Skip to main content
NCBI home page
British Journal of Pharmacology logoLink to British Journal of Pharmacology
. 2013 Oct 28;170(6):1177–1189. doi: 10.1111/bph.12348

An in vitro model for the pro-fibrotic effects of retinoids: mechanisms of action

A C Rankin 1, B M Hendry 1, J P Corcoran 2, Q Xu 1
PMCID: PMC3838693  PMID: 23992207

Abstract

BACKGROUND AND PURPOSE

Retinoids, including all-trans retinoic acid (tRA), have dose-dependent pro-fibrotic effects in experimental kidney diseases. To understand and eventually prevent such adverse effects, it is important to establish relevant in vitro models and unravel their mechanisms.

EXPERIMENTAL APPROACH

Fibrogenic effects of retinoids were assessed in NRK-49F renal fibroblasts using picro-Sirius red staining for collagens and quantified by spectrophotometric analysis of the eluted stain. Other methods included RT-qPCR, immunoassays and matrix metalloproteinase (MMP) activity assays.

KEY RESULTS

With or without TGF-β1, tRA was dose-dependently pro-fibrotic, notably increasing collagen accumulation. tRA and TGF-β1 additively suppressed expression of mRNA for MMP2, 3 and 13 and suppressed MMP activity. tRA, in the presence of TGF-β1, induced plasminogen activator inhibitor-1 (PAI-1) mRNA and they additively induced PAI-1 protein expression. A PAI-1 inhibitor, a pan-retinoic acid receptor (RAR) antagonist and a pan-retinoid X receptor (RXR) antagonist each partially prevented the pro-fibrotic effect of tRA. The dose-dependent pro-fibrotic effects of a pan-RXR agonist were similar to those of tRA. A pan-RAR agonist showed weaker, less dose-dependent pro-fibrotic effects and the pro-fibrotic effects of RARα and RARβ-selective agonists were even smaller. An RARγ-selective agonist did not affect fibrogenesis.

CONCLUSIONS AND IMPLICATIONS

An in vitro model for the pro-fibrotic effects of retinoids was established in NRK-49F cells. It was associated with reduced MMP activity and increased PAI-1 expression, and was probably mediated by RXR and RAR. To avoid or antagonize the pro-fibrotic activity of tRA, further studies on RAR isotype-selective agonists and PAI-1 inhibitors might be of value.

Keywords: fibroblasts, kidney, fibrosis, retinoic acid, retinoids, TGF-β, PAI-1, MMP, RAR, RXR

Introduction

Chronic kidney disease (CKD) represents a significant health problem in the United Kingdom and worldwide (El Nahas and Bello, 2005). Among all predictors of CKD progression, tubulointerstitial fibrosis (TIF), characterized by excessive accumulation of collagens and other extracellular matrix (ECM) proteins in the renal interstitial compartment, is regarded the common pathway to end-stage renal disease (Nangaku, 2004).

It is now known that TIF is mediated by fibrogenic factors such as TGF-β1, which lead to increased production and/or reduced degradation of ECM by effector cells, especially fibroblasts (Liu, 2011). ECM degradation occurs via two major pathways, the serine protease plasminogen–plasmin system and the matrix metalloproteinase (MMP) system. Inhibitors of these systems include plasminogen activator inhibitor-1 (PAI-1) for the former and tissue inhibitors of metalloproteinases (TIMPs) for the latter (Catania et al., 2007; Eddy, 2009).

PAI-1 is an inhibitor of plasminogen activators, which convert plasminogen to plasmin. The latter can degrade matrix proteins directly or indirectly through activating MMPs (Eddy, 2009). PAI-1 has been proposed as an important mediator of TIF: (i) it is up-regulated in many CKDs with progressing TIF (Eddy, 2009); (ii) its non-inhibitory mutant inhibits renal collagen deposition in experimental nephritis and diabetic nephropathy (Huang et al., 2003; 2008); (iii) its deficiency reduces TIF in unilateral ureteric obstruction (UUO; Oda et al., 2001) and TGF-β1 transgenic mice (Krag et al., 2005); and (iv) UUO induces more fibrosis in PAI-1 transgenic mice than in wild-type mice (Matsuo et al., 2005).

The MMPs, a family of zinc-dependent ECM-degrading enzymes, have been implicated in the pathogenesis of TIF and are regarded as potential therapeutic targets (Catania et al., 2007). However, roles for MMPs are likely to be more complex. For example, MMP-2 has been proposed to mediate macrophage infiltration (Nishida et al., 2007) and epithelial-to-mesenchymal transition (Cheng and Lovett, 2003), thus promoting inflammation and fibrosis.

Retinoids are vitamin A and its derivatives, including all-trans retinoic acid (tRA) and 9-cis retinoic acid (9-cis RA). While tRA is a pan-agonist of all the three isotypes (α, β and γ) of RA receptors (RARs), 9-cis RA is a pan-agonist of not only RARs, but also retinoid X receptors (RXRs; Chambon, 1996; receptor nomenclature follows Alexander et al., 2011). Pharmacologically, retinoids have shown promise in treating CKD models, with a suggested anti-fibrotic potential (Xu et al., 2004). However, more recently, dose-dependent pro-fibrotic effects of retinoids were reported in experimental models of CKD, although the cellular and molecular mediators remained elusive (Morath et al., 2009; Xu et al., 2010). If retinoids are to be considered further for the treatment of CKD, their pro-fibrotic effects must be better defined.

We hypothesized that tRA was pro-fibrotic by activating renal fibroblasts. This hypothesis was supported by our in vitro findings of a dose-dependent, net pro-fibrotic effect of tRA in NRK-49F normal rat kidney fibroblasts, which was associated with reduced MMP activity and increased PAI-1 expression. Studies of RXR and RAR agonists and antagonists indicated that tRA was likely to be acting through retinoid receptor-dependent pathways and that isotype-selective RAR agonists may have reduced pro-fibrotic activities.

Methods

Cell culture

NRK-49F normal rat kidney fibroblasts (LGC Standards, Teddington, UK) were maintained in DMEM (PAA Laboratories GmbH, Pasching, Austria) supplemented with 5% fetal calf serum (FCS; Sigma-Aldrich Company Ltd., Gillingham, UK), penicillin 100 IU·mL−1, streptomycin 100 μg·mL−1 (PAA Laboratories GmbH) and amphotericin B 2.5 μg·mL−1 (Invitrogen, Paisley, UK) under humidified conditions at 37°C and 5% CO2. A human foreskin fibroblast primary culture (a kind gift from Dr. Carole Yee, National Institutes of Health, Bethesda, MD, USA) was maintained in DMEM supplemented with 10% FCS and antibiotics and anti-fungals as reported before (Xu et al., 2007). Cells between passages 8 and 12 were used for experiments. NRK-49F cells and human foreskin fibroblasts showed typical morphological appearances of fibroblasts and stained positive for the fibroblast marker vimentin, but negative for the epithelial marker cytokeratin.

Experimental design

NRK-49F cells and human foreskin fibroblasts were cultured in DMEM supplemented with 2.5% FCS and 2.5% Nu-Serum serum replacement (BD Biosciences, Oxford, UK) for 3 days followed by DMEM with insulin-transferrin-selenium supplement (ITS; Sigma-Aldrich) for 4 days as previously defined (Grotendorst et al., 2004). Cells, with or without pretreatment with retinoid receptor antagonists and the PAI-1 inhibitor tiplaxtinin, were then treated with retinoids with or without 5 ng·mL−1 TGF-β1 in DMEM supplemented with ITS for 48 h. Vehicle-treated cells were included as controls.

