|Home | About | Journals | Submit | Contact Us | Français|
To ascertain whether resveratrol affects the expression of free fatty acids (FFA)-induced profibrogenic genes, death receptors, and/or apoptosis-related molecules in human hepatic stellate cells, using the LX-2 cell line.
Cells were cultured in the presence of FFAs (2:1 oleate : palmitate) and subsequently treated with resveratrol. Gene expression rates were determined by quantitative realtime PCR. The 50% lethal dose (LD50) of resveratrol in the presence of FFAs was assessed with the MTT viability test.
Compared to vehicle controls, incubation of LX-2 cells with 0.5 mM FFAs induced profibrogenic genes (α-SMA × 2.9; TGF-β1 × 1.6; TIMP-1 × 1.4), death receptors (CD95/Fas × 3.8; TNFR-1 × 1.4), and anti-apoptotic molecules (Bcl-2 × 2.3; Mcl-1 × 1.3). Subsequent addition of 15 μM resveratrol (LD50 = 23.2 μM) significantly (P < 0.05) upregulated further these genes (α-SMA × 6.5; TGF-β1 × 1.9; TIMP-1 × 2.2; CD95/Fas × 13.1, TNFR-1 × 2.1; Bcl-2 × 3.6; Mcl-1 × 1.9). Importantly, this effect was only observed in the presence of FFAs.
Resveratrol amplifies the profibrogenic activation of human hepatic LX-2 stellate cells. This finding raises the possibility that in obese patients with elevated FFAs reserveratrol could provoke hepatic fibrogenesis. In-vivo studies are necessary to further validate this conclusion.
In 1979, ST Leger and colleagues found an inverse relationship between coronary heart disease and wine consumption;1 this observation is now known as the “French Paradox”.2 In the years following, further investigations revealing a potential benefit of components found in red wine (rather than in other alcoholic beverages) led to the “red wine hypothesis”.3,4 The beneficial effects of red wine were attributed to certain phenols that had been earlier identified by Frankel et al. as its dominant antioxidants.5,6 Since then, several effects could be associated with polyphenol intake, such ascardioprotection,7 the prevention of atherosclerosis7 and the promotion of NO release from vascular endothelial cells.8
More recently, resveratrol has been implicated as the most important polyphenol responsible for the beneficial effects of moderate red wine consumption. This non-flavonoid polyphenol exerts anti-oxidative,9 anti-neoplastic10 and anti-inflammatory properties.11 It was further shown to have anti-fibrogenic activity by preventing TGF-β induced cardiac myofibroblast differentiation.12 Also, resveratrol significantly reduced mortality, transaminase concentrations, and liver lesions in alcohol-treated mice,13 improved health and survival in obese mice,14 and was also suggested to act beneficially in chronic liver diseases.15 Resveratrol was further found to inhibit the proliferation of activated primary rat hepatic stellate cells,16 as well as of the activated murine hepatic stellate cell line, GRX;17 these findings suggested that resveratrol could therapeutically intervene with liver injury,16 and that polyphenol-rich foods may serve as an adjuvant treatment in chronic liver diseases.17
In contrast – and as prominent features of the non-alcoholic fatty liver diseases (NAFLD) – free fatty acids (FFAs) promote the process of liver fibrogenesis as they induce hepatocyte apoptosis18 and act profibrogenically.19,20 At least in part, these effects can be attributed to the increased expression of genes encoding for profibrogenic proteins, death receptors, and apoptosis-related molecules.18 Cumulatively, it thus seems possible that resveratrol might counteract certain detrimental consequences in liver that result from pathologically increased FFA levels. However, no data is available on the effect of resveratrol on fibrogenic cells in the presence of pathologic FFA concentrations.
