Resveratrol inhibits high glucose-induced PI3K/Akt/ERK-dependent interleukin-17 expression in primary mouse cardiac fibroblasts

Kaliyamurthi Venkatachalam, Srinivas Mummidi, Dolores M. Cortez, Sumanth D. Prabhu, Anthony J. Valente, Bysani Chandrasekar

Abstract

We investigated the expression of the proinflammatory cytokine interleukin (IL)-17 in cardiac fibroblasts and its induction by high glucose (HG). Our results show that primary mouse cardiac fibroblasts (mCFs) secrete low basal levels of IL-17 and that HG (25 mM d-glucose) as opposed to low glucose (5 mM d-glucose + 20 mM mannitol) significantly enhances its secretion. HG induces IL-17 mRNA expression by both transcriptional and posttranscriptional mechanisms. HG induces phosphoinositide 3- kinase [PI3K; inhibited by adenoviral (Ad).dominant negative (dn)PI3Kp85], Akt (inhibited by Ad.dnAkt1), and ERK (inhibited by PD-98059) activation and induces IL-17 expression via PI3K→Akt→ERK-dependent signaling. Moreover, mCFs express both IL-17 receptors A and C, and although IL-17RA is upregulated, HG fails to modulate IL-17RC expression. Furthermore, IL-17 stimulates net collagen production by mCFs. Pretreatment with the phytoalexin resveratrol blocks HG-induced PI3K-, Akt-, and ERK-dependent IL-17 expression. These results demonstrate that 1) cardiac fibroblasts express IL-17 and its receptors; 2) HG upregulates IL-17 and IL-17RA, suggesting a positive amplification loop in IL-17 signaling in hyperglycemia; 3) IL-17 enhances net collagen production; and 4) resveratrol can inhibit these HG-induced changes. Thus, in hyperglycemic conditions, IL-17 may potentiate myocardial inflammation, injury, and remodeling through autocrine and paracrine mechanisms, and resveratrol has therapeutic potential in ameliorating this effect.

  • signal transduction
  • proinflammatory cytokines

fibroblasts are the most abundant cell type in the mammalian heart, residing in the interstitium between contracting cardiomyocytes and providing structural support by regulating extracellular matrix deposition and turnover. They play a central role in the cardiac remodeling and fibrosis that follow pressure overload or myocardial injury, by generating high levels of proinflammatory cytokines, chemokines, and growth factors and by increasing the production of extracellular matrix (ECM) proteins (5, 15, 16, 37, 39, 48). Both experimental and clinical studies have reported that diabetes mellitus (DM) is a causative as well as a compounding factor in cardiomyopathy. DM can result in cardiomyopathy in the absence of coronary artery disease, hypertension, and alcoholism (2, 3). Prolonged hyperglycemia results in left ventricular hypertrophy and cardiac fibrosis characterized by fibroblast migration, proliferation, and deposition of excess amounts of collagen and other ECM proteins (2, 3), indicating that DM is an independent risk factor in myocardial remodeling in vivo. In vitro, high glucose (HG) has also been shown to induce oxidative stress; to activate various oxidative stress-responsive transcription factors; to stimulate proinflammatory cytokine, chemokine, adhesion molecule expression, cell growth, cell proliferation, and cell death; and to induce collagen and fibronectin synthesis (2, 3, 7, 18, 23).

Interleukin-17 (IL-17 or IL-17A) is a proinflammatory cytokine belonging to a new family of cytokines that share little or no homology with other interleukins and that signal via a unique family of IL-17 receptors (21, 26, 27). IL-17 is expressed by activated immune cells, particularly the Th17 subset of T cells, which are important mediators of autoimmune responses (20, 32). Of note, both IL-17 and IL-17-secreting Th17 cells have been shown to play critical roles in β-cell toxicity and Type 1 diabetes (29, 45). Although IL-17 is secreted mainly by Th17 cells, IL-17 receptors are expressed ubiquitously, suggesting that nonimmune cell types are targets of IL-17 signaling (21, 26, 27). However, using multiplex assays, we recently demonstrated the presence of low levels of IL-17 in the medium of primary human cardiac fibroblasts (13). Since these were essentially pure cultures, we hypothesized that nonimmune cells such as fibroblasts may secrete IL-17 and that IL-17 exerts its effects via autocrine and paracrine mechanisms.

Resveratrol, a phytoalexin found largely in the skins of red grapes and other fruits, has been shown to exert both vascular and cardioprotective effects (22, 38, 42). It exerts potent antioxidant, anti-inflammatory, and antiproliferative effects. For example, resveratrol exerts cardioprotective effects in myocardial ischemia-reperfusion injury (38). It inhibits angiotensin II-mediated cardiomyocyte hypertrophy (10) and promotes myocardial angiogenesis. In streptozotocin-induced diabetes, resveratrol improves myocardial dysfunction via enhanced antioxidant activity and reduced blood glucose levels (42). Its therapeutic effects have also been described in an experimental model of myocarditis (49). Furthermore, resveratrol has been shown to exert antiproliferative, antifibrotic, anti-inflammatory, and proapoptotic effects (38, 42, 46).

Given these facts, the purpose of this study was to investigate whether cardiac fibroblasts express IL-17 and, since diabetes is a compounding factor in the progression of cardiomyopathy (2, 3, 24, 35), whether HG induces IL-17 expression and to determine the signaling mechanisms responsible. Furthermore, since resveratrol exerts a number of cardioprotective effects, including antiproliferative and anti-inflammatory effects (22, 38, 42), we also investigated whether resveratrol can inhibit HG-mediated IL-17 expression. Our results demonstrate that HG is a potent inducer of IL-17 mRNA and protein in primary mouse cardiac fibroblasts (mCFs). HG induces IL-17 expression via a phosphoinositide 3-kinase (PI3K)→Akt→ERK-dependent signaling pathway, and that this pathway is inhibited by resveratrol.

MATERIALS AND METHODS

Cell culture.

