and FK506 propagate profibrotic cell responses in MC

In a next step, we investigated whether the activation of SBE-driven genes would functionally correlate with an up-regulation of CTGF and TIMP-1. Stimulation of MC for 16 h with either CsA or FK506 caused a dose-dependent increase in steady-state CTGF and TIMP-1 protein levels, and preincubation with either SB203580 or with a neutralizing TGFβ Ab (+ a.-TGFβ a.b.) caused a strong reduction in CNI-induced expression of these proteins (Fig. 9, A and B). In contrast, U0126 and SP600125 had no inhibitory effect on CTGF or TIMP-1 levels, thus confirming that in MC the ERK and JNK signaling cascades are not involved in the CNI-triggered Smad activation (Fig. 9, A and B). To prove that a pharmacological inhibition of these target genes is primarily due to an upstream modulation of Smad signaling, we tested the effects of Smad-2 silencing using RNA interference. Attenuation of Smad-2 expression that specifically prevented the CNI-induced increase in Smad-2 phosphorylation was concomitant with a strong reduction in CNI-triggered CTGF and TIMP-1 expression (Fig. 9C). These data strongly suggest that the activation of Smad-2 by both CNI is functionally relevant and causative for an up-regulation of profibrotic genes.


Figure 9
View larger version (42K):
[in this window]
[in a new window]

  		FIGURE 9. CNI-induced expression of CTGF and TIMP-1 is mediated by Smad-2 and critically depends on p38 activity and TGFβ. Quiescent MC were stimulated for 16 h (A and B) with either vehicle (–), 1 µM CsA (A), or with 0.01 µM FK506 (B) in the absence (+ vehicle) or presence of SP600125 (10 µM), U0126 (20 µM), SB203580 (10 µM), or 20 µg of a pan-specific TGFβ Ab (+ a.-TGFβ a.b.). C, Induction of CTGF and TIMP-1 expression by CNI is abrogated after RNA interference of Smad-2. MC were transfected with siRNA duplexes of rat Smad-2 (si-Smad-2) or with a non-gene-related control siRNA (control-siRNA). After transfection, cells were treated for 8 h with vehicle (–), with CsA, or with FK506 as indicated. After stimulation, MC were harvested for total protein extracts and Western blots were successively probed with anti-Smad-2 Ab (Smad-2), anti-phospho-Smad-2-specific Ab (p-Smad-2), CTGF, anti-TIMP-1, and β-actin-specific antisera. The data shown are representative of two independent experiments giving similar results.

 
FK506 induces a rapid and TGFβ-dependent activation of Smad-2 in vivo

To test whether the CNI-mediated activation of Smads observed in cultured MC would also occur in vivo, we analyzed a possible Smad-2 activation by FK506 in rat kidneys. We limited our animal experiments by testing only FK506 for two reasons: 1) in MC FK506, independent from CN, inhibition has displayed a similar efficiency to activate Smad signaling cascades as CsA; and, more importantly, 2) FK506, due to its somewhat lower nephrotoxicity, is currently the preferentially used CNI in clinical regimes. Western blot analysis from time-course experiments using nuclear extracts from whole kidney samples revealed a marked increase in the nuclear phospho-Smad-2 levels already at 4 h after rats had received a single dose of FK506 (1 mg/kg body weight) but which disappeared at the later time point of 24 h (Fig. 10A). Importantly, measurement of TGFβ levels in the plasma of FK506-treated animals revealed that the increase in phosphorylated Smad-2 level within the renal tissue was accompanied by a transient increase in plasma TGFβ level, with a peak measured after 4 h, which thereafter declined back to a level that was measured in vehicle-treated animals (Fig. 10B). In contrast to the plasma TGFβ levels, the amount of active TGFβ in the urine of short-term-treated animals was below the detection limit of the ELISA (data not shown). Finally, by testing the modulatory effects after neutralizing of TGFβ, we found that the rapid effect on nuclear Smad-2 translocation and phosphorylation was significantly reduced in those animals which had been additionally treated with a neutralizing TGFβ Ab (a.-TGFβ a.b.) (Fig. 10C). By contrast, the administration of control IgG had no significant effect on Smad-2 activity (Fig. 10C). These data provide evidence for a functional role of activated TGFβ in the rapid stimulation of Smad-2 by FK506 in vivo.