In vitro model of fibrosis

An in vitro model of fibrosis was used to quantify global fibrogenesis (Xu et al., 2007). In brief, 1×104 cells per well cultured in collagen type I-coated 96-well plates (BD Biosciences) were treated with TGF-β1 and other compounds as described earlier, fixed in ice-cold methanol overnight at −20°C, washed with Dulbecco's PBS and then incubated with 0.1% picro-Sirius red (PSR; Sigma-Aldrich) for 4 h. Following rinses in 0.1% acetic acid (VWR International Ltd., Lutterworth, UK), plates were air-dried and examined by photomicroscopy. For quantification of PSR staining, 200 μL 0.1 M sodium hydroxide (Sigma-Aldrich) was added and the plate rocked on a rocking platform for 2 h to elute the stain. Optical density was determined using a spectrophotometer (Prior Laboratory Supplies Ltd., East Sussex, UK) at 540 nm.

Cytotoxicity assays

Cell detachment index was recorded to estimate cytotoxicity and to ensure the reliability of subsequent PSR staining results, which required minimal cell (and matrix) detachment and disruption of cell monolayer. Scores of 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5 and 4 represented an area of approximately 0, 5, 10, 20, 30, 40, 60, 80 and 100%, respectively, of the total adherent surface not covered by cells. Furthermore, LDH release was determined by CytoTox 96® Non-Radioactive Cytotoxicity Assay according to manufacturer's instructions (Promega, Southampton, UK).

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA was extracted using the Qiagen RNeasy Mini Kit (Qiagen, Crawley, UK) according to manufacturer's instructions. The RT2 First Strand Kit (Tebu-bio, Peterborough, UK) and the Omniscript Reverse Transcription Kit (Qiagen) were used in reverse transcription for qPCR array and standard qPCR studies, respectively. In a pilot screening, a PCR array kit for rat ECM and adhesion molecules (Qiagen) was used according to the manufacturer's instructions on an ABI7900HT qPCR system (Applied Biosystems, Warrington, UK). For standard RT-qPCR, cDNA was mixed with 20 × TaqMan primer and probe mixtures, TaqMan Universal Master Mix (Applied Biosystems) and water, and qPCR was run on an ABI7900HT system. Where TaqMan gene expression assays were unavailable, SABiosciences RT2 qPCR primer assays were used with RT2 SYBER Green/ROX 2 × qPCR Master Mix (Qiagen). Supporting Information Table S1 lists the gene expression assays used in this study. Results were presented as fold change of gene expression relative to the untreated control sample and normalized to GAPDH (2-ΔΔCT; Livak and Schmittgen, 2001).

Immunocytochemistry and photomicroscopy

NRK-49F cells cultured in 35 mm dishes were fixed with 3.7% paraformaldehyde (Sigma-Aldrich) on ice for 10 min, washed and then permeabilized with 0.1% Triton X-100 (Sigma-Aldrich) in PBS for 5 min. Cells were washed again and then blocked with 1% BSA (PAA Laboratories GmbH) for 1 h, before incubation with primary antibody (Supporting Information Table S2) in PBS for 1 h. After three washes, cells were incubated with 1:1000 diluted secondary antibody labelled with Alexa Fluor 488 (Invitrogen) overnight at 4°C. Cells were again washed and counterstained with 1 μg·mL−1 DAPI in PBS, followed by further washing and examination under a Nikon Eclipse TE2000-S epifluorescence microscope equipped with a standard RGB filter wheel (Nikon Instruments Europe B.V., Amstelveen, The Netherlands). Images were captured with a DXM1200F Nikon digital camera (Nikon UK Limited, Surrey, UK), processed and merged with Adobe Photoshop (Adobe Systems Europe Ltd., Uxbridge, UK).

Protein extraction

Conditioned media of NRK-49F cells cultured in 10 cm dishes were stored at −80°C until use. 500 μL PBS-TDS cell lysis buffer without (for MMP activity assays) or with (for all other applications) protease inhibitors (leupeptin, pepstatin, PMSF and EDTA, all Sigma-Aldrich) was added to the cell monolayer and the dishes incubated on ice for 1 h. Total cell lysate was transferred to a microcentrifuge tube on ice and sheared through a 25 G needle using a syringe. Samples were then centrifuged and supernatants collected and stored at −80°C until use. Protein concentration of samples was determined using the bicinchoninic acid method (Pierce Protein Research Products, Thermo Scientific, Rockford, IL, USA).

Western blotting

Total cell lysates, biotinylated protein ladder (Cell Signalling Technology, Danvers, MA, USA) and Amersham's full range rainbow molecular weight marker (GE Healthcare, Little Chalfont, UK) were separated by routine SDS-PAGE, and protein was transferred onto Amersham's Hybond C-super nitrocellulose membrane (GE Healthcare). Membrane was blocked in 5% skimmed milk blocking solution for 2 h and then incubated with primary antibody (Supporting Information Table S2) overnight at 4°C on a rocking platform. Following washing with Tris-buffered saline with 0.1% Tween 20 the membrane was incubated with secondary antibody (Supporting Information Table S2) conjugated to HRP for 1 h and then washed. The membrane was developed using Amersham's ECL Plus Western Blotting detection Reagents (GE Healthcare) and exposed to Amersham's Hyperfilm ECL chemiluminescence film (GE Healthcare). Band densitometry was analysed using Adobe Photoshop. Results were normalized to a loading control.

MMP activity assay

The Molecular Probes EnzChek Gelatinase/Collagenase Assay Kit (Invitrogen) was used to determine MMP activity in total cell lysates according to the manufacturer's instructions. In brief, 100 μL total cell lysates were added to each assay well in a 96-well assay plate and mixed with 20 μL 1 mg·mL−1 DQ™ gelatin and 80 μL 1× reaction buffer. Reaction buffer alone was used as a negative control and Clostridium collagenase as a positive control. Reactions were performed in triplicate and the plate was incubated at room temperature protected from light for 2 h. Fluorescence intensity was measured using a BioTek FLx800 fluorescence microplate reader (BioTek UK, Potton, UK) at an absorption of 485 nm and fluorescence emission detection of 530 nm. Results were corrected for background fluorescence by subtracting the value derived from the negative control.

Data analysis

PCR array data were analysed using SABiosciences qPCR Array Data Analysis Web portal (Qiagen). All other data were analysed using GraphPad Prism software (GraphPad Software, San Diego, CA, USA). Parametric data were analysed using a paired t-test when comparing two groups. A repeated measures anova with Tukey's post-test was used for analysis of multiple groups. Results presented as fold change were logarithmically transformed before statistical analysis. P < 0.05 was defined as statistically significant.

Materials

Human platelet TGF-β1 (R&D Systems, Abingdon, UK) was reconstituted in sterile 4 mM HCl and 0.1% BSA to make a stock solution of 10 ng·μL−1. tRA (Sigma-Aldrich, Gillingham, UK) and (E)-4-[2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1-propenyl] benzoic acid (TTNPB; a kind gift from R.A.S. Chandraratna, Irvine, California, USA) were dissolved in 100% ethanol. AGN194204, AGN193109 (kind gifts from R.A.S. Chandraratna), HX531 (a kind gift from H. Kagechika, Tokyo, Japan), CD437, CD2019, AGN195183 (synthesized in-house by the J Corcoran group) and tiplaxtinin (Axon Medchem, Groningen, The Netherlands) were dissolved in DMSO. All aliquots were stored at −80°C until use.

The dissociation constants (Kd) and EC50 of pan-RAR agonist TTNPB are 2.5, 2.7 and 1.8 nM and 10, 3.5 and 2.5 nM for mouse RARα, RARβ and RARγ, respectively (Pignatello et al., 1997; 1999). RARα-selective agonist AGN195183 binds to RARα with a Kd of 3 nM and has an EC50 of 200 nM with negligible binding to RARβ and RARγ (Beard et al., 2002). RARβ-selective agonist CD2019 binds to RARβ with a Kd of 26 nM (compared with 920 nM and 160 nM for RARα and RARγ) and has an EC50 of 3.80 nM for RARβ (compared with 19.80 nM and 46.90 nM for RARα and RARγ; Bernard et al., 1992). RARγ-selective agonist CD437 binds to RARγ with a Kd of 77 nM (compared with 6500 nM and 2480 nM for RARα and RARβ) and has an EC50 of 7.3 nM for RARγ (compared with 140 nM and 28.40 nM for RARα and RARβ; Bernard et al., 1992). Pan-RXR agonist AGN194204 has Kd values of 0.4, 3.6 and 3.8 nM and EC50 of 0.2, 0.8 and 0.08 nM for RXRα, RXRβ and RXRγ, respectively. It does not bind to RARs (Vuligonda et al., 2001). Pan-RAR antagonist AGN193109 binds to RARα, RARβ and RARγ with Kd values of 2, 2 and 3 nM, respectively (Johnson et al., 1995). Pan-RXR antagonist HX531 inhibits both RXR homodimers and RXR heterodimers and has an IC50 of 0.018 nM (Ebisawa et al., 1999). The Kd of the PAI-1 inhibitor tiplaxtinin is 480 nM and its IC50 is 2.7 μM (Elokdah et al., 2004).