All animal studies were approved by the Institutional Care and Use Committee and conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. mCFs were isolated from the hearts of 8–10-wk-old male C57Bl/6 mice using a modification of methods developed in our laboratory (11, 12). Briefly, after an induction of deep anesthesia with an injection (0.1 ml im) of ketamine-xylazine (9:1), the hearts were rapidly removed, rinsed, and mounted via the aorta onto a 27-gauge cannula attached to a Langendorff-type apparatus allowing retrograde perfusion of the coronary arteries. Hearts were perfused at 80 mmHg for 5 min with 37°C sterile calcium-free Krebs-Ringer bicarbonate buffer (KRB) containing (in mM) 110 NaCl, 2.6 KCl, 1.2 KH2PO4, 1.2 MgSO4, 25 NaHCO3, and 11 glucose. They were then perfused for 20–25 min with KRB enzyme solution containing 0.5 mg/ml type II collagenase (Worthington Biochemical, Freehold, NJ), 2.5 mM CaCl2, and 1 mg/ml fatty-acid free albumin. After digestion, the ventricles were trimmed free and minced in KRB enzyme solution containing 10 mg/ml albumin, filtered through sterile nylon mesh, and centrifuged at 25 g for 5 min to remove cardiomyocytes, red blood cells, and debris. The resultant supernatant was then centrifuged at 1,000 g for 8 min. The cell pellet was resuspended in 5 ml RPMI 1640 medium with 5 mM glucose, 10% heat-inactivated FBS, and antibiotics (pH 7.3) and plated into T25 tissue-culture flasks (Falcon, Becton-Dickinson Labware, Franklin Lakes, NJ). Nonadherent cells were removed by aspiration after 4 h and discarded. The cardiac fibroblasts from 10 mice were cultured until confluent, pooled, plated into T175 flasks, and fed with fresh medium. On reaching confluency, the pooled cells were then used for experiments. Cell phenotype was investigated by immunofluorescence using the following antibodies; FITC-conjugated monoclonal anti-β-actin (Sigma Chemical, St. Louis, MO), anti-platelet endothelial cell adhesion molecule 1 (CD-31; Research Diagnostics, Flanders, NJ), anti-vimentin (Santa Cruz Biotechnology, Santa Cruz, CA), and anti-smooth muscle actin (Sigma). After two serial passages, >99.7% of cells in these cultures exhibited vimentin and β-actin immunoreactivity, were CD-31 and smooth muscle actin negative, and displayed typical fibroblast-like morphology. Nonfibroblast cells typically accounted for <0.1% of total cells as determined by immunofluorescence. mCFs were used in these experiments between the second and third passages.

At 70% confluency, the cells were made quiescent by incubating in medium containing 0.5% BSA (serum free) for 48 h. The cells were maintained in serum-free medium with 5 mM d-glucose for 24 h, followed by incubation with fresh serum-free medium containing 25 mM d-glucose (HG) for the indicated time periods. Control cells were incubated with 5 mM d-glucose + 20 mM mannitol [osmotic control; low glucose (LG)] for up to 48 h. For dose-response studies, d-glucose was added at 10, 15, and 25 mM for 48 h. l-Glucose (Sigma-Aldrich), the metabolically inactive enantiomer of d-glucose, also served as a control. At the end of the experimental period, culture supernatants were collected and snap frozen. Cells were harvested, snap frozen, and stored at −80°C.

Adenoviral vectors, propagation, and infection.

Recombinant, replication-deficient adenoviral vectors encoding green fluorescent protein (Ad.GFP), dominant negative (dn) Akt1 (Ad.dnAkt1), and dominant negative p85 subunit of class 1 PI3K in which the inner SH2 domain is deleted (Ad.dnPI3Kp85) have all been previously described (34). mCFs were infected as previously described (34). In brief, mCFs were infected in PBS at ambient temperature with the adenoviruses at a multiplicity of infection of 100. After 1 h, the adenovirus-containing medium was replaced with medium containing 0.5% BSA.

IL-17 mRNA expression-RT-PCR, real-time RT-PCR, and Northern blot analysis.

DNA-free total RNA was extracted using the RNAqueous-4PCR kit (Ambion). RNA quality was assessed by capillary electrophoresis using the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). All RNA samples used for quantitative (q)PCR had RNA integrity numbers >9.1 (scale = 1–10), as assigned by default parameters of the Expert 2100 Bioanalyzer software package (v. 2.02). IL-17 mRNA expression was analyzed by Northern blot analysis as previously described (11–13). Twenty-five micrograms of total RNA per lane were denatured, transferred onto nitrocellulose membrane, UV cross-linked, and probed with 32P-labeled cDNA. IL- 17 cDNA was amplified by RT-PCR (GenBank Accession No. NT_039169; 166 bp amplicon), sense, 5′-AGT CCA GGG AGA GCT TCA TC-3′, antisense, 5′-GGG AGT TAA AGA CTT GGT-3′, subcloned into pCR 2.1-TOPO (Invitrogen), and the sequence confirmed by nucleotide sequencing. The insert was excised with EcoR1, gel purified, 32P-labeled, and used for probing the membranes. 28S rRNA served as an internal control. IL-17 mRNA expression was also analyzed by real-time qPCR using the following primers: IL-17 (25), sense primer, 5′-CCA CGT CAC CCT GGA CTC TC-3′ (exon 1; 185–204 bp); antisense primer, 5′-CTC CGC ATT GAC ACA GCG-3′ (exon 2; 268–285 bp); and probe, 5′-CCT CTG TGA TCT GGG AAG CTC AGT GCC-3′ (exon 2; 233–259 bp). β-Actin served as a control and was quantified using the following primers: sense, 5′-AGA GGG AAA TCG TGC GTG ACT-3′, reverse, 5′-CAA TAG TGA TGA CCT GGC CGT-3′, and probe, 5′-CAC TGC CGC ATC CTC TTC CTC CC-3′ (31). Fluorogenic probes were FAM labeled at the 5′ end and TAMRA labeled at the 3′ end. Samples run without the reverse transcriptase step served as negative controls and gave no signal.