Figure 10
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 10. In vivo activation of Smad-2 by FK506 critically depends on TGFβ and is furthermore accompanied by increased plasma levels of active TGFβ1. A, Western blot analysis showing the time-dependent activation of Smad-2 in total kidney extracts from either vehicle (–) or FK506-treated rats. Male Wistar rats weighting 180–200 g either received a single dose of FK506 (1 mg/kg body weight) or the same volume of PBS (–) used as vehicle for the indicated time points. The data shown are representative for four individually treated animals giving similar results. B, Serum levels of activated TGFβ1 in rats treated for the indicated time points with FK506 (1 mg/kg body weight). Data represent means ± SD (n = 4) compared with animals treated for 4 h with vehicle (–). C, A statistically significant decrease in the FK506-induced Smad-2 phosphorylation was observed in those animals, which, in addition to FK506, had been treated with a single dose of neutralizing anti-TGFβ (0.5 mg/kg body weight) Ab (+ a.-TGFβ-a.b.), but not in animals that received a similar dose of IgG1 (+ IgG). Statistical analysis demonstrates the stimulatory effects on nuclear phospho-Smad-2 compared with nuclear HDAC-1 levels 4 h after application of the indicated agents. Data represent means ± SD (n = 4). *, p ≤ 0.05 compared with animals treated for 4 h with vehicle, or #, p ≤ 0.05 compared with FK506-treated animals.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
To best of our knowledge, this is the first study demonstrating that CsA and FK506 both at nanomolar concentrations can induce a rapid induction of the canonical Smad signaling cascade starting from a ligand- and TGFβ receptor-dependent activation, leading to phosphorylation and nuclear translocation of the R-Smad protein Smad-2, and ending with an enhancement of SBE-driven profibrotic gene expression (Fig. 11). We have previously demonstrated that CsA, but not FK506, independent from its calcineurin inhibitory capacity specifically interferes with cytokine-induced expression of matrix metalloproteinase-9 (30), a finding that may reveal the beneficial role of FK506 observed in some clinical trials when compared with CsA (32, 33, 34). Even though CsA has been found to exert a higher nephrotoxic potential when compared with tacrolimus, long-term fibrogenic effects were observed with both compounds, suggesting a similar mechanism underlying these side effects (34, 35, 36). There are several reports demonstrating that the profibrotic actions of both drugs are attributable to an autocrine synthesis of TGFβ (9, 38, 39), and some studies have demonstrated a transcriptional effect on TGFβ expression by CsA both in vitro (40) and in vivo (41, 42). Therefore, an increase in urinary TGFβ excretion was discussed as a useful parameter for monitoring CNI-induced side effects. However, taking into account that patients in most clinical settings undergo long-term immunosuppressive therapy with combinations of different immunomodulatory and anti-inflammatory compounds, which each of them can exert a modulatory effect on TGFβ expression and/or excretion (41, 42, 43), neither urinary nor plasma TGFβ levels are reliable prognostic markers for CNI-mediated fibrosis. Our observation that FK506 similar to CsA causes a rapid activation of Smads in MC but also in the in vivo setting strongly implies that these rapid effects in addition to long-term effects may be of pathophysiological relevance. This notion is further supported by the finding that activation of Smad-2 after a single dose of FK506 is accompanied by a transient increase in active TGFβ in the plasma (Fig. 10B). Consistent with the increase in Smad binding to cognate promoter sites, we found a rapid activation of TGFβ-inducible promoters by CNI, including those of TIMP-1 and CTGF. Using a specific kinase inhibitor of the type I TGFβ receptor denominated as ALK-5 inhibitor, siRNAs, and a neutralizing TGFβ Ab, we furthermore proved that both types of TGFβ receptors and active TGFβ are indispensable for Smad activation by CNI.


Figure 11
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 11. Schematic summary of renal profibrotic signaling pathways induced within minutes by CNI. CsA and FK506 via a cyclophilin-independent pathway, via increased ROS production, can activate TGFβreceptor-triggered signaling pathways by activating rapid release of latent ECM-bound TGFβ. Thereby, the Smad-2-dependent pathway is convergently activated through p38 and TGFβ-triggered signaling. The association of Smad-2 with Smad-4 (Smad-2/Smad-4) gives rise to an active Smad transcription factor complex that specifically binds to a cognate SBE present in the promoters of CTGF and TIMP-1 genes. The diagram also depicts the different approaches used to specifically inhibit single devices in the CNI-induced signaling cascade. ATF indicates activating transcription factor; Elk, E26-like protein; SOD, superoxide dismutase.