Results

Net pro-fibrotic effect of tRA in NRK-49F cells and human foreskin fibroblasts

In NRK-49F cells, tRA dose-dependently increased PSR staining for total collagen, with or without 5 ng·mL−1 TGF-β1 (Figure 1A, B). Up to 5 μM tRA alone did not increase LDH release, but with TGF-β1, 5 μM tRA did (Figure 1C). 0.02–5 μM tRA did not increase cell detachment with and without TGF-β1 (Figure 1D), but higher concentrations of tRA (e.g. 10 μM) induced cell detachment and LDH release, indicating cytotoxicity (data not shown). Immunocytochemistry showed that both with and without 5 ng·mL−1 TGF-β1, 2 μM tRA induced collagen types I and III (Figure 1E), while no collagen type IV was detected in any groups (data not shown). Of note, the pro-fibrotic effect of tRA was not unique to NRK-49F cells. A similar dose-dependent pro-fibrotic effect of tRA was also observed in human foreskin fibroblasts (Supporting Information Figure S1A, B).

Figure 1.

Figure 1

Open in a new tab

tRA promoted fibrogenesis in NRK-49F cells in the absence and presence of TGF-β1, as indicated by collagen deposition. NRK-49F cells were treated with or without 0.02–5 μM tRA and 5 ng·mL−1 TGF-β1 for 48 h. Effects on fibrogenesis were illustrated by representative PSR staining images, which visualized total collagen deposition in red colour (×100 magnification) (A); eluted PSR was quantified by spectrophotometric analysis (B); cytotoxicity was assessed by LDH release (C) and cell detachment (D). Immunocytochemical analysis was used to determine protein expression of collagen types I and III in cells treated without and with 2 μM tRA and 5 ng·mL−1 TGF-β1 (fluorescence microscopy ×100 magnification) (E). Data in panels A and E are representative of three independent experiments; data in panels B–D are mean ± SEM of three or more independent cell culture studies. *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle (0.1% ethanol) group; ψP < 0.05, ψψP < 0.01 versus TGF-β1-treated group.

tRA and TGF-β1 down-regulated MMP expression and activity in NRK-49F cells

RT-qPCR array analysis of mRNA expression in a single biological study showed that, both with and without TGF-β1, tRA tended to suppress many and induce a few MMPs, while it tended to also suppress TIMPs (Supporting Information Figure S2). Since these data were from a single biological study, standard RT-qPCR was also performed for MMPs-2, -3 and -13. tRA and TGF-β1 reduced MMP-3 and MMP-13 mRNA expression at 24 and 48 h; although MMP-2 mRNA was reduced at 24 h, complex changes were observed at 48 h (Figure 2B, C and D). To determine net MMP activities, we examined total cell lysates of NRK-49F cells subjected to different treatments. Although tRA and TGF-β1 both significantly reduced MMP activity, no additive effect was observed in the combined treatment group (Figure 2A).

Figure 2.

Figure 2

Open in a new tab

tRA affected MMP activity and MMP-2, MMP-3 and MMP-13 mRNA expression in NRK-49F cells. A. NRK-49F cells were treated with and without 2 μM tRA and 5 ng·mL−1 TGF-β1 for 48 h and total cell lysates were subjected to MMP activity assay. Data were normalized to vehicle-treated group (Vehicle) and are shown as means ± SEM from three independent experiments. ***P < 0.001 versus vehicle-treated group. B–D. NRK-49F cells were treated with and without 2 μM tRA and 5 ng·mL−1 TGF-β1 for the last 6, 24 and 48 h. RNA was extracted and subjected to RT-qPCR assays for the mRNAs of MMP-2 (B), MMP-3 (C) and MMP-13 (D). Results were normalized to GAPDH and are presented as fold change compared with 0 h. Data represent mean ± SEM of three independent cell culture studies. *P < 0.05, **P < 0.01, ***P < 0.001 in comparison with 0 h; +P < 0.05, ++P < 0.01, +++P < 0.001 between groups indicated.

tRA and TGF-β1 additively induced PAI-1, which might play a role in tRA/TGF-β1 dual treatment induced fibrogenesis in NRK-49F cells

In view of the important role for PAI-1 in renal fibrosis its mRNA expression was examined by RT-qPCR. TGF-β1 caused significant increases in PAI-1 mRNA expression at 6 h and 24 h, which showed a trend of further increases in the combined treatment group (Figure 3A). In total cell lysates, both tRA and TGF-β1 induced a trend towards an increase in PAI-1 protein levels and combined treatment significantly increased PAI-1 protein expression (Figure 3Bi, 3Bii). In conditioned media, tRA, TGF-β1 and combined treatment all significantly increased PAI-1 protein levels (Figure 3Bi, 3Bii).

Figure 3.

Figure 3

Open in a new tab

tRA- and TGF-β1- induced PAI-1 expression and the effect of the PAI-1 inhibitor tiplaxtinin on tRA-induced fibrogenesis in NRK-49F cells. A. NRK-49F cells were treated in the last 0, 6, 24, 48 h with 2 μM tRA, 5 ng·mL−1 TGF-β1 or both. Total RNA was extracted and subjected to RT-qPCR for PAI-1 mRNA, with GAPDH used as a loading control. B. NRK-49F cells were treated with or without 2 μM tRA and 5 ng·mL−1 TGF-β1 for 48 h. Total cell lysates and conditioned media were collected for Western blot analysis, results of which are shown representatively in Bi; the PAI-1 bands of three independent experiments were densitometrically analysed, normalized to the GAPDH band and results were presented as fold change compared with vehicle-treated group (Bii). C. NRK-49F cells were treated with 0–2.5 μM tiplaxtinin, with and without 2 μM tRA and 5 ng·mL−1 TGF-β1 for 48 h. The effect on total collagen deposition was determined by PSR staining and quantified by spectrophotometric analysis of eluted PSR (Ci). Cytotoxicity was assessed by LDH release (Cii) and cell detachment index (Ciii). Data represent mean ± SEM of 3 independent cell culture studies. *P < 0.05, **P < 0.01, ***P < 0.001 in comparison with 0 h/vehicle-treated group; ψψP < 0.01, ψψψP < 0.001 versus TGF-β1-treated group; ΩP < 0.05 versus tRA-only treated group; ΦP < 0.05 versus combined treatment (Dual) group only.