IL-17 receptor A and C expressions were analyzed by RT-PCR using mCF cDNA and the following primers: IL-17RA (GenBank Accession No. NM_008359, 300 bp amplicon), sense, 5′-CTG CAG CTG AAC ACC AAT-3′, antisense, 5′-ATG CTG TGT GTC CAA GGT C-3′; IL-17RC (GenBank Accession No. NM_134159; 348 bp amplicon), sense, 5′-CTG CCT GAT GGT GAC AAT G-3′, antisense, 5′-CCA GCA CAG TGT TAC CTT G-3′. PCR was conducted using a “hot-start and touch-down” program of 94°C for 2 min, followed by one cycle of denaturing at 94°C for 30 s, annealing at 66°C (one degree down per cycle until 58°C is reached) for 30 s, and extension at 68°C for 45 s, and then 30 cycles of 94°C for 30 s, 57°C for 30 s, and 68°C for 45 s. A final extension was performed at 68°C for 10 min. PCR products were analyzed by electrophoresis on 2% agarose gels and visualized using ethidium bromide staining.

ELISA.

IL-17 secretion was quantified by ELISA (Mouse IL-17 Quantikine ELISA Kit, No. M1700, R&D Systems). To determine whether HG-induced IL-17 expression is regulated at the transcriptional or posttranscriptional levels, serum-deprived mCFs were treated for 1 h with the protein synthesis inhibitor cyclohexamide (CHX; 10 μM in DMSO) or the inhibitor of RNA polymerase II, actinomycin D (ActD; 10 μM in DMSO) before the addition to HG (25 mM for 48 h). CHX (No. C4859) and ActD (No. A1410) were purchased from Sigma-Aldrich.

PI3K lipid kinase assay.

After overnight incubation in RPMI medium containing 0.5% BSA, mCFs were incubated with HG for 1 h. PI3K lipid kinase assays were performed as previously described (8) using p85 immunoprecipitates. Western blot analysis for p85 confirmed equal loading of cleared cell lysates. Phosphoinositide 3-phosphate (PI3P) levels were quantified by autoradiography and densitometry. Resveratrol (trans-3,4′,5-trihydroxystilbene, No. 554325, EMD Biosciences), a grape skin extract, has been recently shown to inhibit the activation of class IA PI3K (4, 17). Therefore, we examined whether resveratrol inhibits HG-mediated PI3K-dependent IL-17 expression. Serum-starved mCFs were treated with resveratrol (100 μM in DMSO for 30 min; No. 554325, EMD Biosciences) before incubation with HG (25 mM) for the indicated time periods.

Western blotting and immunecomplex kinase assays.

Levels of IL-17 protein in concentrated culture supernatants were analyzed by Western blot analysis under reducing conditions using affinity-purified goat anti-mouse IL-17 antibodies (Cat. No. AF-421-NA; R&D Systems). Culture supernatants were concentrated using 10-kDa molecular mass cut-off Millipore Centricon Centrifugal filter units. IL-17 protein levels were also analyzed by Western blot analysis using 30 μg of cleared whole cell homogenates. α-Tubulin served as a loading control. Akt, phospho- (p)Akt, ERK, and pERK levels in whole cell homogenates were also analyzed by Western blot analysis. The immunoreactive bands were detected by chemiluminescence (ECL Plus; GE Healthcare) and quantified by densitometry. Akt (No. 9272), pAkt (No. 9275), ERK1/2 (No. 9102), pERK1/2 (No. 9101S), and α-tubulin (No. 2144) antibodies were purchased from Cell Signaling Technology (Beverly, MA).

Akt kinase activity was analyzed using a commercially available colorimetric assay kit (Cell Signaling Technology). The assay is based on Akt-induced phosphorylation of glycogen synthase kinase-3 (GSK-3).

ERK enzyme activity was analyzed in whole cell homogenates using a colorimetric assay kit (p44/42 MAP Kinase Assay Kit; Cell Signaling Technology) (9, 13). In brief, mCFs were treated with HG for 1 h. The cells were then harvested and lysed in 1× lysis buffer provided by the manufacturer. The protein content in the lysates was determined by the Bradford method, and 200 μg of cleared cell lysate were incubated with 15 μl of immobilized phospho-p44/42 MAPK (Thr202/Tyr204) monoclonal antibody with gentle rocking at 4°C for 12 h. Immunecomplexes were collected by centrifugation at 8,000 g for 30 s, followed by a wash with lysis buffer, and twice with the kinase buffer containing (in mM) 25 Tris (pH 7.5), 5 glycerophosphate, 2 dithiothreitol, 0.1 Na3VO4, and 10 MgCl2. The complex was then incubated with 50 μl of kinase buffer containing 200 μM ATP and 2 μg of E-26-like protein-1 (Elk-1) fusion protein at 30°C for 30 min. The reaction was terminated by adding 12.5 μl of 5× SDS sample buffer. Samples were then loaded onto 10% polyacrylamide gels using a Bio-Rad Mini Protean 3 system. After electrophoresis, proteins were transferred onto polyvinylidene difluoride membrane and probed with phosphospecific anti-Elk-1 (Ser383) antibodies diluted in 2% nonfat dry milk in Tris-buffered saline containing 0.05% Tween 20 (TTBS). After incubation overnight at 4°C, blots were rinsed in TTBS for 30 min and incubated in horseradish peroxidase-conjugated secondary antibodies in 5% milk in TTBS for 1 h. Following a 30-min rinse in TTBS, blots were incubated in chemiluminescent substrate (SuperSignal Pico West; Pierce) supplemented with 5% SuperSignal Femto (Pierce), and exposed to film. Actin served as a loading control. Actin antibodies were purchased from Santa Cruz Biotechnology, and ERK activation was targeted by PD-98059 (10 μM in DMSO for 1 h; EMD Biosciences, San Diego, CA).

Collagen content.

At 70–80% confluency, the complete media was replaced with media containing 0.5% BSA and ascorbic acid at 50 μg/ml. Thereafter, ascorbic acid was added every 12 h. Cells were incubated (10 μg/ml for 1 h) with IL-17R/fraction crystallizable (Fc) chimera (Catalog No. 4481-MR) or anti-IL-17A neutralizing antibodies (Catalog No. AF-421-NA), or treated with either mouse IL-17RA (No. sc-40038, Santa Cruz Biotechnology,) or IL-17RC (Catalog No. SI01075487, Qiagen) siRNA (100 nM for 48 h) before the addition of recombinant mouse IL-17 (rmIL-17; 100 ng/ml for 48 h; R&D Systems). Fc, normal goat IgG, nontargeting siRNA (control siRNA-A, No. sc-37007; consists of a scrambled sequence that will not lead to the specific degradation of any known cellular mRNA) served as controls. Knockdown of IL-17RC was confirmed by Western blot analysis. Collagen content in cell homogenates and culture supernatants was determined with a hydroxyproline-based assay as described by Villarreal et al. (44). In brief, mCFs were scraped, and both whole cell homogenates and culture medium were combined and then boiled for 20 min. The samples were then precipitated at −20°C following the addition of two volumes of absolute ethanol. The samples were then spun at 30 000 g for 30 min. The pellets were then air dried, resuspended in 2 ml 6N HCl, and hydrolyzed overnight at 108°C. After evaporation to dryness, the samples were reconstituted with 2 ml water and adjusted to a pH between 6 and 8. Hydroxyproline content was determined colorimetrically. A standard curve was generated with known concentrations of hydroxyproline (2 to 10 μg/ml). The experiments were performed in triplicate and at least 3 times, and the results were expressed as fold increases against values from untreated mCFs.