 
Unlike their immunosuppressive action, which predominantly results from an inhibition of calcineurin and its downstream target NFAT, the activation of Smads in MCs seems independent of immunophilin binding, as is clearly demonstrated by the results from RNA interference experiments (Fig. 2).

While searching for the molecular mechanism responsible for CNI-triggered Smad activation, we found that both CNI activate Smad signaling by a signaling mechanism that is triggered by increased ROS production and activation of latent TGFβ1 (Fig. 11). Activating latent TGFβ1 in amounts that suffice to activate Smads (see Fig. 4A) is implicated by ELISA and by the strong inhibitory effects observed after neutralizing TGFβ by a pan-specific Ab. Most importantly, this inhibitory effect of TGFβ neutralization was also observed in vivo. Previously, a release of latent TGFβ by CsA as a result of drug-induced cell death has been demonstrated in lymphocytes. In contrast to our findings, the effects on TGFβ release in T cells were only observed after prolonged incubation and at higher doses of CsA (10 µM), which most probably induce cytotoxicity (44).

Physiologically, activation of latent TGFβ is achieved by either proteolytic or nonproteolytic events (for review, see Ref. 45) and in some cases may include a redox-sensitive mechanism (46, 47). Previous studies have demonstrated that CsA at different concentrations promotes the generation of ROS by a mechanism that is independent of cytochrome P-450 oxidases (27) and NADPH oxidases (48), the major cellular superoxide-generating enzymes. In full agreement with these studies on smooth muscle cells, we found a similar increase in ROS production by both CNI that is blocked by the antioxidant NAC. In contrast to NAC, ROS production was neither affected by DPI, an inhibitor of several important ROS-generating enzymes, nor by catalase. The latter finding indicates that H2O2 is not the relevant ROS. Despite NADPH oxidases and xanthine oxidase, other well-known sources of cellular superoxide, including those of the inner and outer mitochondrial membrane but also the arachidonic acid-metabolizing enzymes, cylooxygenases and lipoxygeneases (for review, see Ref. 49), may represent the primary source of CNI-triggered ROS in the kidney. This important issue has to be addressed by future investigations. In addition to activation of latent TGFβ, we herein demonstrate that Smad activation by CNI depends on the activation of p38, which itself is blocked by the ROS scavenger NAC (Fig. 6A). This is corroborated by the finding that SB203580 specifically interfered with the CNI-induced Smad-2 phosphorylation, although it had no effect on the TGFβ-triggered Smad-2 activation (Fig. 5, A and B). From these data it is tempting to speculate that ROS, by a TGFβ-dependent mechanism, additionally activate the p38 MAPK via the TGFβ type II receptor (Fig. 11). A critical involvement of the p38 MAPK in CNI-induced Smad activation is in a certain line with former studies demonstrating functional crosstalk with the MAP kinase pathway (28, 29, 50, 51, 52). However, most of these studies have demonstrated a Smad-independent activation of MAPK by TGFβ in a way that is furthermore independent of a functional TGFβ type I receptor (28, 52). In clear contrast to these findings, the rapid activation of p38 by CsA or FK506 reported herein strongly depends on TGFβ receptor signaling. Furthermore, in mesangial cells the release of CNI-triggered TGFβ is not affected by p38 inhibition, which clearly demonstrates that ROS generation by CNI is upstream of p38 MAPK activation and upstream of TGFβ activation (Fig. 11). Therefore, both pathways initiated by CNI convergently can activate Smad-2 signaling and subsequent gene expression by a mechanism that requires the TGFβ receptor (Fig. 11). The potential blockade of these drug-induced signaling events, either by anti-TGFβ Abs or antioxidant treatment, may emphasize the concept of therapeutic TGFβ neutralization in combination with antioxidant therapy as a valuable approach for the prevention of CNI-induced renal fibrosis. Furthermore, our study sheds light on an undescribed mechanistic facet of the complicated and intimate crosstalk between the TGFβ and MAPK pathways, which besides tissue fibrosis may be important for the anticancer properties observed for CNI.