To examine if tRA-induced fibrogenesis could be prevented by PAI-1 inhibition, cells were pretreated for 2 h with the PAI-1 inhibitor tiplaxtinin before addition of tRA and/or TGF-β1 for 48 h. As expected, a significant increase in fibrogenesis was induced by tRA with and without TGF-β1 (Figure 3Ci). Low concentrations (1 and 2.5 μM) of tiplaxtinin alone had no effect on fibrogenesis in vehicle and TGF–β1-treated cells; however, 2.5 μM tiplaxtinin significantly reduced the increase in PSR staining caused by tRA and combined treatment (Figure 3Ci). As 2.5 μM tiplaxtinin also increased LDH release (Figure 3Cii), the relationship between suppression of PAI-1, suppression of fibrogenesis and induction of cytotoxicity needs further examination.

tRA-induced fibrogenesis was partially inhibited by pan-RAR and pan-RXR antagonists

As shown in Supporting Information Figure S3, NRK-49F cells expressed both RAR and RXR, including all three RAR isotypes. To test the possible roles of RAR and RXR in tRA-induced fibrogenesis, we preincubated cells with a pan-RAR antagonist AGN193109 or a pan-RXR antagonist HX531 before tRA and/or TGF-β1 treatment. To abolish RAR and RXR activity of tRA or 9-cis RA, a 5–10 fold excess of AGN193109 and HX531 has been recommended (Johnson et al., 1995; Ebisawa et al., 1999; Xu et al., 2002). We found that neither 1 μM AGN193109 nor 1 μM HX531 had any effects on TGF-β1-induced fibrogenesis, however, they both significantly suppressed fibrogenesis induced by combined treatment with tRA (0.2 μM) and TGF-β1 (Figure 4Ai, Bi). Neither antagonist caused significant cell detachment (Figure 4Aii, Bii).

Figure 4.

Figure 4

Open in a new tab

Effects of pan-RAR and pan-RXR antagonists on tRA-induced fibrogenesis in the presence of TGF-β1 in NRK-49F cells. NRK-49F cells were pretreated with 1 μM pan-RAR antagonist AGN193109, 1 μM pan-RXR antagonist HX531 or vehicle for 2 h before treatment with 5 ng·mL−1 TGF-β1 with or without 0.2 μM tRA for 48 h. Fibrogenesis was indicated by changes of total collagen deposition as quantified by spectrophotometric analysis of eluted PSR staining (Ai, Bi). Cytotoxicity was assessed by cell detachment index (Aii, Bii). Data represent mean ± SEM of three independent cell culture studies. *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle-treated group; ψP < 0.05, ψψψP < 0.001 versus TGF-β1-treated group; +P < 0.05, ++P < 0.01 between groups indicated.

Effects of pan-RAR and pan-RXR agonists on fibrogenesis in NRK-49F cells

We next examined the effects of a pan-RAR agonist TTNPB and a pan-RXR agonist AGN194204 on fibrogenesis with and without TGF-β1. We found that without TGF-β1, TTNPB caused a small but significant increase in fibrogenesis at the highest dose tested (2 μM), in contrast to tRA, which was pro-fibrotic at ten-fold lower concentrations (Figure 5Ai). In the presence of TGF-β1, 0.02–2 μM TTNPB caused a significant, non-dose-dependent increase in PSR staining, similar to the effects of 0.02 μM tRA (Figure 5Ai), without causing significant cell detachment (Figure 5Aii). 0.02–2 μM AGN194204 induced a dose-dependent increase in PSR staining both without and with TGF-β1, similar to that observed with tRA (Figure 5Bi). In the presence but not the absence of TGF-β1, 2 μM AGN194204 significantly increased cell detachment, suggestive of cytotoxicity (Figure 5Bii).

Figure 5.

Figure 5

Open in a new tab

Effects of synthetic RAR and RXR agonists on fibrogenesis in the absence and presence of TGF-β1 in NRK-49F cells. NRK-49F cells were treated for 48 h with vehicle or different concentrations of synthetic RAR and RXR agonists, in the absence and presence of 5 ng·mL−1 TGF-β1 and then subjected to quantification of fibrogenesis by spectrophotometric analysis of eluted PSR staining (Ai, Bi and Ci). Cytotoxicity and the integrity of the cell monolayer were assessed by cell detachment index (Aii, Bii and Cii). A. Pan-RAR agonist TTNPB; B. Pan-RXR agonist AGN194204; C. Isotype-selective RAR agonists, including RARα-selective agonist AGN195183, RARβ-selective agonist CD2019 and RARγ-selective agonist CD437. Data represent mean ± SEM of at least three independent cell culture studies. *P < 0.05, **P < 0.01, ***P < 0.001 versus vehicle-treated group; ψP < 0.05, ψψP < 0.01, ψψψP < 0.001 versus TGF-β1-treated group; +P < 0.05, ++P < 0.01, +++P < 0.001 between groups indicated.

Effects of RAR isotype-selective agonists on fibrogenesis in NRK-49F cells

As a pan-RAR agonist showed smaller pro-fibrotic effects, we further explored the effects of RAR isotype-selective agonists on fibrogenesis. As shown in Figure 5Ci, with or without TGF-β1, 0.2 μM RARα-selective agonist AGN195183 induced significant fibrogenesis. 0.2 and 2 μM RARβ-selective agonist CD2019 did not cause any changes in fibrogenesis alone, however in the presence of TGF-β1 both concentrations mildly, but significantly, increased fibrogenesis. Finally, both with or without TGF-β1, 0.2 and 2 μM RARγ-selective agonist CD437 did not affect fibrogenesis. These isotype-selective RAR agonists did not induce any significant cell detachment (Figure 5Cii).

Discussion

Previous work has identified dose-dependent pro-fibrotic effects of tRA and other retinoids in animal models of CKD, but the cellular and molecular mediators of such an unwanted effect remained elusive. In the present study, a novel in vitro model for the dose-dependent pro-fibrotic effects of tRA and other retinoids has been established in a kidney fibroblast cell line, supporting our hypothesis that the pro-fibrotic effects of retinoids might be, at least in part, mediated by their effects on renal fibroblasts. It has also allowed us to identify potential mechanisms behind the pro-fibrotic effect of tRA, including modulation of the expression and activities of MMP and PAI-1, as well as activation of RXR and RAR.

To establish an in vitro model of the pro-fibrotic effect of retinoids, one must first try to understand the seemingly opposite effects of retinoids on fibrogenesis documented in the literature. Some retinoids have been suggested to be anti-fibrotic according to their effects on molecular markers of fibrogenesis. For example, in human lung fibroblasts, tRA (but not 9-cis RA) decreased the steady-state level of α1(I) collagen mRNA (Krupsky et al., 1994), and suppressed collagen type I and III basal synthesis, while tRA, 13-cis RA, 9-cis RA and a collection of other retinoids inhibited TGF-β1-stimulated type I collagen production (Redlich et al., 1995). In human keratinocytes, retinol, retinal, tRA and 13-cis RA all suppressed collagen type VII mRNA and protein expression (Chen et al., 1997). tRA, 9-cis RA, synthetic RAR agonist TTNPB and RXR agonist AGN194204 have also been reported to reduce synthesis of collagen type I and III and fibronectin in rat hepatic stellate cells (Hellemans et al., 2004). However, in activated hepatic stellate cells, although tRA inhibited synthesis of procollagens type I, III and IV, as well as fibronectin and laminin, 9-cis RA increased procollagen I mRNA without affecting the expression of other matrix proteins (Hellemans et al., 1999). The likely ligand type- and cell condition-dependent effects of retinoids have also been reported in skin fibroblasts (Shigematsu and Tajima, 1995; Meisler et al., 1997). In scleroderma and normal human skin fibroblasts, tRA and 13-cis RA dose-dependently reduced procollagen production and type I and type III procollagen mRNA expression (Ohta and Uitto, 1987), but in growth-inhibited human skin fibroblasts, tRA induced production of fibronectin, thrombospondin, laminin and collagen type I (Varani et al., 1990). These in vitro findings are of in vivo relevance because, in photodamaged skin, tRA induced collagen types I and III synthesis (Schwartz et al., 1991; Griffiths et al., 1993). Thus, selectivity of retinoids, cell type specificity, choice of pro-fibrotic marker, as well as dose- and condition-dependent factors could all be important in explaining these seemingly opposing effects of retinoids.