Cell death.

After 48 h incubation in DMEM containing 0.5% BSA, mCFs were treated with resveratrol (100 μM in DMSO for 48 h). Cells were harvested and analyzed for cell death by ELISA (The Cell Death Detection ELISAPLUS, Roche Diagnostics, Roche Applied Science, Indianapolis, IN). This is a photometric enzyme immunoassay used for the quantitative in vitro determination of cytoplasmic histone-associated DNA fragments (mono- and oligonucleosomes) in the cytoplasm after induced cell death. The HT29 colon adenocarcinoma cell line (Catalog No. HTB-38, ATCC, Manassas, VA) and the human breast adenocarcinoma cell line MCF-7 (HTB-22, ATCC) were used as positive controls (33, 36). Both cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS, 2 mM l-glutamine, 5 mM d-glucose, 100 U/ml penicillin G, 100 μg/ml streptomycin sulfate, and 0.25 μg/ml amphotericin B. Cultures were maintained at 37°C in a 5% CO2-95% room air atmosphere. Cells were treated with resveratrol (100 μM in DMSO for 48 h) with the addition of fresh resveratrol to culture medium at 24 h. Cells were cultured in a humidified atmosphere with 5% CO2-95% room air. DMSO served as a solvent control.

Cell death was also analyzed by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay (the MTT Cell Proliferation Assay, ATCC) according to the manufacturer's instructions. The assay is based on the reduction of yellow MTT by dehydrogenase activity of metabolically active cells to generate the reducing equivalents NADH and NADPH. The intracellular formazan thus formed is proportional to cell numbers and was quantified spectrophotometrically. mCFs, HT29, and MCF-7 cells were seeded in 96-well flat-bottom plates at 1 × 104 cells per well. Cells were cultured for 12 h, serum-starved for 24 h, and then incubated with resveratrol (100 μM in DMSO) or DMSO for 48 h. Cell viability was assessed using the mitochondrial-dependent reduction of MTT to formazan. MTT was added at a final concentration of 0.5 mg/ml. Cells were incubated for 2 h, the medium was aspirated, and the cells were dissolved in acidic isopropanol (90% isopropanol, 0.5% sodium dodecyl sulfate, and 40 mmol/l HCl). Optical density was read on a microplate reader (SpectraMAX 190, Molecular Devices) at 570 nm using isopropanol as blank. Cell numbers in DMSO-treated cells at 48 h were considered as 1, and the results expressed as fold induction from DMSO-treated controls.

Statistical analysis.

Results are expressed as means ± SE. For statistical analysis, we used ANOVA followed by an appropriate post hoc multiple comparison test (Tukey method). Data were considered statistically significant at P < 0.05.

RESULTS

HG stimulates IL-17 secretion by primary mCFs.

IL-17, a proinflammatory cytokine, has been implicated in the initiation and progression of a variety of inflammatory and autoimmune diseases (21, 26, 27). Although it is known to be secreted by a novel subset of activated CD4+ T cells called Th17 (20, 32), it is unclear to what extent IL-17 can also be expressed by nonimmune cells. Using multiplexed immunoassays, we previously reported that under basal conditions, primary human cardiac fibroblasts secrete various proinflammatory cytokines including low levels of IL-17 (4.1 pg/ml; 11). Since these cardiac fibroblast cultures were >99% pure, we hypothesized that fibroblasts may express IL-17 and potentially signal via autocrine and paracrine mechanisms. Recently, LaFramboise et al. (28) have reported low basal levels of IL-17 in rat cardiac fibroblast-conditioned medium, further supporting our hypothesis. Since diabetes is an independent risk factor in myocardial fibrosis, we investigated whether HG induces IL-17 expression in mCFs.

Our results show that under conditions of LG (5 mM d-glucose + 20 mM mannitol), the mCFs secreted low levels of IL-17 into the medium (Fig. 1A). Incubation with HG (25 mM d-glucose), however, significantly elevated IL-17 secretion by approximately sevenfold (P < 0.001; n = 12; Fig. 1A). The results obtained by ELISA were confirmed by Western blot analysis using concentrated culture supernatants (Fig. 1B, top) and cleared whole cell homogenates (Fig. 1A, bottom; α-tubulin served as a loading control). rmIL-17 served as a positive control. Our results also indicated that the induction of IL-17 secretion was dose dependent over the range of d-glucose testing, with significant increases seen with as little as 10 mM d-glucose (Fig. 1C). In contrast to the response to d-glucose, no significant changes in IL-17 secretion were observed when its enantiomer l-glucose was used at similar concentrations (data not shown), demonstrating that the stimulatory effect was specific to d-glucose.

Fig. 1.