   Acknowledgments
 
We thank R. Goldschmeding (Department of Pathology, University Medical Center, Utrecht, The Netherlands) for kindly providing the plasmid pGL-CTGF, Dr. Bert Vogelstein (The Johns Hopkins Oncology Center, Baltimore, MD) for kindly providing the plasmids SBE4-Luc and MBE6-Luc, and Miriam Plesková for technical assistance with the ROS measurement. Furthermore, we thank Prof. I. A. Hauser (Department of Nephrology, Klinikum der Johann Wolfgang Goethe-Universität Frankfurt, Germany) for helpful discussions.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflicts of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by the German Research Foundation (DFG) Grants EB 257/3-1, PF 361/2-2, SCHA 1082/2-1, the Excellence Cluster Cardiopulmonary System (ECCPS) EXC 147/1, and the Interdisciplinary Center of Clinical Research, Münster (Schae 2/026/06). Back

2 Address correspondence and reprint requests to Dr. Wolfgang Eberhardt, Pharmazentrum Frankfurt/ZAFES, Klinikum der Johann Wolfgang Goethe-Universität, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany. E-mail address: w.eberhardt{at}em.uni-frankfurt.de Back

3 Abbreviations used in this paper: CsA, cyclosporin A; ALK, activin receptor-like kinase; CNI, calcineurin inhibitors; CTGF, connective tissue growth factor; CyPA, cyclophilin A; DPI, diphenylene iodonium; ECM, extracellular matrix; HDAC, histone deacetlyase; MC, mesangial cells; NAC, N-acetyl cysteine; PEG-SOD, polyethylene glycol-superoxide dismutase; ROS, reactive oxygen species; R-Smad, receptor-bound Smad protein; SBE, Smad binding element; siRNA, small interfering RNA; TIMP-1, tissue inhibitor of matrix metalloproteinase. Back