To establish an in vitro model for the role of retinoids in TIF, we chose NRK-49F renal fibroblasts for the following reasons: (i) this cell type is highly related to TIF and has been widely used in fibrogenesis studies, especially those induced by TGF-β (Roberts et al., 1986; Grotendorst et al., 2004) and (ii) the culture conditions to minimize cell proliferation and maximize the fibrogenic response to TGF-β have already been established (Grotendorst et al., 2004) and used in our efforts to establish TGF-β-induced in vitro models of fibrosis in NRK-49F cells (Hu et al., 2009; Wong et al., 2012). In addition, we also screened a number of other cells, which failed to produce a useful fibrotic model due to insignificant increase in PSR staining in response to TGF-β1, with either low or high PSR basal staining (data not shown). The exception was the human foreskin fibroblasts, which showed responses to both tRA and TGF-β1, similar to those of NRK-49F cells (Supporting Information Figure S1). Given the aforementioned condition-dependent effect of retinoids in skin fibroblasts, the pro-fibrotic effect of tRA in human foreskin fibroblasts in our particular conditions could be a useful model for investigating the pro-fibrotic effects of retinoids in skin disorders, if any in vivo relevance is established. Likewise, retinoids have been reported to either inhibit or promote liver fibrosis in different animal models (Okuno et al., 1997; Wang et al., 2007) and in hepatic stellate cells, natural and synthetic retinoids are reported to regulate TGF-βl activation, proliferation and synthesis of different ECM proteins (Okuno et al., 1997; Hellemans et al., 2004). However, the net effects of these retinoids on global fibrogenesis and how retinoids could both suppress and potentiate fibrogenesis in hepatic stellate cells are unknown. Thus, it is possible that some of our observations could shed light not only on the dual potential of retinoids in CKD, but also in liver fibrosis.

As demonstrated in Supporting Information Figure S2, tRA and/or TGF-β1 can differentially regulate different markers of fibrosis at the mRNA level. Thus, by selecting only a few molecular markers for measurements, quite different conclusions could be drawn. To avoid this, we visualized total collagen deposition through clinically relevant PSR staining and further quantified it by spectrophotometric analysis, allowing us to define changes of net fibrogenesis (Figure 1A, B), as earlier (Xu et al., 2007; Hu et al., 2009). This was further supported by immunostaining of collagen types I and III (Figure 1E).

Using NRK-49F cells cultured in defined conditions and PSR staining-based fibrogenesis assays, we established dose-dependent, net pro-fibrotic effects of tRA both in the absence and presence of TGF-β1. In view of the dose-dependent pro-fibrotic effects of tRA in TGF-β1 transgenic mice (Xu et al., 2010) and the synergistic and/or additive effects of tRA and TGF-β1 in inducing net collagen accumulation (Figure 1B), suppressing MMPs (Figure 2) and inducing PAI-1 expression (Figure 3B) in this study, it seems appropriate to call for special caution when tRA is prescribed for CKD patients. This patient group often have increased renal TGF-β1 activity and are thus more prone to fibrosis, therefore they may be more at risk if prescribed tRA either in clinical trials for the treatment of CKD or in other prescriptions for well established indications for tRA. In these in vivo and in vitro experiments, the pro-fibrotic effect of tRA increased with dose. Thus avoiding over-dose is the first step to consider if pro-fibrotic adverse effects and other dose-related toxicities are to be prevented in CKD.

The pro-fibrotic effects of retinoids might, at least in part, be explained by tRA-induced suppression of ECM-degrading enzymes, including both MMP expression and activation. The latter might be the consequence of tRA-induced PAI-1, which suppresses conversion of plasminogen into plasmin, an activator of MMPs. In fact, both tRA- and TGF-β- induced MMP suppression through inhibition of the AP-1 binding sites in MMP genes has been well documented long before (Benbow and Brinckerhoff, 1997). For example, tRA has been reported to suppress ultraviolet irradiation-induced MMPs in human skin (Fisher and Voorhees, 1998) and to suppress plasma MMP activity and MMP-9/TIMP-1 ratio in cultured macrophages in patients with emphysema (Mao et al., 2003). However, the effect of tRA on PAI-1 expression was somehow surprising. Both tRA and 9-cis RA were previously reported to suppress PAI-1 expression in TGF-β-stimulated mesangial cells (Wen et al., 2005; Liu et al., 2008) and in experimental glomerulosclerosis induced by partial nephrectomy (Liu et al., 2011). Furthermore, RAR agonists and antagonists have been previously reported to suppress and potentiate TGF-β-induced PAI-1 promoter activity, respectively (Pendaries et al., 2003). Whether this represents a true cell type-specificity in tRA effect on PAI-1 expression between renal fibroblasts and mesangial cells, two main effectors in TIF and glomerulosclerosis, respectively, deserves further examination. Nevertheless, tRA induction of PAI-1 in NRK-49F cells is somewhat similar to that observed in cultured smooth muscle cells, in which both tRA and 9-cis RA induced PAI-1 mRNA expression in a tyrosine kinase-dependent manner (Watanabe et al., 2002). Furthermore, in our study, the PAI-1 inhibitor tiplaxtinin reduced tRA-induced collagen accumulation supporting the hypothesis that PAI-1 was contributing to the pro-fibrotic effect of tRA (Figure 3C). Because chemical inhibitors can have off-target effects as well as their predicted actions, in future work, the results generated from the use of tiplaxtinin will require confirmation via additional interventions such as another PAI-1 inhibitor and/or gene-silencing.

Our data indicated that RXR and RAR were likely to be the mediators of the pro-fibrotic effects of tRA because: (i) NRK-49F cells expressed these receptors (Supporting Information Figure S3); (ii) pan-RAR and pan-RXR agonists, some RAR isotype-selective agonists (Figure 5), but not RAR and RXR antagonists (Figure 4) showed pro-fibrotic effects; and (iii) RAR and RXR antagonists partially blocked tRA-induced total collagen accumulation (Figure 4).

In this study, RAR and RXR antagonists AGN193109 and HX531 both partially blocked PSR staining induced by tRA and TGF-β1 combined treatment probably by preventing the activation of RARs and RXRs by tRA and its metabolites. Of note, although tRA per se is a pan-RAR agonist, in the presence of isomerases, it is believed to produce metabolites such as 9-cis RA that can activate RXR as well (Urbach and Rando, 1994). We also found that there were differences between retinoids in terms of their pro-fibrotic potency (Figure 5), ranging from no activity (∼0.2–2 μM RARγ-selective agonist CD437) through low (RARα-selective agonist AGN195183, RARβ-selective agonist CD2019) and moderate (pan-RAR agonist TTNPB) to high pro-fibrotic activity (tRA and pan-RXR agonist AGN194204). None of these RAR and RXR agonists showed net anti-fibrotic effects in our system. Nevertheless, given that isotype-selective retinoids have shown promising efficacy in CKD models (Lehrke et al., 2002; Schaier et al., 2004; Zhong et al., 2011), the use of RAR isotype-selective retinoids could be a plausible strategy to avoid unwanted pro-fibrotic effects of tRA.

In summary, we have characterized an in vitro model of the pro-fibrotic effect of retinoids in NRK-49F cells, which is associated with reduced MMPs, increased PAI-1 and activation of the RXRs and RARs (Figure 6). In view of the emerging clinical use of retinoids in patients with CKD (Yamane et al., 2009; Kinoshita and Funauchi, 2012), the fibrogenic effects of retinoids need to be better defined in vitro and in vivo so that the long-term safety profile of retinoids can be improved and eventually, a viable retinoid-based therapy for CKD, devoid of the unwanted pro-fibrotic effects can be developed.

Figure 6.

Figure 6

Open in a new tab

Possible mechanisms of tRA-induced fibrogenesis in this experimental system. tRA-induced pro-fibrotic effect is associated with increased PAI-1 expression and reduced MMP expression. tRA-induced PAI-1 could lead to reduced plasminogen activation into plasmin, which can degrade ECM proteins directly or through activation of MMPs. Thus, both PAI-1 induction and MMP reduction could contribute to decreased degradation and increased accumulation of ECM. Both RXR and RAR pathways are likely to be involved in the pro-fibrotic effects. tRA activation of RXR signalling might be mediated by 9-cis RA converted from tRA by isomerases. In terms of the pro-fibrotic effects, RAR agonists were weak and the RXR agonist AGN194204 was as potent as tRA, suggesting that the pro-fibrotic effects of tRA were mainly through RXR-activating metabolites. Solid lines represent pathways directly supported by our existing data; dashed lines are hypothesized pathways.