High glucose (HG) stimulates IL-17 secretion by primary mouse cardiac fibroblasts (mCFs). A: serum-deprived mCFs were incubated with 5 mM d-glucose plus 20 mM mannitol [low glucose (LG)] or 25 mM d-glucose (HG) for 48 h and IL-17 levels in culture supernatants quantified by ELISA. *P < 0.001 vs. LG control, n = 12 experiments. B: Western blot analysis: HG increases IL-17 protein levels in concentrated culture supernatants (top) and whole cell homogenates (α-tubulin served as a loading control; bottom). Numbers at the top of each panel represent an individual sample. These experiments were repeated at least 3 times with similar results. The “+ve” control was recombinant mouse (rm)IL-17 (5 ng) and was used to verify specificity of IL-17 antibodies. C: HG stimulates IL-17 secretion in a dose-dependent manner. Serum-deprived mCFs were incubated with increasing concentrations of d-glucose for 48 h. IL-17 levels in culture supernatants were analyzed by ELISA. Mannitol was used as an osmotic control. *P < 0.05 vs. LG; **P < 0.01 vs. LG; n = 12 experiments. D: HG-mediated IL-17 expression is regulated at transcriptional and posttranscriptional levels. Serum-deprived mCFs were treated with actinomycin D (ActD) or cyclotexamine (CHX) before the addition of HG. IL-17 secretion at 48 h was quantified by ELISA. *P < 0.001 vs. LG; †P < 0.01 vs. HG alone; n = 12 experiments. E and F: HG induces IL-17 mRNA expression. Serum-deprived mCFs were treated with HG for 12 h. IL-17 transcripts were analyzed by Northern blot analysis (n = 3 experiments/group; E) and confirmed by RT-quantitative (q)PCR (F). 28S rRNA (Northern blot analysis) and β-actin (RT-qPCR) served as controls. β-Actin levels in F were not significantly different between samples as evidenced by the β-actin amplicon in gel electrophoresis (F, top). *P < 0.001 vs. LG; n = 6 experiments by ANOVA.

To determine whether HG-induced IL-17 secretion was regulated by transcriptional or posttranscriptional mechanisms, mCFs were incubated with ActD, CHX, or with their solvent vehicle DMSO before treatment with HG. Whereas treatment with ActD or CHX significantly attenuated HG-induced IL-17 release from the mCFs (Fig. 1D), the low basal levels of IL-17 secreted in the presence of LG were not affected (data not shown). Thus the response to HG required both RNA and protein synthesis.

To quantify the amount of IL-17 mRNA induction, mCFs were incubated with HG for 12 h. Total RNA was extracted and analyzed for IL-17 mRNA by Northern blot abalysis. Fig. 1E shows low to undetectable levels of IL-17 mRNA when mCFs were incubated with LG. However, an increase in IL-17 mRNA was observed following incubation with HG. 28S rRNA was used as an internal control and shows similar levels of RNA loading. The effects of HG on IL-17 mRNA expression were also confirmed by RT-qPCR. As shown in Fig. 1E, right, incubation with HG for 12 h resulted in a significant increase in the IL-17 transcript. β-Actin served as an internal control (shown as an insert). Together, these results indicate that 1) mCFs express low levels of IL-17 mRNA and protein at basal conditions, 2) HG is a potent inducer of IL-17 expression, and 3) both enhanced transcription and translation play a role in HG-mediated IL-17 induction (Fig. 1).

HG induces PI3K and Akt activation.

We have demonstrated that HG stimulates IL-17 expression by mCFs (Fig. 1). We next investigated the signaling mechanisms involved in HG-induced IL-17 expression. Since HG activates diverse cellular second messengers including PI3K (41), we analyzed PI3K activation in mCFs following incubation with HG. Activation of PI3K was determined as described previously using PI3K lipid kinase assays (8). HG treatment induced a significant fourfold increase in PI3K-dependent PI3P formation in mCFs (Fig. 2A, arrow; densitometric analysis from 3 independent experiments is shown in Fig. 2B). Equal loading of p85 immunoprecipitates was confirmed by Western blot analysis (Fig. 2A, bottom). Because the serine/threonine kinase Akt/protein kinase B is one of the major downstream targets of PI3K (1, 6), we next analyzed total Akt and pAkt (Ser473) levels in whole cell homogenates using Western blot analysis. Treatment with HG rapidly induced Akt phosphorylation in a time-dependent manner without modulating total Akt levels (Fig. 2C). pAkt levels were quantified by densitometry, and results from three independent experiments are summarized in Fig. 2D. Furthermore, HG induced Akt kinase activity, as seen by the increased levels of phosphorylated glycogen synthase kinase-3 (Fig. 2E). These results indicate that HG potently induces PI3K and Akt activation in mCFs (Fig. 2).

Fig. 2.

HG induces phosphoinositide 3-kinase (PI3K) and Akt activation in primary mCFs. A: HG induces PI3K activity. Serum-deprived mCFs were incubated with LG or HG for 1 h. Cleared cell lysates were immunoprecipitated with anti-p85 antibodies, followed by the PI3K assay as described in materials and methods. WB, Western blotting. Phosphoinositide 3-phosphate (PI3P) was separated by thin-layer chromatography. PI3P levels were quantified by densitometry, and results from 3 independent experiments are summarized (B). *P < 0.001 vs. LG. C: HG activates Akt. Serum-deprived mCFs were incubated with HG as in A. Cleared cell lysates were immunoblotted with phospho- (p)Akt or Akt antibodies. D: ratio of intensity of pAkt to that of corresponding total Akt. *P < 0.05, P < 0.001 vs. control by ANOVA; n = 3 experiments. E: HG stimulates Akt kinase activity. mCFs treated as in C (30 min) were analyzed for Akt kinase activity using immunecomplex kinase assays. Glycogen synthase kinase-3 (GSK3) served as a substrate. These experiments were performed at least 3 times using n = 3 experiments/group.

HG induces Akt activation via PI3K.

Since HG stimulated both PI3K and Akt in mCFs (Fig. 2), we next determined whether HG-mediated Akt activation is PI3K dependent. Our results show that adenoviral transduction of dnPI3Kp85 significantly attenuated HG-mediated PI3K-dependent PI3P formation in mCFs, whereas the control adenovirus, Ad.GFP, had no effect (Fig. 3A ; densitometric analysis from 3 independent experiments is shown in Fig. 3A, bottom). Our results also show that adenoviral transduction of dnAkt1 inhibited HG-mediated Akt phosphorylation (Fig. 3B; densitometric analysis from 3 independent experiments is shown in Fig. 3B, bottom) and kinase activity (Fig. 3C; densitometric analysis from 3 independent experiments is shown in Fig. 3C, bottom). Furthermore, Ad.dnPI3Kp85 inhibited HG-induced Akt kinase activity (Fig. 3D; densitometric analysis from 3 independent experiments is shown in Fig. 3D, bottom), indicating that HG induces Akt activation via PI3K (Fig. 3).

Fig. 3.

PI3K inhibition blocks HG-induced Akt activation. A: adenoviral (Ad) transduction of dominant-negative (dn)PI3Kp85 inhibits HG-induced PI3K activity. mCFs infected with Ad.dnPI3Kp85 or Ad.green fluorescent protein (GFP) (control) for 24 h were treated with HG. PI3K activity was determined as described in materials and methods. Origin of samples is not shown. Autoradiographic signals seen in A were semiquantified, and results from 3 independent experiments are presented (bottom). *P < 0.001 vs. LG + Ad.dnPI3Kp85; †P < 0.01 vs. HG. B: Ad.dnAkt1 inhibits HG-induced Akt phosphorylation. mCFs infected with Ad.dnAkt1 or Ad.GFP were treated with HG for 1 h. Total and pAkt levels in cleared cell lysates were analyzed by Western blot analysis. Autoradiographic signals seen in B were semiquantified, and results from 3 independent experiments are presented (bottom). C: mCFs treated as in C were analyzed for Akt kinase activity by immunecomplex kinase assays using GSK3 as substrate. Autoradiographic signals seen in C were semiquantified, and results from 3 independent experiments are presented (bottom). D: Ad.dnPI3Kp85 inhibits HG-induced Akt kinase activity. mCFs treated as in A were analyzed for Akt kinase activity. Autoradiographic signals seen in D were semiquantified, and results from 3 independent experiments are presented (bottom).

HG induces IL-17 expression via PI3K- and Akt-dependent ERK activation.

HG is a potent inducer of various MAPKs, such as ERK1/2, that play critical roles in cell death, cell proliferation, inflammation, and extracellular matrix regulation (47). Therefore, we investigated whether HG induces ERK1/2 activation in mCFs. Treatment with HG induced ERK1/2 activation in mCFs, an effect that was significantly inhibited by 10 μM PD-98059 (Fig. 4A ; densitometric analysis from 3 independent experiments is shown in Fig. 4A, bottom). Similarly, pretreatment with PD-98059, but not its solvent control DMSO, inhibited HG-induced ERK kinase activity (Fig. 4B; densitometric analysis from 3 independent experiments is shown in Fig. 4B, bottom). Furthermore, Ad.dnPI3Kp85 and Ad.dnAkt1 attenuated ERK kinase activity (Fig. 4C; densitometric analysis from 3 independent experiments is shown in Fig. 4C, bottom). Importantly, inhibition of PI3K, Akt, and ERK blocked HG-induced IL-17 mRNA expression as seen by RT-qPCR (Fig. 4D; n = 6/group). Inhibition of HG-mediated IL-17 expression by Ad.dnPI3Kp85 (Fig. 4E), Ad.dnAkt1 (Fig. 4F), and PD-98059 (Fig. 4G) was also confirmed by Northern blot analysis, with respective densitometric analysis from 3 independent experiments shown in Fig. 4, F–H, bottom. Furthermore, inhibition of PI3K, Akt, and ERK1/2 blunted HG-mediated IL-17 secretion (Fig. 4H; n = 6/group). These results demonstrate that HG 1) activates ERK1/2, 2) induces ERK activation via PI3K and Akt, and 3) upregulates IL-17 mRNA expression and protein secretion via PI3K/Akt/ERK-dependent signaling (Fig. 4).

Fig. 4.

HG induces IL-17 expression via PI3K and Akt-dependent ERK activation. A: HG induces ERK activation. Serum-starved mCFs were treated with PD-98059 (PD; 10 μg/ml in DMSO) for 1 h before HG addition (25 mM for 1 h). Cleared cell lysates were analyzed for total and pERK levels by Western blot analysis using activation-specific antibodies. Signals were semiquantified by densitometric analysis, and results from 3 independent experiments are summarized (bottom). *P < 0.01 vs. LG; †P < 0.01 vs. HG or HG + DMSO. B: HG induces ERK kinase activity. Cleared cell lysates (100 μg) were immunoprecipitated with anti-ERK antibody, followed by immunecomplex kinase assay using E-26-like protein-1 (Elk-1) fusion protein as substrate and [γ-32P]ATP. Signals were semiquantified by densitometric analysis, and results from 3 independent experiments are summarized (bottom). *P < 0.01 vs. LG; †P < 0.01 vs. HG or HG + DMSO. C: HG induces ERK kinase activity in PI3K- and Akt-dependent manner. mCF treated as in Fig. 3, C and D, were analyzed for ERK kinase activity. Signals were semiquantified by densitometric analysis, and results from 3 independent experiments are summarized (bottom). *P < 0.01 vs. LG; †P < 0.01 vs. HG or HG + GFP. D: inhibition of PI3K, Akt, and ERK blunts HG-induced IL-17 mRNA expression. Serum-starved mCFs were infected with Ad.dnPI3Kp85 and Ad.dnAkt1 for 24 h and then incubated with HG for 12 h. Ad.GFP served as a control. IL-17 mRNA expression was analyzed by RT-qPCR. *P < 0.001 vs. LG; †P < 0.05 vs. HG, HG + Ad.GFP or HG + DMSO by ANOVA. EG: inhibition of PI3K (E), Akt (F), and ERK (G) inhibits HG-mediated IL-17 mRNA expression. Serum-starved mCFs treated as in D were analyzed for IL-17 mRNA expression by Northern blot analysis. 28S rRNA served as an internal control. Signals were semiquantified by densitometric analysis, and results from 3 independent experiments are summarized (bottom). *P < 0.01 vs. LG; †P < 0.01 vs. HG or HG + Ad.GFP. H: inhibition of PI3K, Akt, and ERK attenuates HG-mediated IL-17 secretion. mCF were treated as in D, and IL-17 levels in culture supernatants at 48 h post-HG treatment were analyzed by ELISA (n = 9 experiments). *P < 0.01 vs. LG; †P < 0.01 vs. HG or HG+Ad.GFP by ANOVA.

IL-17 stimulates net collagen production by mCFs via IL-17RA and IL-17RC.

We have demonstrated that mCFs express IL-17 (Fig. 1). Recently, Toy et al. (43) have demonstrated that IL-17 signals via an IL-17R heterodimer consisting of IL-17RA and IL-17RC. Therefore, we investigated whether mCFs express both the components. RT-PCR revealed that mCFs expressed both IL-17RA and IL-17RC at basal conditions (Fig. 5A), and HG upregulated IL-17RA but not IL-17RC (Fig. 5B). Furthermore, IL-17 increased net collagen production in mCFs, an effect that was significantly attenuated by IL-17R/Fc chimera or IL-17RA neutralizing antibodies (Fig. 5C). Similarly, the knockdown of IL-17RA or IL-7RC significantly attenuated IL-17-mediated net collagen production (Fig. 5C). The knockdown of IL-17RA and IL-17RC was confirmed by Western blot analysis (Fig. 5D; β-actin served as a loading control). Together, these results indicate that 1) mCFs express IL-17RA and IL-1RC, 2) HG stimulates IL-17RA expression, and 3) IL-17 stimulates net collagen production by mCFs in an IL-17RA and IL-17RC-dependent manner and suggest that IL-17 may affect mCF biology via autocrine and paracrine mechanisms (Fig. 5).

Fig. 5.

IL-17 stimulates collagen production. A: mCFs express IL-17 receptors A (IL-17RA) and C (IL-17RC) at basal conditions. DNA-free total RNA extracted from serum-deprived untreated mCFs was analyzed for IL-17RA and IL-17RC expression by RT-PCR. Samples run without the reverse transcriptase step served as a negative control (lane 4). A representative of 3 independent experiments is shown. B: HG induces IL-17RA, but not IL-17RC, gene expression. Serum-starved mCFs were incubated with HG for 12 h. IL-17RA and IL-17RC expressions were analyzed as in A by RT-PCR. β-Actin served as an internal control. A representative of 3 independent experiments is shown. C: IL-17 stimulates collagen production by mCF. Serum-starved mCFs were incubated with IL-17R/fraction crystallizable (Fc) chimera or anti-IL-17A neutralizing antibodies for 1 h before treatment with rmIL-17A (100 ng/ml). Fc and normal goat IgG served as respective controls. D: IL-17RA and IL-17RC expressions were targeted by small-interfering (si)RNA. Nontargeting siRNA served as a control. mCFs were treated with IL-17RA, IL-17RC or nontargeting siRNA (100 nM) for 48 h, followed by incubation with rmIL-17 (100 ng/ml). Collagen content in cell homogenates and culture supernatants was determined with a hydroxyproline-based assay as described in materials and methods. Experiments were performed in triplicate and at least 3 times. *P < 0.001 vs. untreated; †P < 0.05 vs. IL-17; ††P < 0.01 vs. IL-17 by ANOVA.

Resveratrol inhibits HG-induced PI3K-, Akt-, and ERK-dependent IL-17 expression and IL-17-mediated collagen production.

Resveratrol has been shown to exert both vascular and cardioprotective effects (22, 38, 42). Although it targets various second messengers depending on the cell type and stimulus, it has been shown recently to inhibit class IA PI3K and its downstream signaling (4, 17). Since HG induces IL-17 gene transcription through a PI3K-dependent mechanism, we investigated whether resveratrol inhibits this pathway. Pretreatment with resveratrol (100 μm for 30 min) potently inhibited HG-induced PI3K activity (Fig. 6A ; densitometric analysis from 3 independent experiments is shown in Fig. 6A, bottom), Akt kinase activity (Fig. 6B; densitometric analysis from 3 independent experiments is shown in Fig. 6B, bottom), ERK kinase activity (Fig. 6C; densitometric analysis from 3 independent experiments is shown in Fig. 6C, bottom), IL-17 mRNA expression (Fig. 6D; Fig. 6D, inset, shows Northern blot analysis), and IL-17 secretion (Fig. 6E). The vehicle DMSO had no effect. Importantly, treatment with resveratrol attenuated IL-17-mediated net collagen production (Fig. 6F). Together, these results demonstrate that resveratrol inhibits not only HG-mediated PI3K-, Akt-, and ERK1/2-dependent IL-17 expression but also IL-17-dependent collagen production and suggest that resveratrol has therapeutic potential in myocardial remodeling (Fig. 6).

Fig. 6.

Resveratrol inhibits HG-induced PI3K, Akt, ERK-dependent IL-17 mRNA expression. A: serum-starved mCFs were incubated with resveratrol (100 μM in DMSO for 30 min) followed by incubation with HG for 1 h (AC). PI3K lipid kinase assays were performed as Fig. 2A. PI3P levels were quantified by densitometry, and results from 3 independent experiments are summarized (bottom). *P < 0.001 vs. LG; †P < 0.01 vs. IL-17 + resveratrol. Similarly, resveratrol inhibited HG-mediated Akt kinase activity (B), ERK kinase activity (C), and IL-17 mRNA expression (D); IL-17 mRNA expression was also analyzed by Northern blot analysis and is shown (inset). BD: *P < 0.001 vs. LG; †P < 0.01 vs. IL-17 + resveratrol. E: resveratrol inhibits HG-mediated IL-17 secretion. Serum-starved mCFs were treated with resveratrol for 30 min before HG for 48 h. IL-17 levels in culture supernatants were analyzed by ELISA. *P < 0.001 vs. LG; †P < 0.01 vs. IL-17 + resveratrol. F: resveratrol blocks IL-17-mediated collagen production. Serum-starved mCFs were treated with resveratrol for 30 min followed by incubation with IL-17 (10 ng/ml for 48 h). Net collagen production was quantified as described in materials and methods. *P < 0.001 vs. untreated; †P < 0.001 vs. IL-17 + resveratrol by ANOVA.

Resveratrol is not toxic to mCFs.

Since resveratrol exerts either proliferative or cyotoxic effects in a concentration and cell type-dependent manner, we investigated whether resveratrol-mediated inhibition in IL-17 expression is due to its cytotoxic effects on mCFs. The colon adenocarcinoma cell line HT29 and the human breast adenocarcinoma cell line MCF-7 served as positive controls. Results in Fig. 7A show that neither HG nor resveratrol at various doses exerts growth inhibitory or cytotoxic effects in mCFs. In addition, the combination of HG and resveratrol also failed to induce cell death in mCFs. In contrast, significant cell death was detected in HT29 and MCF7 cells treated with resveratrol at the concentrations used. These results were confirmed by the MTT assay (Fig. 7B). Together, these results indicate that resveratrol at the concentrations employed is not cytotoxic to mCFs (Fig. 7).