Received for publication March 6, 2008. Accepted for publication June 18, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Schreiber, S. L., G. R. Crabtree. 1992. The mechanism of action of cyclosporine A and FK506. Immunol. Today 13: 136-142. [Medline]
  2. Mihatsch, M. J., M. Kyo, K. Morozumi, Y. Yamaguchi, V. Nickeleit, B. Ryffel. 1998. The side-effects of ciclosporine-A and tacrolimus. Clin. Nephrol. 49: 356-363. [Medline]
  3. Fellstrom, B.. 2004. Cyclosporine nephrotoxicity. Transplant. Proc. 36: S220-S223. [Medline]
  4. Duymelinck, C., J. T. Deng, S. E. Dauwe, M. E. De Broe, G. A. Verpooten. 1998. Inhibition of the matrix metalloproteinase system in a rat model of chronic cyclosporine nephropathy. Kidney Int. 54: 804-818. [Medline]
  5. Woessner, J. F., Jr. 1991. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 5: 2145-2154. [Abstract]
  6. Eddy, A. A.. 2000. Molecular basis of renal fibrosis. Pediatr. Nephrol. 15: 290-301. [Medline]
  7. Gore-Hyer, E., D. Shegogue, M. Markiewicz, S. Lo, D. Hazen-Martin, E. L. Greene, G. Grotendorst, M. Trojanowska. 2002. TGFβ and CTGF have overlapping and distinct fibrogenic effects on human renal cells. Am. J. Physiol. 283: F707-F716.
  8. Okada, H., T. Kikuta, T. Kobayashi, T. Inoue, Y. Kanno, M. Takigawa, T. Sugaya, J. B. Kopp, H. Suzuki. 2005. Connective tissue growth factor expressed in tubular epithelium plays a pivotal role in renal fibrogenesis. J. Am. Soc. Nephrol. 16: 133-143. [Abstract/Free Full Text]
  9. Khanna, A. K., V. R. Cairns, C. G. Becker, J. D. Hosenpud. 1999. Transforming growth factor (TGF)-β mimics and anti-TGF-β antibody abrogates the in vivo effects of cyclosporine: demonstration of a direct role of TGF-β in immunosuppression and nephrotoxicity of cyclosporine. Transplantation 67: 882-889. [Medline]
  10. Ziyadeh, F. N., B. B. Hoffman, D. C. Han, M. C. Iglesias-De La Cruz, S. W. Hong, M. Isono, S. Chen, T. A. McGowan, K. Sharma. 2000. Long-term prevention of renal insufficiency, excess matrix gene expression, and glomerular mesangial matrix expansion by treatment with monoclonal antitransforming growth factor-β antibody in db/db diabetic mice. Proc. Natl. Acad. Sci. USA 97: 8015-8020. [Abstract/Free Full Text]
  11. Islam, M., J. F. Burke, Jr, T. A. McGowan, Y. Zhu, S. R. Dunn, P. McCue, J. Kanalas, K. Sharma. 2001. Effect of anti-transforming growth factor-β antibodies in cyclosporine-induced renal dysfunction. Kidney Int. 59: 498-506. [Medline]
  12. Ling, H., X. Li, S. Jha, W. Wang, L. Karetskaya, B. Pratt, S. Ledbetter. 2003. Therapeutic role of TGF-β-neutralizing antibody in mouse cyclosporine A nephropathy: morphologic improvement associated with functional preservation. J. Am. Soc. Nephrol. 14: 377-388. [Abstract/Free Full Text]
  13. Fukasawa, H., T. Yamamoto, H. Suzuki, A. Togawa, N. Ohashi, Y. Fujigaki, C. Uchida, M. Aoki, M. Hosono, M. Kitagawa, A. Hishida. 2004. Treatment with anti-TGF-β antibody ameliorates chronic progressive nephritis by inhibiting Smad/TGF-β signaling. Kidney Int. 65: 63-74. [Medline]
  14. Massague, J.. 1998. TGF-β signal transduction. Annu. Rev. Biochem. 67: 753-791. [Medline]
  15. Piek, E., C. H. Heldin, P. Ten Dijke. 1999. Specificity, diversity, and regulation in TGF-β superfamily signaling. FASEB J. 13: 2105-2124. [Abstract/Free Full Text]
  16. Miyazono, K., P. ten Dijke, C. H. Heldin. 2000. TGF-β signaling by Smad proteins. Adv. Immunol. 75: 115-157. [Medline]
  17. Chen, Y., I. E. Blom, S. Sa, R. Goldschmeding, D. J. Abraham, A. Leask. 2002. CTGF expression in mesangial cells: involvement of SMADs, MAP kinase, and PKC. Kidney Int. 62: 1149-1159. [Medline]
  18. Marti, H. P., L. Lee, M. Kashgarian, D. H. Lovett. 1994. Transforming growth factor-beta 1 stimulates glomerular mesangial cell synthesis of the 72-kd type IV collagenase. Am. J. Pathol. 144: 82-94. [Abstract]
  19. Akool, E. S., A. Doller, R. Muller, P. Gutwein, C. Xin, A. Huwiler, J. Pfeilschifter, W. Eberhardt. 2005. Nitric oxide induces TIMP-1 expression by activating the transforming growth factor β-Smad signaling pathway. J. Biol. Chem. 280: 39403-39416. [Abstract/Free Full Text]