Acknowledgments

We are grateful to Kidney Research UK for providing a clinical fellowship to AC Rankin (TF2/2009) and to the King's Renal Research Fund for partially supporting this project. The kind gifts of a human foreskin fibroblast primary culture by Dr. Carole Yee (National Institutes of Health, Bethesda, MD, USA) and synthetic retinoids from Dr. RAS Chandraratna (Allergan, Irvine, CA, USA) and Dr. H Kagechika (University of Tokyo, Tokyo, Japan) are sincerely acknowledged.

Glossary

CKD

chronic kidney disease

ECM

extracellular matrix

ITS

insulin-transferrin-selenium

MMP

matrix metalloproteinase

PAI-1

plasminogen activator inhibitor-1

PSR

picro-Sirius red

RA

retinoic acid

RAR

retinoic acid receptor

RT-qPCR

reverse transcription quantitative polymerase chain reaction

RXR

retinoid X receptor

TIF

tubulointerstitial fibrosis

TIMP

tissue inhibitors of metalloproteinases

tRA

all-trans retinoic acid

TTNPB

(E)-4-[2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1-propenyl] benzoic acid

UUO

unilateral ureteric obstruction

Conflict of interest

None.

Supporting information

Additional Supporting Information may be found in the online version of this article at the publisher's web-site:

http://dx.doi.org/10.1111/bph.12348

Figure S1 Effect of tRA and TGF-β1 on fibrogenesis in human foreskin fibroblasts. Fibrogenesis was indicated by total collagen deposition, which was visualized and quantified by PSR staining and spectrophotometric analysis of eluted PSR. Human foreskin fibroblasts cultured in collagen type-I-coated 96-well plates for 3 days in DMEM supplemented with 2.5% FCS and 2.5% Nu then for 4 days in DMEM supplemented with ITS were treated with increasing doses of tRA in the absence and presence of TGF-β1 for 48 h. The effect on total collagen deposition was determined by photomicroscopy of PSR staining (×100 magnification) (A) and quantified by spectrophotometric analysis of eluted PSR (B). Data represent mean ± SEM of three independent cell culture studies. *** P< 0.001 versus vehicle; ψ P < 0.05, ψψψ P < 0.001 versus TGF-β1-treated group.

Figure S2 RT-qPCR array analysis of a pilot study on the effects of tRA and/or TGF-β1 on mRNA expression of selected molecular markers of fibrosis in NRK-49F cells. NRK-49F cells cultured for 3 days in DMEM supplemented with 2.5% FCS and 2.5% Nu then for 4 days in DMEM supplemented with ITS were treated with 2 μM tRA in the absence and presence of 5 ng·mL−1 TGF-β1 for 48 h. RNA was extracted and subjected to RT-qPCR using an RT-qPCR array. Fold change in mRNA expression of selected collagens, connective tissue growth factor and fibronectin in TGF–β1- and tRA-treated cells versus vehicle-treated cells (Ai, Bi) and (TGF-β1 + tRA) dual-treated cells versus TGF–β1-treated cells (Ci) are shown in the left panels. The right panels are fold-changes in mRNA expression of the MMPs and TIMPs in TGF–β1-treated cells and tRA-treated cells versus vehicle-treated cells (Aii, Bii) and (TGF-β1 + tRA) dual-treated cells versus TGF–β1-treated cells (Cii). Data represent triplicates from a single cell culture study, therefore statistical analysis was not performed.

Figure S3 Expression of RARs and RXRs in NRK-49F cells without and with tRA and TGF-β1 treatment. (A) Total RNA was extracted from NRK-49F cells cultured for 3 days in DMEM supplemented with 2.5% FCS and 2.5% Nu then for 4 days in DMEM supplemented with ITS and was subjected to RT-qPCR analysis of RAR and RXR mRNAs. The mRNA expression of each receptor was relative to the lowest expressed receptor, RXRγ. Data represent mean ± SEM of nine independent experiments. (B) NRK-49F total cell lysate was extracted from cells cultured as before. Western blotting was performed for RARα (Bi), RARβ (Bii) and RARγ (Biii). A431 cell nuclear extract supplied with the RARγ antibody was used as a positive control for RARγ. (C) NRK-49F cells were cultured with 2 μM tRA and/or 5 ng·mL−1 TGF-β1 for 0–48 h. Total RNA was extracted and subjected to RT-qPCR analysis of the mRNA expression of RARα (Ci), RARβ (Cii), RARγ (Ciii), RXRα (Civ), RXRβ (Cv) and RXRγ (Cvi). Data represent mean ± SEM of four independent cell culture studies. * P < 0.05, ** P < 0.01, *** P < 0.001 versus 0 h group; + P < 0.05, ++ P < 0.01 between groups indicated.

Table S1 Gene expression assays used in this study.

Table S2 Antibodies used in this study.

bph0170-1177-SD1.docx (1.1MB, docx)