Fig. 7.

Resveratrol is not cytotoxic to mCFs. A: resveratrol does not induce mCF cell death. Serum-starved mCFs were treated with either resveratrol alone or in combination with HG for 24 h (n = 12 experiments). Cell death was analyzed by ELISA that quantifies mono- and oligonucleosomal-fragmented DNA in the cytoplasmic extracts. HT29 and MCF7 cells served as positive controls. *P < 0.0001 vs. LG + DMSO. These results were confirmed by MTT assay (B). Serum-starved mCF, HT29, and MCF7 cells were treated with resveratrol (100 μM in DMSO) or DMSO alone for 24 h (n = 12 experiments). Cell viability was assessed by MTT assay as described in materials and methods. *P < 0.0001 vs. respective DMSO. C: signaling mechanism involved in HG-mediated IL-17 expression in mCFs and that are targeted by resveratrol.

DISCUSSION

Here we show for the first time that primary mCFs express IL-17 and IL-17 receptors A and C. Furthermore, HG induced IL-17 mRNA expression and protein secretion in a PI3K-, Akt-, and ERK1/2-dependent manner, and this was inhibited by resveratrol. Since diabetes is a causative and a compounding factor in various cardiovascular diseases, our results suggest that HG/IL-17/IL-17R signaling may contribute to myocardial remodeling and fibrosis in diabetic cardiomyopathy and that resveratrol has therapeutic potential in ameliorating these processes.

IL-17 (IL-17A) was the first member of the IL-17 family to be cloned and characterized (20, 32). The IL-17 family contains six ligands (A, B, C, D, E, and F) and five receptors (IL-17RA, B, C, D, and E). These ligands and receptors are unique in their amino acid sequence and share little to no homology with other interleukins (21, 26, 27). IL-17 is secreted mainly by a subset of CD+ T helper cells called Th17. In contrast, IL-17 receptors are expressed on a variety of cell types and therefore are targets of IL-17 signaling. In fact, we have previously reported that human cardiac fibroblasts express IL-17RA (8) and that treatment with IL-17 induces matrix metalloproteinase-1 expression. In that study we also reported the detection of low levels of IL-17 in the culture supernatants (13). Since these human cardiac fibroblast cultures were essentially pure, we hypothesized that IL-17 detected in the culture medium was derived from the fibroblasts. In support of this hypothesis, LaFramboise et al. (28) have also reported the detection of low levels of IL-17 in rat fibroblast-conditioned medium. In the present study we confirmed these reports and extended our study to determine the molecular mechanisms involved. Our results clearly demonstrate that mCFs express low basal levels of IL-17 mRNA and protein and that their expression is significantly augmented by HG. Since they also express IL-17 receptor, it is possible that HG levels enhance the IL-17 signaling pathway in cardiac fibroblasts via autocrine or paracrine mechanisms and stimulate the release of inflammatory cytokines and extracellular matrix proteins.

The functional IL-17 receptor is a heteromeric receptor complex comprised of IL-17RA and IL-17RC (43). Although IL-17RA is critical for IL-17 biological activity, particularly the inflammatory response, IL-17RC signaling promotes cell survival (50). In this study we show that cardiac fibroblasts express both IL-17RA and IL-17RC at the transcript level and that HG stimulates IL-17RA, but not IL-17RC, expression, indicating a positive amplification in IL-17 signaling in hyperglycemic conditions. However, the signaling pathways involved in HG-mediated upregulation in IL-17RA expression are not known. Our results also demonstrate that IL-17 stimulates net collagen deposition in a IL-17RA- and IL-17RC-dependent manner, indicating that IL-17 may play a role in myocardial remodeling.

We also demonstrated that treatment with resveratrol inhibits HG-induced IL-17 expression in mCFs. Since resveratrol has been previously shown to exert proapoptotic effects (33, 36), we examined whether resveratrol blocks HG-mediated IL-17 expression via increased cell death. Our results demonstrate that resveratrol at the concentrations used in the present study, although cytotoxic to HT29 and MCF-7 cells, is not cytotoxic to mCFs, indicating that the inhibitory effects of resveratrol are not due to cell death. Resveratrol blocked HG-induced IL-17 expression via inhibition of PI3K activation. Frojdo et al. (17) have reported that resveratrol is a potent class IA PI3K inhibitor. Resveratrol also inhibited Akt and ERK1/2 activation in mCFs, indicating that the inhibition of PI3K also affected the IL-17-mediated activation of the downstream PI3K-signaling intermediates. Resveratrol has recently been shown to inhibit fibroblast proliferation. Using primary adult rat cardiac fibroblasts, Olson et al. (30) have reported that resveratrol inhibited angiotensin II-induced cardiac fibroblast growth and proliferation. In addition, resveratrol was also reported to inhibit fibroblast differentiation into myofibroblasts by attenuating ERK1/2 activation (30). However, in the latter study, resveratrol failed to affect ANG II-induced Akt and p70S6 kinase activation. In contrast, our results show that the inhibition of HG-induced IL-17 expression by resveratrol clearly involved both the inhibition of ERK1/2 and the inhibition of Akt kinase activity. Whether resveratrol blocked Akt phosphorylation directly, or as a consequence of PI3K inhibition, needs to be investigated. Of note, resveratrol has been shown to inhibit Akt activation in myocardium and nonmyocardial cells and tissues (14, 19, 40, 42, 51). Whether the observed differences in the effects of resveratrol on Akt activity in mouse and rat cardiac fibroblasts are due to differences in species, the agonist employed or the dose of resveratrol is unclear.

In summary, our results indicate that 1) nonimmune cells such as fibroblasts can express significant quantities of IL-17; therefore, IL-17 expression is not exclusive to Th17 cells; 2) in cardiac fibroblasts, the IL-17 pathway has the potential to be activated by both autocrine and paracrine mechanisms; 3) by stimulating production of collagen and other extracellular matrix proteins by cardiac fibroblasts, IL-17 may play a role in myocardial remodeling and fibrosis; and 4) resveratrol has therapeutic potential in blocking IL-17 expression and IL-17-dependent inflammatory signaling.

GRANTS

This work was supported in part by the merit review grant from the Research Service of the Department of Veterans Affairs.

Footnotes

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

REFERENCES

View Abstract