  20. Pfeilschifter, J., K. Vosbeck. 1991. Transforming growth factor beta 2 inhibits interleukin 1 beta- and tumour necrosis factor alpha-induction of nitric oxide synthase in rat renal mesangial cells. Biochem. Biophys. Res. Commun. 175: 372-379. [Medline]
  21. Zawel, L., J. L. Dai, P. Buckhaults, S. Zhou, K. W. Kinzler, B. Vogelstein, S. E. Kern. 1998. Human Smad3 and Smad 4 are sequence-specific transcription activators. Mol. Cell 1: 611-617. [Medline]
  22. Eberhardt, W., M. Schulze, C. Engels, E. Klasmeier, J. Pfeilschifter. 2002. Glucocorticoid-mediated suppression of cytokine-induced matrix metalloproteinase-9 expression in rat mesangial cells: involvement of nuclear factor-{kappa}B and Ets transcription factors. Mol. Endocrinol. 16: 1752-1766. [Abstract/Free Full Text]
  23. Xin, C., S. Ren, B. Kleuser, S. Shabahang, W. Eberhardt, H. Radeke, M. Schafer-Korting, J. Pfeilschifter, A. Huwiler. 2004. Sphingosine 1-phosphate cross-activates the Smad signaling cascade and mimics transforming growth factor-β-induced cell responses. J. Biol. Chem. 279: 35255-35262. [Abstract/Free Full Text]
  24. Plesková, M., K. F. Beck, M. H. Behrens, A. Huwiler, B. Fichtelscherer, O. Wingerter, R. P. Brandes, A. Mülsch, J. Pfeilschifter. 2005. Nitric oxide down-regulates the expression of the cytalytic NADPH oxidase subunit Nox 1 in rat renal mesangial cells. FASEB J. 20: 139-141. [Medline]
  25. Sawada, S., G. Suzuki, Y. Kawase, F. Takaku. 1987. Novel immunosuppressive agent, FK506. In vitro effects on the cloned T cell activation. J. Immunol. 139: 1797-1803. [Abstract]
  26. Sawyer, J. S., D. W. Beight, K. S. Britt, B. D. Anderson, R. M. Campbell, T. Goodson, Jr, D. K. Herron, H.Y. Li, W. T. McMillen, N. Mort, et al 2004. Synthesis and activity of new aryl- and heteroaryl-substituted 5,6-dihydro-4H-pyrrolo[1,2-b] pyrazole inhibitors of the transforming growth factor-β type I receptor kinase domain. Bioorg. Med. Chem. Lett. 14: 3581-3584. [Medline]
  27. Krauskopf, A., T. M. Buetler, N. S. D. Nguyen, K. Macé, U. T. Ruegg. 2002. Cyclosporin A-induced free radical generation is not mediated by cytochrome P-450. Br. J. Pharmacol. 135: 977-986. [Medline]
  28. Moustakas, A., C. H. Heldin. 2005. Non-Smad TGF-β signals. J. Cell Sci. 118: 3573-3584. [Abstract/Free Full Text]
  29. Mulder, K. M.. 2000. Role of Ras and Mapks in TGFβ signaling. Cytokine Growth Factor Rev. 11: 23-35. [Medline]
  30. Doller, A., E. S. Akool, R. Muller, P. Gutwein, C. Kurowski, J. Pfeilschifter, W. Eberhardt. 2007. Molecular mechanisms of cyclosporine A inhibition of the cytokine-induced matrix metalloproteinase-9 in glomerular mesangial cells. J. Am. Soc. Nephrol. 18: 581-592. [Abstract/Free Full Text]
  31. Lin, H. Y., X. F. Wang, E. Ng-Eaton, R. A. Weinberg, H. F. Lodish. 1992. Expression cloning of the TGF-β type II receptor, a functional transmembrane serine/threonine kinase. Cell 68: 775-785. [Medline]
  32. Bicknell, G. R., S. T. Williams, J. A. Shaw, J. H. Pringle, P. N. Furness, M. L. Nicholson. 2000. Differential effects of cyclosporine and tacrolimus on the expression of fibrosis-associated genes in isolated glomeruli from renal transplants. Br. J. Surg. 87: 1569-1575. [Medline]
  33. Reynolds, N. J., W. I. Al-Daraji. 2002. Calcineurin inhibitors and sirolimus: mechanisms of action and applications in dermatology. Clin. Exp. Dermatol. 27: 555-561. [Medline]
  34. Pirsch, J. D., J. Miller, M. H. Deierhoi, F. Vincenti, R. S. Filo. 1997. A comparison of tacrolimus (FK506) and cyclosporine for immunosuppression after cadaveric renal transplantation: FK506 Kidney Transplant Study Group. Transplantation 63: 977-983. [Medline]
  35. Waiser, J., K. Dell, J. Kreutzkamp, T. Bohler, K. Budde, H. Peters, H. H. Neumayer. 2006. FK506, transforming growth factor-β1 and mesangial matrix synthesis: parallels and differences compared with cyclosporine A. Cytokine 33: 59-65. [Medline]
  36. Mihatsch, M. J., M. Kyo, K. Morozumi, Y. Yamaguchi, V. Nickeleit, B. Ryffel. 1998. The side-effects of ciclosporine-A and tacrolimus. Clin. Nephrol. 49: 356-363. [Medline]