References

  1. Alexander SP, Mathie A, Peters JA. Guide to Receptors and Channels (GRAC), 5th edition. Br J Pharmacol. 2011;164(Suppl. 1):S1–324. doi: 10.1111/j.1476-5381.2011.01649_1.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Beard RL, Duong TT, Teng M, Klein ES, Standevan AM, Chandraratna RA. Synthesis and biological activity of retinoic acid receptor-α specific amides. Bioorg Med Chem Lett. 2002;12:3145–3148. doi: 10.1016/s0960-894x(02)00647-9. [DOI] [PubMed] [Google Scholar]
  3. Benbow U, Brinckerhoff CE. The AP-1 site and MMP gene regulation: what is all the fuss about? Matrix Biol. 1997;15:519–526. doi: 10.1016/s0945-053x(97)90026-3. [DOI] [PubMed] [Google Scholar]
  4. Bernard BA, Bernardon JM, Delescluse C, Martin B, Lenoir MC, Maignan J, et al. Identification of synthetic retinoids with selectivity for human nuclear retinoic acid receptor γ. Biochem Biophys Res Commun. 1992;186:977–983. doi: 10.1016/0006-291x(92)90842-9. [DOI] [PubMed] [Google Scholar]
  5. Catania JM, Chen G, Parrish AR. Role of matrix metalloproteinases in renal pathophysiologies. Am J Physiol Renal Physiol. 2007;292:F905–F911. doi: 10.1152/ajprenal.00421.2006. [DOI] [PubMed] [Google Scholar]
  6. Chambon P. A decade of molecular biology of retinoic acid receptors. FASEB J. 1996;10:940–954. [PubMed] [Google Scholar]
  7. Chen M, Goyal S, Cai X, O'Toole EA, Woodley DT. Modulation of type VII collagen (anchoring fibril) expression by retinoids in human skin cells. Biochim Biophys Acta. 1997;1351:333–340. doi: 10.1016/s0167-4781(96)00226-6. [DOI] [PubMed] [Google Scholar]
  8. Cheng S, Lovett DH. Gelatinase A (MMP-2) is necessary and sufficient for renal tubular cell epithelial-mesenchymal transformation. Am J Pathol. 2003;162:1937–1949. doi: 10.1016/S0002-9440(10)64327-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ebisawa M, Umemiya H, Ohta K, Fukasawa H, Kawachi E, Christoffel G, et al. Retinoid X receptor-antagonistic diazepinylbenzoic acids. Chem Pharm Bull (Tokyo) 1999;47:1778–1786. doi: 10.1248/cpb.47.1778. [DOI] [PubMed] [Google Scholar]
  10. Eddy AA. Serine proteases, inhibitors and receptors in renal fibrosis. Thromb Haemost. 2009;101:656–664. [PMC free article] [PubMed] [Google Scholar]
  11. El Nahas AM, Bello AK. Chronic kidney disease: the global challenge. Lancet. 2005;365:331–340. doi: 10.1016/S0140-6736(05)17789-7. [DOI] [PubMed] [Google Scholar]
  12. Elokdah H, Abou-Gharbia M, Hennan JK, McFarlane G, Mugford CP, Krishnamurthy G, et al. Tiplaxtinin, a novel, orally efficacious inhibitor of plasminogen activator inhibitor-1: design, synthesis, and preclinical characterization. J Med Chem. 2004;47:3491–3494. doi: 10.1021/jm049766q. [DOI] [PubMed] [Google Scholar]
  13. Fisher GJ, Voorhees JJ. Molecular mechanisms of photoaging and its prevention by retinoic acid: ultraviolet irradiation induces MAP kinase signal transduction cascades that induce AP-1-regulated matrix metalloproteinases that degrade human skin in vivo. J Investig Dermatol Symp Proc. 1998;3:61–68. [PubMed] [Google Scholar]
  14. Griffiths CE, Russman AN, Majmudar G, Singer RS, Hamilton TA, Voorhees JJ. Restoration of collagen formation in photodamaged human skin by tretinoin (retinoic acid) N Engl J Med. 1993;329:530–535. doi: 10.1056/NEJM199308193290803. [DOI] [PubMed] [Google Scholar]
  15. Grotendorst GR, Rahmanie H, Duncan MR. Combinatorial signaling pathways determine fibroblast proliferation and myofibroblast differentiation. FASEB J. 2004;18:469–479. doi: 10.1096/fj.03-0699com. [DOI] [PubMed] [Google Scholar]
  16. Hellemans K, Grinko I, Rombouts K, Schuppan D, Geerts A. All-trans and 9-cis retinoic acid alter rat hepatic stellate cell phenotype differentially. Gut. 1999;45:134–142. doi: 10.1136/gut.45.1.134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hellemans K, Verbuyst P, Quartier E, Schuit F, Rombouts K, Chandraratna RA, et al. Differential modulation of rat hepatic stellate phenotype by natural and synthetic retinoids. Hepatology. 2004;39:97–108. doi: 10.1002/hep.20015. [DOI] [PubMed] [Google Scholar]
  18. Hu Q, Noor M, Wong YF, Hylands PJ, Simmonds MS, Xu Q, et al. In vitro anti-fibrotic activities of herbal compounds and herbs. Nephrol Dial Transplant. 2009;24:3033–3041. doi: 10.1093/ndt/gfp245. [DOI] [PubMed] [Google Scholar]
  19. Huang Y, Haraguchi M, Lawrence DA, Border WA, Yu L, Noble NA. A mutant, non-inhibitory plasminogen activator inhibitor type 1 decreases matrix accumulation in experimental glomerulonephritis. J Clin Invest. 2003;112:379–388. doi: 10.1172/JCI18038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Huang Y, Border WA, Yu L, Zhang J, Lawrence DA, Noble NA. A PAI-1 mutant, PAI-1R, slows progression of diabetic nephropathy. J Am Soc Nephrol. 2008;19:329–338. doi: 10.1681/ASN.2007040510. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Johnson AT, Klein ES, Gillett SJ, Wang L, Song TK, Pino ME, et al. Synthesis and characterization of a highly potent and effective antagonist of retinoic acid receptors. J Med Chem. 1995;38:4764–4767. doi: 10.1021/jm00024a003. [DOI] [PubMed] [Google Scholar]
  22. Kinoshita K, Funauchi M. Therapeutic effect of retinoic acid in lupus nephritis. Nihon Rinsho Meneki Gakkai Kaishi. 2012;35:1–7. doi: 10.2177/jsci.35.1. [DOI] [PubMed] [Google Scholar]
  23. Krag S, Danielsen CC, Carmeliet P, Nyengaard J, Wogensen L. Plasminogen activator inhibitor-1 gene deficiency attenuates TGF-β1-induced kidney disease. Kidney Int. 2005;68:2651–2666. doi: 10.1111/j.1523-1755.2005.00737.x. [DOI] [PubMed] [Google Scholar]
  24. Krupsky M, Fine A, Berk JL, Goldstein RH. Retinoic acid-induced inhibition of type I collagen gene expression by human lung fibroblasts. Biochim Biophys Acta. 1994;1219:335–341. doi: 10.1016/0167-4781(94)90057-4. [DOI] [PubMed] [Google Scholar]
  25. Lehrke I, Schaier M, Schade K, Morath C, Waldherr R, Ritz E, et al. Retinoid receptor-specific agonists alleviate experimental glomerulonephritis. Am J Physiol Renal Physiol. 2002;282:F741–F751. doi: 10.1152/ajprenal.00026.2001. [DOI] [PubMed] [Google Scholar]
  26. Liu X, Lu L, Tao BB, Zhu YC. All-trans retinoic acid inhibits the increases in fibronectin and PAI-1 induced by TGF-β1 and Ang II in rat mesangial cells. Acta Pharmacol Sin. 2008;29:1035–1041. doi: 10.1111/j.1745-7254.2008.00849.x. [DOI] [PubMed] [Google Scholar]
  27. Liu X, Lü L, Tao BB, Zhou AL, Zhu YC. Amelioration of glomerulosclerosis with all-trans retinoic acid is linked to decreased plasminogen activator inhibitor-1 and α-smooth muscle actin. Acta Pharmacol Sin. 2011;32:70–78. doi: 10.1038/aps.2010.200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liu Y. Cellular and molecular mechanisms of renal fibrosis. Nat Rev Nephrol. 2011;7:684–696. doi: 10.1038/nrneph.2011.149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)) method. Methods. 2001;25:402–408. doi: 10.1006/meth.2001.1262. [DOI] [PubMed] [Google Scholar]
  30. Mao JT, Tashkin DP, Belloni PN, Baileyhealy I, Baratelli F, Roth MD. All-trans retinoic acid modulates the balance of matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 in patients with emphysema. Chest. 2003;124:1724–1732. doi: 10.1378/chest.124.5.1724. [DOI] [PubMed] [Google Scholar]
  31. Matsuo S, Lopez-Guisa JM, Cai X, Okamura DM, Alpers CE, Bumgarner RE, et al. Multifunctionality of PAI-1 in fibrogenesis: evidence from obstructive nephropathy in PAI-1-overexpressing mice. Kidney Int. 2005;67:2221–2238. doi: 10.1111/j.1523-1755.2005.00327.x. [DOI] [PubMed] [Google Scholar]
  32. Meisler NT, Parrelli J, Gendimenico GJ, Mezick JA, Cutroneo KR. All-trans-retinoic acid inhibition of Pro α1(I) collagen gene expression in fetal rat skin fibroblasts: identification of a retinoic acid response element in the Pro α1(I) collagen gene. J Invest Dermatol. 1997;108:476–481. doi: 10.1111/1523-1747.ep12289723. [DOI] [PubMed] [Google Scholar]
  33. Morath C, Ratzlaff K, Dechow C, Schwenger V, Schaier M, Zeier B, et al. Chronic low-dose isotretinoin treatment limits renal damage in subtotally nephrectomized rats. J Mol Med (Berl) 2009;87:53–64. doi: 10.1007/s00109-008-0404-5. [DOI] [PubMed] [Google Scholar]
  34. Nangaku M. Mechanisms of tubulointerstitial injury in the kidney: final common pathways to end-stage renal failure. Intern Med. 2004;43:9–17. doi: 10.2169/internalmedicine.43.9. [DOI] [PubMed] [Google Scholar]
  35. Nishida M, Okumura Y, Ozawa S, Shiraishi I, Itoi T, Hamaoka K. MMP-2 inhibition reduces renal macrophage infiltration with increased fibrosis in UUO. Biochem Biophys Res Commun. 2007;354:133–139. doi: 10.1016/j.bbrc.2006.12.165. [DOI] [PubMed] [Google Scholar]
  36. Oda T, Jung YO, Kim HS, Cai X, Lopez-Guisa JM, Ikeda Y, et al. PAI-1 deficiency attenuates the fibrogenic response to ureteral obstruction. Kidney Int. 2001;60:587–596. doi: 10.1046/j.1523-1755.2001.030002587.x. [DOI] [PubMed] [Google Scholar]
  37. Ohta A, Uitto J. Procollagen gene expression by scleroderma fibroblasts in culture. Inhibition of collagen production and reduction of pro α1(I) and pro α1(III) collagen messenger RNA steady-state levels by retinoids. Arthritis Rheum. 1987;30:404–411. doi: 10.1002/art.1780300407. [DOI] [PubMed] [Google Scholar]
  38. Okuno M, Moriwaki H, Imai S, Muto Y, Kawada N, Suzuki Y, et al. Retinoids exacerbate rat liver fibrosis by inducing the activation of latent TGF-β in liver stellate cells. Hepatology. 1997;26:913–921. doi: 10.1053/jhep.1997.v26.pm0009328313. [DOI] [PubMed] [Google Scholar]
  39. Pendaries V, Verrecchia F, Michel S, Mauviel A. Retinoic acid receptors interfere with the TGF-β/Smad signaling pathway in a ligand-specific manner. Oncogene. 2003;22:8212–8220. doi: 10.1038/sj.onc.1206913. [DOI] [PubMed] [Google Scholar]
  40. Pignatello MA, Kauffman FC, Levin AA. Multiple factors contribute to the toxicity of the aromatic retinoid, TTNPB (Ro 13-7410): binding affinities and disposition. Toxicol Appl Pharmacol. 1997;142:319–327. doi: 10.1006/taap.1996.8047. [DOI] [PubMed] [Google Scholar]
  41. Pignatello MA, Kauffman FC, Levin AA. Multiple factors contribute to the toxicity of the aromatic retinoid TTNPB (Ro 13-7410): interactions with the retinoic acid receptors. Toxicol Appl Pharmacol. 1999;159:109–116. doi: 10.1006/taap.1999.8726. [DOI] [PubMed] [Google Scholar]
  42. Redlich CA, Delisser HM, Elias JA. Retinoic acid inhibition of transforming growth factor-β-induced collagen production by human lung fibroblasts. Am J Respir Cell Mol Biol. 1995;12:287–295. doi: 10.1165/ajrcmb.12.3.7873195. [DOI] [PubMed] [Google Scholar]
  43. Roberts AB, Sporn MB, Assoian RK, Smith JM, Roche NS, Wakefield LM, et al. Transforming growth factor type β: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc Natl Acad Sci U S A. 1986;83:4167–4171. doi: 10.1073/pnas.83.12.4167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Schaier M, Liebler S, Schade K, Shimizu F, Kawachi H, Grone HJ, et al. Retinoic acid receptor α and retinoid X receptor specific agonists reduce renal injury in established chronic glomerulonephritis of the rat. J Mol Med (Berl) 2004;82:116–125. doi: 10.1007/s00109-003-0510-3. [DOI] [PubMed] [Google Scholar]
  45. Schwartz E, Cruickshank FA, Mezick JA, Kligman LH. Topical all-trans retinoic acid stimulates collagen synthesis in vivo. J Invest Dermatol. 1991;96:975–978. doi: 10.1111/1523-1747.ep12476385. [DOI] [PubMed] [Google Scholar]
  46. Shigematsu T, Tajima S. Modulation of collagen synthesis and cell proliferation by retinoids in human skin fibroblasts. J Dermatol Sci. 1995;9:142–145. doi: 10.1016/0923-1811(94)00360-q. [DOI] [PubMed] [Google Scholar]
  47. Urbach J, Rando RR. Isomerization of all-trans-retinoic acid to 9-cis-retinoic acid. Biochem J. 1994;299(Pt 2):459–465. doi: 10.1042/bj2990459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Varani J, Mitra RS, Gibbs D, Phan SH, Dixit VM, Mitra R, Jr, et al. All-trans retinoic acid stimulates growth and extracellular matrix production in growth-inhibited cultured human skin fibroblasts. J Invest Dermatol. 1990;94:717–723. doi: 10.1111/1523-1747.ep12876294. [DOI] [PubMed] [Google Scholar]
  49. Vuligonda V, Thacher SM, Chandraratna RA. Enantioselective syntheses of potent retinoid X receptor ligands: differential biological activities of individual antipodes. J Med Chem. 2001;44:2298–2303. doi: 10.1021/jm0100584. [DOI] [PubMed] [Google Scholar]
  50. Wang L, Potter JJ, Rennie-Tankersley L, Novitskiy G, Sipes J, Mezey E. Effects of retinoic acid on the development of liver fibrosis produced by carbon tetrachloride in mice. Biochim Biophys Acta. 2007;1772:66–71. doi: 10.1016/j.bbadis.2006.08.009. [DOI] [PubMed] [Google Scholar]
  51. Watanabe A, Kanai H, Arai M, Sekiguchi K, Uchiyama T, Nagai R, et al. Retinoids induce the PAI-1 gene expression through tyrosine kinase-dependent pathways in vascular smooth muscle cells. J Cardiovasc Pharmacol. 2002;39:503–512. doi: 10.1097/00005344-200204000-00006. [DOI] [PubMed] [Google Scholar]
  52. Wen X, Li Y, Hu K, Dai C, Liu Y. Hepatocyte growth factor receptor signaling mediates the anti-fibrotic action of 9-cis-retinoic acid in glomerular mesangial cells. Am J Pathol. 2005;167:947–957. doi: 10.1016/S0002-9440(10)61185-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wong YF, Qu S, Kong Q, Zhang XL, Liang XM, Hu Q. Knowledge-based discovery of anti-fibrotic and pro-fibrotic activities from Chinese materia medica. In: Kuang X, et al., editors. Recent Advances in Theories and Practice of Chinese Medicine. Rijeka: InTech; 2012. pp. 337–352. Online. [Google Scholar]
  54. Xu Q, Konta T, Furusu A, Nakayama K, Lucio-Cazana J, Fine LG, et al. Transcriptional induction of mitogen-activated protein kinase phosphatase 1 by retinoids. Selective roles of nuclear receptors and contribution to the antiapoptotic effect. J Biol Chem. 2002;277:41693–41700. doi: 10.1074/jbc.M207095200. [DOI] [PubMed] [Google Scholar]
  55. Xu Q, Lucio-Cazana J, Kitamura M, Ruan X, Fine LG, Norman JT. Retinoids in nephrology: promises and pitfalls. Kidney Int. 2004;66:2119–2131. doi: 10.1111/j.1523-1755.2004.66002.x. [DOI] [PubMed] [Google Scholar]
  56. Xu Q, Norman JT, Shrivastav S, Lucio-Cazana J, Kopp JB. In vitro models of TGF-β-induced fibrosis suitable for high-throughput screening of antifibrotic agents. Am J Physiol Renal Physiol. 2007;293:F631–F640. doi: 10.1152/ajprenal.00379.2006. [DOI] [PubMed] [Google Scholar]
  57. Xu Q, Hendry BM, Maden M, Lu H, Wong YF, Rankin AC, et al. Kidneys of Alb/TGF-β1 transgenic mice are deficient in retinoic acid and exogenous retinoic acid shows dose-dependent toxicity. Nephron Exp Nephrol. 2010;114:e127–e132. doi: 10.1159/000276587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yamane A, Tsukamoto N, Saitoh T, Uchiumi H, Handa H, Karasawa M, et al. Successful treatment by all-trans retinoic acid in a patient with acute promyelocytic leukemia complicated by liver cirrhosis and polycystic kidney. Intern Med. 2009;48:1691–1694. doi: 10.2169/internalmedicine.48.2358. [DOI] [PubMed] [Google Scholar]
  59. Zhong Y, Wu Y, Liu R, Li Z, Chen Y, Evans T, et al. Novel retinoic acid receptor α agonists for treatment of kidney disease. PLoS ONE. 2011;6:e27945. doi: 10.1371/journal.pone.0027945. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

bph0170-1177-SD1.docx (1.1MB, docx)

Articles from British Journal of Pharmacology are provided here courtesy of The British Pharmacological Society

RESOURCES