In this study, we have explored the effects of reservatrol on the hepatic stellate cell, since this is the primary cell type implicated in fibrogenesis.21,22 Moreover, these cells have many more intrahepatic and systemic roles than previously assumed: they coordinate and integrate hepatic progenitor cell amplification and differentiation; they serve as a highly plastic functional nexus in a complex and tightly regulated sinusoidal milieu; and they function in a variety of metabolic and immunologic roles that contribute to overall homeostasis.23
Given the emerging role of FFAs in fibrosis, and the suggestion that resveratrol may have therapeutic value, we have investigated the combined effects of FFAs and resveratrol on hepatic stellate cells. Specifically, should resveratrol be able to neutralize, or reduce, FFA FFA-mediated detrimental effects in this crucial cell type, its benefit might even reach beyond protection from liver fibrogenesis. To our surprise, however, resveratrol appears to amplify the fibrogenic effects of FFAs.
LX-2 cells were cultured24 at 2.5 × 106/mL at total volumes of 14 mL in 75 cm2 culture flasks in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 100 U/mL penicillin, 0.1 mg/mL streptomycin, and 2 mM L-glutamine (PAA Laboratories, Pasching, Austria). Cell cultures were maintained in a 5% CO2 atmosphere at 37°C in a steam-saturated incubator.
We employed freshly prepared solutions of FFAs, that is 2:1 oleate : palmitate (Sigma–Aldrich, Seelze, Germany) or of trans-resveratrol (i.e. (E)-5-(2-[4-hydroxyphenyl]ethenyl)-1,3-benzendiol) (Sigma-Aldrich) in phosphate-buffered saline without Ca2+ and Mg2+ (PBS) plus 1% fat-free albumin (Fraction V; Roche, Mannheim, Germany). Where indicated, LX-2 cells were cultured for 48 h in the presence of 0.5 mM FFAs and/or treated for an additional 24 h with or without resveratrol at 5, 10 or 15 μM, respectively.
Based on Sagawa et al.,25 we employed the MTT viability assay26 to determine the 50% lethal dose (LD50) of resveratrol in FFA-supplemented LX-2 cells. To this end, reseveratrol was employed at concentrations of 0, 5, 10, 15, 20, 25, and 30 μM under the general culture conditions specified above and in the presence of 0.5 mM FFAs. MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, which is converted into its H2O-unsoluble derivative, 1-(4,5-dimethylthiazol-2-yl)-3,5-diphenylformazan] (Sigma–Aldrich), was dissolved at 5 mg/mL in phosphate buffered saline (PBS) and employed without interim storage. At 24 h or 48 h after onset of culture, cells were centrifuged at 260 g, medium was removed, and 200 μL of MTT solution was added per well. Cells were then incubated for an additional 26 h. After centrifugation at 260 g, removal of the MTT solution, and washing with PBS, MTT solubilization solution (10% Triton X-100 plus 0.01 N HCl in anhydrous isopropanol) was added to each flask. The viability rate of LX-2 cells was determined photometrically at an absorbance/optical density of 570 nm on a SpectraMax 250 microtiter plate reader (Molecular Devices, Sunnyvale, CA, USA).
The mRNA levels of profibrogenic genes (α-SMA; TGF-β1; TIMP-1), of death receptor genes (CD95/Fas; TNFR-1), and of genes encoding for anti-apoptotic molecules (Bcl-2; Mcl-1) were determined by quantitative realtime reverse-transcriptase PCR (qRT-PCR) using hypoxanthine phosphoribosyltransferase 1 (HPRT1) as a housekeeping gene (for primers, see Tab. 1). QRT-PCRs of the respective cDNAs were performed on an iQ thermal iCycler with real-time detection system software 3.0a and Genex software (BioRad, Hercules, CA, USA) in 30-μL reactions containing 15 μL QuantitTect Sybr Green master mix (Qiagen), 5 μL cDNA, 1 μL forward primer, 1 μL reverse primer (at 10 pmol/μL, each) and 8 μL Aq. dest. Amplifications were performed for 15 min at 95°C, followed by 40 cycles of 30 s at 95°C, 30 s at 55°C, and 30 s at 72°C. Melting-curve data were collected from 95°C to 55°C, at −0.5°C steps for 10 s, each. Relative gene expressions were calculated from the threshold cycles in relation to housekeeping gene or to untreated controls, respectively.
After 72 h of incubation with or without mixed FFAs and with or without resveratrol, LX-2 cultures were added 100 ng/mL of the CD95/Fas-stimulating mouse anti-human CD95/Fas-specific monoclonal antibody (clone CH11; Upstate Biochemicals, Lake Placid, NY, USA) for 4 h. CH11-negative controls revealed the unstimulated apoptosis rate in treated versus untreated cells. At the end of culture, cells were incubated with lysis buffer (R&D Systems) to generate cell lysates that were stored at −20°C until the assessment of apoptosis by M30 ELISA. The apoptosis marker, M30, was measured in LX-2 cells using the M30-Apoptosense ELISA kit (Peviva, Bromma, Sweden). The M30 neoepitope is only exposed upon cleavage of cytokeratin-18 (CK18) by activated caspase-3 at aspartate 396 in the process of apoptosis.27 The M30 ELISA was chosen as a result of an initial comparison between the M30 and TUNEL assays, which revealed a superior specificity of M30 data.
Descriptive statistics were performed for all variables; these include means and standard deviations or medians. Among experimental cell culture conditions, differences between FAT concentrations, death receptor expression rates and M30 neoepitope concentrations were evaluated statistically by one-way anova, repeated-measures anova or paired t-test. A P-value of ≤0.05 was considered as statistically significant. Analyses were performed with SPSS, version 15.0.1 (SPSS, Chicago, IL, USA).
Uptake of FFAS by LX-2 hepatic stellate cells after treatment with 0.5 mM of FFAs for 24 h was verified by fluorescence microscopy after Nile Red staining for lipids and blue nuclear counterstaining (Fig. 1). Negative control cells also revealed intracellular lipids – as, for example, obtained from the fetal bovine serum source – but lipid staining was much more intense in cells that had received additional FFAs.
Relative to FFA-free controls, incubation of LX-2 cells with 0.5 mM FFAs resulted in an upregulation of the mRNAs of activation- and fibrosis-related genes (i.e. α-SMA: × 2.9; TGF-β1: × 1.8; and TIMP-1: × 1.4) by this stellate cell line (Fig. 2, left panel). The presence of additional FFAs also upregulated the expression of death receptor genes (i.e. CD95/Fas: × 3.8; and TNFR-1: × 1.1) (Fig. 3, left panel), as well as the expression of anti-apoptotic proteins (Bcl-2: × 2.3; and Mcl-1: × 1.3) (Fig. 3, left panel). Of all mRNAs investigated, CD95/Fas was upregulated most in the presence of 0.5 mM FFAs when compared to the negative controls. This is in line with our results in the HepG2 hepatoblastoma cell line representing hepatocytes.28
Under FFA-supplemented conditions mimicking steatosis in vitro, resveratrol was first employed at concentrations of up to 30 μM for determining the 50% lethal-dose level. In the presence of 0.5 mM FFAs, the LD50 of resveratrol in human LX-2 hepatic stellate cells was thus calculated as 23.2 μM.
Next, resveratrol was employed at concentrations of 0 (negative control), 5, 10, or 15 μM, respectively. The most distinct effects were measured at 15 μM for all resveratrol-treated conditions tested (see below). Importantly, this finding corresponds to results by Lancon et al. and Baur et al. who showed that correlating concentrations proved beneficial in mice when applied in vivo.14,29 We thus selectively depict results obtained with 15 μM of resveratrol. At lower concentrations, the parameters' expression rates were less pronounced than at 15 μM, but they did not differ qualitatively (not shown).
As shown in Figures 2 and and3,3, resveratrol upregulated the expression of key mRNAs associated with activated, fibrogenic stellate cells. However, this effect was less marked than in the presence of FFAs. Specifically, among the activation- and fibrosis-related genes, only α-SMA and TIMP-1 were slightly upregulated (i.e. α-SMA: × 1.7; and TIMP-1: × 1.4) (Fig. 2, central panel); among the death receptor genes, only CD95/Fas was upregulated by × 2.2 (Fig. 3. central panel); and only one of the anti-apoptotic molecules, Bcl-2, was found increased by × 1.6 (Fig. 3, central panel). Therefore, and despite statistical significance in the cases of α-SMA and CD95/Fas (see Figs 2,,3,3, central panels), resveratrol alone induced gene expression much less than FFAs. Nevertheless, and similar to the findings in the presence of FFAs only, resveratrol again had the greatest influence on the expression of CD95/Fas mRNA; this supports the view that CD95/Fas is highly sensitively upregulated by a broad spectrum of chemically diverse compounds.
Importantly, pre-incubation of LX-2 cells with FFAs before adding resveratrol mimics the conditions encountered by hepatic stellate cells in obese patients who consume red wine for pleasure and/or because of its presumptive health benefits on a background of consistently elevated serum FFA levels. Our in-vitro results indicate that such a concurrent presence of FFAs and resveratrol may enhance the degree of activation in hepatic stellate cells and, hence, their pro-fibrotic activity. When human LX-2 stellate cells had already been incubated for 48 h with 0.5 mM of FFAs, further addition of 15 μM resveratrol significantly (P < 0.05) upregulated the expression of all of the genes investigated. Of note, the expression rates of these genes not only increased in comparison to the negative control values, but also relative to the expression rates determined in cells that received FFAs only. Specifically, mRNA expression of activation- and fibrosis-related genes were upregulated (i.e. α-SMA: × 6.5; TGF-β1: × 1.9; and TIMP-1: × 2.2; Fig. 2, right panel), as were those of death receptor genes (i.e. CD95/Fas: × 13.1; and TNFR-1: × 2.1; Fig. 3, right panel), as well as the mRNA expressions of genes encoding for anti-apoptotic molecules (i.e. Bcl-2: × 3.6; and Mcl-1: × 1.9; Fig. 3, right panel).
As depicted in Figure 4, the presence of FFAs or of resveratrol alone, or the combined presence of FFAs and resveratrol significantly reduced (P < 0.05) the hepatic stellate cells' susceptibility to apoptosis. This finding applied both to the spontaneous rates of apoptosis under either of these conditions, as well as to the provoked induction of apoptosis by the CD95/Fas-agonistic monoclonal antibody, CH11. Even more importantly, under both principal conditions – that is unprovoked and antibody-stimulated apoptosis – the degrees of apoptosis determined for LX-2 cells treated with both FFAs and resveratrol were significantly different (P < 0.05) than cells incubated in the presence of FFAs only.
Hepatic stellate cells are of central importance for the induction and propagation of liver fibrosis, and they provide an increasingly recognized plethora of functions for the liver and the entire organism.23 It thus appeared rewarding to investigate the combined effect of resveratrol and pathologic FFA concentrations on a human stellate cell line. Indeed, this study for the first time investigated the effect of resveratrol on human hepatic stellate cells that were adjusted to mimic the state in morbidly obese individuals by providing FFA-enriched conditions.
Our working hypothesis suggested that resveratrol might counteract the undesirable FFA-mediated activation of this crucial cell type. To verify this assumption could have important implications for the (adjuvant) treatment of patients with chronic liver diseases as, for example, had been suggested by Souza et al.17 Yet, our results revealed a different story. Indeed, FFAs activated human LX-2 stellate cells, which is consistent with results by Lu et al. in primary rat hepatic stellate cells.20 However, the subsequent addition of resveratrol further exacerbated the FFAs' detrimental effects. This red-wine component not only amplified the expression of some select genes, but even of all three groups of genes investigated. The strongest amplification rates were found for the death receptor, CD95/Fas, as well as for the profibrogenic protein α-SMA; however, all other transcripts, including the anti-apoptotic mediators Bcl-2 and Mcl-1 were also super-induced in the parallel presence of FFAs and resveratrol. As a result of the induction of anti-apoptotic key factors, the apoptosis rate of LX-2 cells, as measured by an M30 immunoassay, was found decreased in cells treated with FFAs only, but even more so in cells treated with both FFAs and resveratrol. Still, it appears unlikely that slight increases of only Bcl-2 and Mcl-1 may have so efficiently counter-regulated the sharp upregulation of CD95/Fas. Yet this finding strongly suggests that other pro-apoptotic mediators that were not determined here are also upregulated. As the results argue for themselves, further experimental insight will thus likely reveal that, in fact, various anti-apoptotic factors in concert induce in FFA-laden hepatic stellate cells exposed to resveratrol a reduced susceptibility to die from apoptosis.
Interestingly, both measured on day 3 of culture, Kawada et al. had found suppression of α-SMA expression in the presence of 100 μM resveratrol16 while we arrived at exactly the opposite result at 15 μM resveratrol in the presence of FFAs. However, these results are not easily compared. Basically, it is known that stellate cells obtained from different species may differ in distinct properties as, for example, shown by Zhou et al.30 This indicates that the putatively antagonistic effect of resveratrol may, at least in part, be due to the fact that the compared studies employed rat versus human hepatic stellate cells. Still, as the concentrations employed differ by almost one order of magnitude, it were improper to directly compare their effect. However, our study specifically aimed at identifying effects of resveratrol on an FFA-laden human stellate cell line as a correlate of such cells in obese individuals. Our key finding is thus that resveratrol exacerbates the FFAs' profibrogenic effect. Nevertheless, we still subscribe to the conclusion by Kawada and colleagues that resveratrol might interfere with the stellate cells' signal transduction,16 albeit in a dissimilar manner in cells obtained from different species, and obviously depending on the nature of the (pre-)activating stimulus, such as pathologically elevated FFA levels.
A number of studies demonstrated positive effects of resveratrol as well as, likely, other polyphenols on various non-liver diseases;2,3,5–10,12 these findings suggested that the consumption or application of resveratrol – whether in the form of red wine or as an isolated substance – may also act beneficially in obesity, upon ethanol abuse, as well as in liver disease, which was supported by results in mice.11,13 However, our findings in human stellate cells contradict both such assumptions, as well as the results obtained in rodents.
The role of FFAs in chronic liver diseases, and most prominently in NAFLD, has been extensively characterized as, for example, in a lipidomic analysis by Puri et al.31 Insight from such studies revealed that the detrimental activity of FFAs is most likely mediated via their induction of pro-apoptotic features in hepatocytes.18,19 Moreover, Kupffer cells that are activated in liver injury produce chemokines and over-express death ligands (e.g. FasL, TNF-α, and TRAIL) which induce apoptosis in hepatocytes.32,33 When viewed from this perspective, one might reason that the suppression of apoptosis, as was most strikingly achieved in LX-2 cells in the presence of both FFAs and resveratrol, might be advantageous for patients with fibrotic liver diseases. Yet this response appears restricted to hepatocytes18 while the opposite applies to hepatic stellate cells. As shown following partial hepatectomy, death receptors and their ligands can actually promote liver regeneration.34 However, hepatic stellate cells produce collagen in chronic liver injury so as to replace sites of hepatocyte death and liver injury. The activation of these cells, by promoting fibrogenesis, thus is the dominant event in liver fibrosis. In this context, induction of hepatic stellate-cell apoptosis is a desirable therapeutic strategy that promises to indirectly downregulate collagen production by these cells and thus to counteract liver fibrogenesis.35 Thus, the combined action of FFAs and resveratrol may exacerbate fibrogenesis. In addition, under these conditions the expression of all of the genes investigated – with some of which known to actively participate in the progression of liver fibrosis – were upregulated.
Notwithstanding these considerations, the actual in-vivo setting is probably more complex and may require the integration of additional factors, such as, but not limited to, the following:
Based on both this data and considerations, the recommendation of moderate consumption of red wine to obese individuals should be reconsidered until in vivo models can be used to further clarify this issue. The detrimental “resveratrol effect” is only apparent in the presence of FFAs, which could hint towards a direct lipophilic interaction between these molecules, which requires further analysis.
All authors are most grateful to Ms Svenja Sydor, MS, and Mr Martin Schlattjan, MS, for their excellent experimental contributions. This work was supported by the Deutsche Forschungsgemeinschaft CA (grant 267/4-1); the German Competence Network for Viral Hepatitis, funded by the German Ministry of Education and Research (BMBF; grant No. 16,4); and the Wilhelm Laupitz Foundation.