  37. Shin, G. T., A. Khanna, R. Ding, V. K. Sharma, M. Lagman, B. Li, M. Suthanthiran. 1998. In vivo expression of transforming growth factor-β1 in humans: stimulation by cyclosporine. Transplantation 65: 313-318. [Medline]
  38. Pankewycz, O. G., L. Miao, R. Isaacs, J. Guan, T. Pruett, G. Haussmann, B. C. Sturgill. 1996. Increased renal tubular expression of transforming growth factor β in human allografts correlates with cyclosporine toxicity. Kidney Int. 50: 1634-1640. [Medline]
  39. Wolf, G., E. Mueller, R. A. Stahl, F. N. Ziyadeh. 1993. Angiotensin II-induced hypertrophy of cultured murine proximal tubular cells is mediated by endogenous transforming growth factor-β. J. Clin. Invest. 92: 1366-1372. [Medline]
  40. Prashar, Y., A. Khanna, P. Sehajpal, V. K. Sharma, M. Suthanthiran. 1995. Stimulation of transforming growth factor-β1 transcription by cyclosporine. FEBS Lett. 358: 109-112. [Medline]
  41. Gao, R., Y. Lu, Y. P. Xin, X. H. Zhang, J. Wang, Y. P. Li. 2006. The effects of different immunosuppressants on chronic allograft nephropathy by affecting the transforming growth factor-β and Smads signal pathways. Transplant. Proc. 38: 2154-2157. [Medline]
  42. Citterio, F., U. Pozzetto, J. Romagnoli, E. Tondolo, P. Silvestri, G. Nanni, M. Castagneto. 2004. Plasma levels of transforming growth factor-β1 in renal transplant recipients receiving different immunosuppressive regimens. Transplant. Proc. 36: 698-699. [Medline]
  43. Haramaki, R., K. Tamaki, M. Fujisawa, H. Ikedo, N. Haramaki, S. Okuda. 2001. Steroid therapy and urinary transforming growth factor-β1 in IgA nephropathy. Am. J. Kidney Dis. 38: 1191-1198. [Medline]
  44. Minguillon, J., B. Morancho, S. J. Kim, M. Lopez-Botet, J. Aramburu. 2005. Concentrations of cyclosporine A and FK506 that inhibit IL-2 induction in human T cells do not affect TGF-β1 biosynthesis, whereas higher doses of cyclosporine A trigger apoptosis and release of preformed TGF-β1. J. Leukocyte Biol. 77: 748-758. [Abstract/Free Full Text]
  45. Annes, J. P., J. S. Munger, D. B. Rifkin. 2003. Making sense of latent TGFβ activation. J. Cell Sci. 116: 217-224. [Abstract/Free Full Text]
  46. Barcellos-Hoff, M. H., T. A. Dix. 1996. Redox-mediated activation of latent transforming growth factor-β1. Mol. Endocrinol. 10: 1077-1083. [Abstract/Free Full Text]
  47. Jobling, M. F., J. D. Mott, M. T. Finnegan, V. Jurukovski, A. C. Erickson, P. J. Walian, S. E. Taylor, S. Ledbetter, C. M. Lawrence, D. B. Rifkin, M. H. Barcellos-Hoff. 2006. Isoform-specific activation of latent transforming growth factor β (LTGF-β) by reactive oxygen species. Radiat. Res. 166: 839-848. [Medline]
  48. Krauskopf, A., P. Lhote, O. Petermann, U. T. Ruegg, T. M. Buetler. 2005. Cyclosporin A generates superoxide in smooth muscle cells. Free Radical Res. 39: 913-919. [Medline]
  49. Tkaczyk, J., M. Vízek. 2007. Oxidative stress in the lung tissue-sources of reactive oxygen species and antioxidant defence. Prague Med. Rep. 108: 105-114. [Medline]
  50. Hayashida, T., M. Decaestecker, H. W. Schnaper. 2003. Cross-talk between ERK MAP kinase and Smad signaling pathways enhances TGF-β-dependent responses in human mesangial cells. FASEB J. 17: 1576-1578. [Abstract/Free Full Text]
  51. Javelaud, D., A. Mauviel. 2005. Crosstalk mechanisms between the mitogen-activated protein kinase pathways and Smad signaling downstream of TGF-β: implications for carcinogenesis. Oncogene 24: 5742-5750. [Medline]
  52. Yu, L., M. C. Hebert, Y. E. Zhang. 2002. TGF-β receptor-activated p38 MAP kinase mediates Smad-independent TGF-β responses. EMBO J. 21: 3749-3759. [Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Gastrointest. Liver Physiol.Home page
Y. Satoh, Y. Ishiguro, H. Sakuraba, S. Kawaguchi, H. Hiraga, S. Fukuda, and A. Nakane
Cyclosporine regulates intestinal epithelial apoptosis via TGF-{beta}-related signaling
Am J Physiol Gastrointest Liver Physiol, September 1, 2009; 297(3): G514 - G519.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Akool, E.-S.
Right arrow Articles by Eberhardt, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Akool, E.-S.
Right arrow Articles by Eberhardt, W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS