These authors contributed equally to this work.
Interleukin-27 displays interferon-γ–like functions in human hepatoma cells and hepatocytes†
Article first published online: 6 APR 2009
Copyright © 2009 American Association for the Study of Liver Diseases
Volume 50, Issue 2, pages 585–591, August 2009
How to Cite
Bender, H., Wiesinger, M. Y., Nordhoff, C., Schoenherr, C., Haan, C., Ludwig, S., Weiskirchen, R., Kato, N., Heinrich, P. C. and Haan, S. (2009), Interleukin-27 displays interferon-γ–like functions in human hepatoma cells and hepatocytes. Hepatology, 50: 585–591. doi: 10.1002/hep.22988
Potential conflict of interest: Nothing to report.
- Issue published online: 29 JUL 2009
- Article first published online: 6 APR 2009
- Accepted manuscript online: 6 APR 2009 12:00AM EST
- Manuscript Accepted: 16 MAR 2009
- Manuscript Received: 11 DEC 2008
- Deutsche Forschungsgemeinschaft DFG. Grant Number: SFB542-TPB11
Interleukin-27 (IL-27) is a cytokine belonging to the IL-6/IL-12 cytokine family. It is secreted by antigen-presenting cells, strongly acts on T cells, and also stimulates innate immune cells. In most studies, the effects of IL-27 on T cells were investigated; however, not much is known about possible effects of IL-27 on other cell types. IL-27 signals via the common IL-6–type cytokine receptor chain gp130 and the IL-27–specific chain WSX-1. Given the importance of gp130 in regulating liver responses such as the acute phase response or liver regeneration, we investigated whether IL-27 could also have a function in liver cells. We find that IL-27 stimulates hepatoma cells and hepatocytes by inducing a sustained signal transducer and activator of transcription (STAT)1 and STAT3 activation. Whereas the STAT3 mediated responses to IL-27 (γ-fibrinogen and hepcidin induction) are not detectable, we observe an interferon-gamma (IFN-γ)–like STAT1 response leading to the induction of interferon-regulated proteins such as STAT1, STAT2, interferon response factor (IRF)-1, IRF-9, myxovirus resistance A and guanylate binding protein 2. Conclusion: Our study provides evidence for a function of IL-27 in hepatoma cells and hepatocytes and shows that IL-27 responses are not restricted to the classical immune cells. Our results suggest that IL-27 exerts IFN-like functions in liver cells and that it can contribute to the antiviral response in these cells. (HEPATOLOGY 2009.)
Interleukin-27 (IL-27) is a type I cytokine predominantly secreted by activated macrophages and dendritic cells. It can be allocated to the IL-6/IL-12 superfamily of cytokines. As a heterodimeric cytokine composed of the two subunits p28 and Epstein-Barr virus–induced gene 3,1 IL-27 is a member of the IL-12 cytokine family, also encompassing IL-12 and IL-23. Like these cytokines, IL-27 has profound effects on T-cells and acts on innate immune cells.2, 3 Although IL-27 can have proinflammatory effects, most data point at the dominant role of IL-27 being immunosuppressive. Most studies have investigated the effects of IL-27 on CD4+ T-cells, and not much is known about possible effects of IL-27 on other cell types. IL-27 was shown to promote T helper 1 (TH1) responses through the induction of the transcription factors T-bet, up-regulation of IL-12Rβ2, and interferon-gamma (IFN-γ) production and suppression of the TH2 transcription factor GATA3.1, 4 However, IL-27 is also capable of suppressing both TH1 and TH2 responses during infection with a variety of pathogens.5, 6
IL-27 signaling occurs via a receptor complex composed of the signal transducing receptor chains WSX-1 and glycoprotein (gp)130. Whereas WSX-1 is the IL-27–specific receptor chain,7 gp130 is the common receptor subunit of IL-6–type cytokines.8 Thus, IL-27 also belongs to this family. IL-6–type cytokines activate target genes involved in differentiation, survival, apoptosis, and proliferation. They can exert proinflammatory as well as anti-inflammatory properties and are major players in the acute phase response and the immune response of the organism. IL-6 is a major mediator for the acute phase response of the liver as well as in liver regeneration.9, 10
IL-6–type cytokines stimulate tyrosine phosphorylation of signal transducer and activator of transcription (STAT)1 and STAT3, which can form STAT3 and STAT1 homodimers as well as STAT3/STAT1 heterodimers. However, the importance of the detected STAT1 phosphorylation by IL-6–type cytokines remains elusive. For example, IL-6 and oncostatinM (OSM) only seem to induce an IFN-γ–like response in STAT3 knockout cells.11, 12 There are multiple reasons for this inefficient STAT1 response.13 Not only is STAT1 tyrosine phosphorylation after IL-6–type cytokine stimulation very transient,11, 14 but additionally, most of the phosphorylated STAT1 seems to be trapped in STAT1/STAT3 heterodimers.13
We describe a function of IL-27 in hepatoma cells and hepatocytes. We show that IL-27 elicits an efficient STAT1 response and leads to the expression of IFN-γ–regulated genes in these cells.
Materials and Methods
The human hepatoma cell line HepG2 (DSMZ) was maintained in Dulbecco's modified Eagle medium/Nut. MixF-12 medium with Glutamax supplemented with 10% fetal bovine serum, 100 mg/L streptomycin, and 60 mg/L penicillin. The human hepatocyte cell line PH5CH8 was described previously.15
Isolation and Cultivation of Rat Hepatocytes.
Hepatocytes were isolated from adult male Sprague-Dawley rats as described before.16 Details are provided as Supporting Information.
Cell Lysis, Preparation of Nuclear Extracts for Electrophoretic Mobility Shift Assay, Western Blot Analysis, and Antibodies.
All of these procedures were performed as previously described.13 The antibodies used are listed as Supporting Information.
Viral Infections and Plaque Assay.
Fowl plague virus (FPV) was propagated and used as described previously.17 For infection, 7 × 105 HepG2 cells were left untreated or were pretreated with 50 ng/mL IL-27 for 18 hours. Cells were then washed with phosphate-buffered saline followed by incubation with FPV (0.001 multiplicity of infection) diluted in phosphate-buffered saline/BA (phosphate-buffered saline containing 0.2% bovine serum albumin, 1 mM MgCl2, 0.9 mM CaCl2, 100 U/mL penicillin, and 0.1 mg/mL streptomycin) for 30 minutes at 37°C. The inoculum was aspirated, and cells were incubated for 24 hours with infection medium containing 0.2% bovine serum albumin and antibiotics supplemented either with or without 50 ng/mL IL-27. As a positive control for antiviral activity, infections were performed in the presence of 1000 U/mL interferon-alpha (IFN-α) for 24 hours. Plaque assays were performed as described previously.18 Results are given as plaque-forming units per milliliter, and standard deviations are represented as error bars.
The statistical analysis was performed using a Student t test. P < 0.05 was regarded as being statistically significant.
IL-27 Acts on Human Hepatoma Cells and Cultured Human Hepatocytes.
By screening different cell lines for their response to IL-27, we observed that human hepatoma cells are sensitive to IL-27. HepG2 cells were stimulated with increasing amounts of IL-27, and tyrosine phosphorylation of STAT3 (pY705) and STAT1 (pY701) was assessed by western blot analysis. Stimulation of HepG2 cells for 15 minutes with IL-27 leads to a phosphorylation of both STAT3 and STAT1 in a dose-dependent manner (Fig. 1A). As a control, HepG2 cells were also stimulated with IL-6. STAT1 as well as STAT3 tyrosine phosphorylation occurred on treatment of these cells with IL-6 for 15 minutes. The levels of phosphorylated STAT1 and STAT3 were higher in IL-6–stimulated HepG2 cells than in those treated with IL-27. We next investigated the kinetics of STAT phosphorylation on IL-27 stimulation in HepG2 cells and in the human hepatocyte cell line PH5CH8. Both of these cell lines express the IL-27 receptor WSX-1 on their surface (Supporting Fig. 1). As a control, the cells were also stimulated with IL-6, IFN-γ, and IFN-α. IL-27 induces a sustained phosphorylation of STAT1 and STAT3 (Fig. 1B,C; lanes 2-5) in both HepG2 and PH5CH8 cells, whereas IL-6 only leads to a sustained, albeit more pronounced, STAT3 phosphorylation (lanes 10-13). STAT1 activation after IL-6 stimulation was very transient (Supporting Fig. 2). However, a prominent STAT1 phosphorylation was observed on stimulation of the cells with interferons (lanes 6-9 and lanes 14-17), and STAT2 phosphorylation was only observed on treatment with IFN-α. Of note, up-regulation of both STAT1 and STAT2 was observed when the cells were stimulated with IL-27, IFN-γ, or IFN-α, an indication for an efficient STAT1 activation.
IL-27 Leads to a Prolonged STAT1 and STAT3 Activation in Liver Cells.
As we previously reported, the STAT1 phosphorylation observed on treatment of hepatoma cells and primary human macrophages with IL-6–type cytokines such as IL-6 and OSM does not necessarily lead to the formation of active STAT1 homodimers. Most of the phosphorylated STAT1 is rather trapped in STAT1/STAT3 heterodimeric complexes.13 Thus, we performed electrophoretic mobility shift assays to examine whether phosphorylated STAT1 is forming homodimers on treatment of liver cells with IL-27 (Fig. 2). As controls, we used cells stimulated with IL-6, IFN-γ, or IFN-α. On stimulation of HepG2 or PH5CH8 cells with IL-27, the sustained formation of STAT1/STAT1 (lanes 2-5) complexes shows that IL-27 induces a persistent STAT1 activation. Although STAT3 homodimers also can be detected, STAT3 activation is weak if compared with IL-6. Of note, STAT1/STAT1 dimers are hardly observed on stimulation of the cells with IL-6 at these time points, indicating that the prominent but transient STAT1 phosphorylation observed 15 minutes after stimulation (Fig. 1A) is not translated into a STAT1 response. IL-27, however, should be capable of inducing STAT3 as well as STAT1 responses, because both factors are activated in a sustained manner and bind DNA in their homodimeric form. As expected, IFN-γ and IFN-α mainly induced the formation of STAT1 homodimers. Overall, both interferons show a more sustained STAT1 activation in PH5CH8 cells than in HepG2 cells (Figs. 1B,C, 2).
Compared with IL-6, IL-27 Does Not Induce the STAT3-Dependent Genes γ-Fibrinogen and Hepcidin.
To assess whether IL-27 induces STAT3-dependent genes, we investigated the induction of the acute-phase protein genes γ-fibrinogen and hepcidin in both HepG2− and PH5CH8 cells (Supporting Fig. 3). In contrast to IL-6, neither IL-27, IFN-γ, nor IFN-α seem to induce significant levels of these genes.
IL-27 Displays Antiviral Activity.
Because IL-27 leads to a sustained STAT1 activation, we investigated possible antiviral activities of IL-27 by performing infection assays with HepG2 cells using the fowl plague virus FPV. HepG2 cells were pretreated with IL-27 for 18 hours before infection. Cells were then infected with FPV for 24 hours in the presence or absence of IL-27. As a positive control, infection assays were also performed in the presence of IFN-α, a cytokine that is well known to suppress virus replication. Figure 3 shows that treatment of HepG2 cells with IL-27 reduces the amount of progeny viruses after 24 hours of infection. Similar albeit less pronounced effects could be observed with a human influenza virus A7Puerto-Rico/8/34 isolate (data not shown).
IL-27 Mediates IFN-γ–like, STAT1-Dependent Responses in Liver Cells.
To further investigate whether IL-27 induces an efficient STAT1 response, we investigated whether IL-27 would regulate STAT1-dependent gene transcription and thereby mediate interferon-like responses. We performed reporter gene assays in HepG2 and PH5CH8 cells and found IL-27 to prominently induce an interferon response factor (IRF)-1 promotor luciferase construct (Supporting Fig. 4). We then performed western blot analyses in these cells to monitor STAT1-dependent protein expression on treatment of these cells with IL-27, IL-6, IFN-γ, and IFN-α for different times. Figure 4A (HepG2) and 4B (PH5CH8) show that IL-27 up-regulates the STAT1-dependent genes STAT1 (lanes 4, 5), STAT2 (lanes 4, 5), and IRF-1 (lanes 2-5). Up-regulation of these genes also can be observed on treatment of the cells with IFN-γ or IFN-α, although the STAT2 up-regulation is barely detectable in HepG2 cells stimulated with IFN-γ (lane 9). In contrast, IL-6 fails to up-regulate any of the investigated STAT1-dependent genes. We further checked whether IL-27 would induce other interferon-regulated genes such as guanylate binding protein 2 (GBP2) and myxovirus resistance A (MxA), which are regulated by IFN-γ and IFN-α, respectively. Both genes are implicated in the antiviral response after interferon treatment of cells.19 IL-27 up-regulates GBP2 in a similar manner to IFN-γ (lanes 5 and 9), whereas IFN-α does not induce GPB2. In contrast, IFN-α leads to a prominent up-regulation of MxA protein expression (lanes 15-17), whereas IL-27 only shows a weak induction (lane 5). IFN-γ only induces MxA expression in PH5CH8 cells (lane 9). Because MxA is known to be regulated by type I interferons, we investigated whether its IL-27–mediated induction could be attributable to the up-regulation of IFN-α or IFN-β by IL-27 (Supporting Fig. 5). However, stimulation of HepG2 cells with IL-27 in the presence of neutralizing antibodies directed against IFN-α or IFN-β did not affect IL-27–mediated induction of MxA, suggesting a direct induction by IL-27.
Because the IL-27–regulated transcription factor IRF-1 was recently reported to play a central role in the regulation of the antiviral protein RNA helicase retinoic acid–inducible gene-I (RIG-I),20 we investigated whether IL-27 can induce RIG-I expression. We therefore monitored RIG-I induction in HepG2 and PH5CH8 cells by real-time polymerase chain reaction and western blot analysis. We detected a relatively weak twofold to sixfold increase in RIG-I messenger RNA expression (Supporting Fig. 6), whereas RIG-I protein up-regulation was hardly detectable after 24 hours by western blot (Fig. 4A,B; lane 5). In comparison, RIG-I protein was clearly induced on treatment of both cell lines with IFN-α (lanes 15-17).
Because we found both STAT1 and STAT2 to be up-regulated on stimulation with IL-27 (Fig. 4) and phosphorylation of these factors after treatment with IFN-α, we investigated whether prestimulation with IL-27 could enhance subsequent IFN-α–mediated signaling in PH5CH8 cells. We found that pretreatment with IL-27 enhances subsequent STAT1 and STAT2 phosphorylation on IFN-α treatment (Supporting Fig. 7A) and also induces the expression of IRF-9 (Supporting Fig. 7B), which forms the transcription factor complex interferon-stimulated gene factor 3 together with pYSTAT1 and pYSTAT2. However, we did not detect increased expression of the IFN-α–regulated genes RIG-I and MxA (Supporting Fig. 7C).
IL-27 Acts on Primary Rat Hepatocytes.
To verify whether IL-27 also acts on hepatocytes in primary culture, we isolated primary rat hepatocytes and stimulated these cells with IL-27, IFN-γ, or IL-6 for different times. Treatment of these cells with IL-27 induces a sustained phosphorylation of both STAT1 and STAT3 (Fig. 5), showing that primary hepatocytes respond to IL-27. Whereas the IL-27–mediated STAT1 phosphorylation is comparable to the one obtained after treatment with IFN-γ, the STAT3 response is much weaker than the one initiated by IL-6, confirming the results we obtained in the cell lines.
Within the gastrointestinal system, a role for IL-27 was reported in the context of concanavalin-induced hepatitis,21 Crohn disease,22 as well as colon carcinoma.23 However, all of these studies highlight IL-27 functions in infiltrating immune cells such as T cells and natural killer cells. IL-27 also acts on liver cells, namely, human hepatoma cells, cultured human hepatocytes, and primary rat hepatocytes. We find IL-27 to induce a sustained activation of STAT1 and STAT3 in these cells (Figs. 1 and 5).
IL-6–type cytokine signaling is characterized by a sustained STAT3 activation mediated via the different contributing receptor chains gp130, leukemia inhibitory factor receptor, and oncostatinM receptor. This STAT3 activation is of primordial importance for various functions in the liver such as the acute-phase response and liver regeneration. We therefore investigated the potency of the STAT3 activation mediated by IL-27, the new member of the IL-6–type cytokine family. For this, we selected γ-fibrinogen and hepcidin, two type II acute-phase proteins that are synthesized by hepatocytes in response to IL-6. In our experiments, IL-27 (as well as IFN-γ) was not able to lead to a significant induction of these genes (Supporting Fig. 3). These results suggest that IL-27 at most very weakly contributes to the acute-phase response of the liver and that it is in general a quite weak initiator of STAT3 responses in parenchymal liver cells. Because of the weak but sustained STAT3 activation that we observe after IL-27 stimulation (Figs. 1 and 2), it may be of interest to investigate a larger panel of acute-phase proteins to further dissect the potential contribution of IL-27 to the expression of type II acute-phase response genes.
However, the observed activation of STAT1 on stimulation of hepatoma cells and hepatocytes with IL-27 was of special interest because IL-6–type cytokines such as IL-6 and OSM fail to induce an efficient STAT1 response despite the fact that STAT1 phosphorylation is observed.11–13 Thus, the detection of mere STAT1 phosphorylation (as detected after a 15-minute stimulation in Fig. 1A) does not allow drawing conclusions about STAT1 activity. Together with the very transient phosphorylation of STAT1 after stimulation of hepatoma cells with IL-6, the fact that most of the phosphorylated STAT1 is found in STAT1/STAT3 heterodimers contributes to the lack of STAT1-dependent gene induction after treatment of cells with IL-6.13 It also provides an explanation for the fact that IL-6 and OSM induce an interferon-like response in STAT3 knockout cells,11, 12 because the lack of STAT3 prevents the formation of heterodimers and thereby favors STAT1 homodimer formation. Here we show that IL-27 leads to a sustained STAT1 activation characterized by the formation of STAT1 homodimers (Fig. 2). Investigating whether the observed STAT1 activation translates to the induction of STAT1 target genes, we show that IL-27 up-regulates STAT1, STAT2, IRF-1, and IRF-9 protein expression (Fig. 4; Supporting Fig. 7B). This induction is comparable to the up-regulation after stimulation of these cells with IFN-γ or IFN-α, with the exception of an impaired STAT2 up-regulation by IFN-γ in HepG2 cells (Fig. 4A). This shows that IL-27 mounts an efficient STAT1 response and can mediate interferon-like responses in liver cells. This result corroborates previous data obtained in CD4+T cells and macrophages that highlight the importance of STAT1 for distinct biological activities mediated by IL-27.4, 24, 25
It is an interesting thought that the extent of STAT1 and STAT3 activation may be differently regulated as STAT3 responses are mediated via the gp130 receptor chain, whereas STAT1 responses will most likely only efficiently be mediated via the IL-27–specific WSX-1 receptor chain. This may lead to case-sensitive STAT1 or STAT3 responses. For example, it was recently reported that IL-27 activates both STAT1 and STAT3 in early activated T cells, whereas it displays a preferential activation of STAT3 in fully activated CD4+T cells.26 The reasons for this may be manifold and could involve regulatory proteins or may even solely be attributable to different expression levels of STAT1 and STAT3 and thus to different distributions of STAT-dimer species.
The prolonged activation of STAT1 as well as the up-regulation of STAT1-dependent genes led us to investigate whether IL-27 possesses antiviral activity. We performed a plaque assay in hepatoma cells to assess the antiviral potency of IL-27 and show that, similarly to IFN-α, IL-27 reduces the amount of progeny viruses in HepG2 cells (Fig. 3). This result suggests that IL-27 can mediate antiviral effects in the liver.
To further assess the IL-27–mediated regulation of proteins involved in host resistance to pathogens, we then investigated the regulation of RIG-I, MxA, and GBP2 on stimulation of HepG2 and PH5CH8 cells with IL-27. The RNA helicase RIG-I is induced by retinoic acid as well as interferons and constitutes the first line of defense against viral infections by sensing viral double-stranded RNA.27 Because it was recently shown that IRF-1 plays a central role in the regulation of RIG-I expression, we investigated whether IL-27 would induce this sensor for viral double-stranded RNA. Although we detected an increase in RIG-I messenger RNA levels on IL-27 stimulation in both human hepatoma cells and cultured human hepatocytes (Supporting Fig. 6), RIG-I protein was barely detectable. This shows that although IRF-1 expression may be necessary for RIG-I induction,20 its expression alone is not sufficient. One may speculate that additional cellular signals may lead to an up-regulation of RIG-I protein expression by IL-27 and IFN-γ. Investigating the induction of the IFN-γ–induced protein GBP2 and the IFN-α–regulated MxA protein, we found IL-27 to up-regulate both proteins. IL-27 regulated these genes in a manner comparable to that of IFN-γ treatment. Because MxA is a gene regulated by type I interferons, we tested whether its induction after IL-27 treatment could be mediated through the induction of type I interferons. Experiments with neutralizing antibodies against IFN-α and IFN-β1 did not affect MxA nor GPB2 and STAT1 induction, suggesting that the observed regulation by IL-27 does not involve type I interferon production (Supporting Fig. 5).
Furthermore, we tested whether IL-27 could prime cells for a subsequent IFN-α stimulation (Supporting Fig. 7). Most interestingly, we found that prestimulation with IL-27 enhances subsequent IFN-α–mediated STAT1 and STAT2 phosphorylation and also up-regulates IRF-9. However, we did not detect increased expression of the IFN-α–regulated proteins RIG-I and MxA at the different doses of IFN-α tested. Nevertheless, it would be of interest to investigate other genes induced by the transcription factor complex interferon-stimulated gene factor 3 composed of pYSTAT1, pYSTAT2, and IRF-9.
Recent reports suggest that IL-27 can have antiviral activities in peripheral blood mononuclear cells, CD4+T cells, and macrophages and can inhibit human immunodeficiency virus 1 replication.25, 28 It was shown that IL-27 significantly induces interferon-inducible antiviral genes such as myxovirus protein 1, 2′-5′-oligoadenylate synthetase 2 and RNA-dependent protein kinase in macrophages, suggesting that IL-27 inhibits human immunodeficiency virus replication by eliciting an interferon-like response.25 Together with our data, this suggests that IL-27 can elicit a multifaceted antiviral response.
The results presented in this study suggest that IL-27 may be a potential candidate for studies on combination therapies against hepatitis C. The standard care for a chronic hepatitis C infection is a combination therapy of IFN-α plus ribavirin. For a standard treatment, the response rate is approximately 50% for patients with hepatitis C virus genotype 1 and abut 80% for genotypes 2 and 3.29 For the development of future therapies, interests are focusing on combination therapies with different classes of anti–hepatitis C virus drugs such as protease or polymerase inhibitors. Furthermore, novel IFN-based products are being developed.29 Our results that IL-27 acts on hepatocytes and hepatoma cells and displays IFN-like signaling in these cells as well as the antiviral effects of IL-27 observed in immune cells25, 28 indicate that treatment with IL-27 could be used in the therapy of hepatitis C virus infection.
Taken together, we present data showing that IL-27 acts on hepatocytes and hepatoma cells, elicits IFN-γ–like STAT1-mediated responses in these cells, and is able to regulate genes involved in host resistance to pathogens.
- 22Expression of interleukin-12-related cytokine transcripts in inflammatory bowel disease: elevated interleukin-23p19 and interleukin-27p28 in Crohn's disease but not in ulcerative colitis. Inflamm Bowel Dis 2005; 11: 16–23., , , , , , et al.
- 26Two-sided roles of IL-27: induction of Th1 differentiation on naive CD4+ T cells versus suppression of proinflammatory cytokine production including IL-23-induced IL-17 on activated CD4+ T cells partially through STAT3-dependent mechanism. J Immunol 2006; 177: 5377–5385., , , , , , et al.
Additional Supporting Information may be found in the online version of this article.
|HEP_22988_sm_SupFig1.tif||153K||Supplemental Figure 1.|
|HEP_22988_sm_SupFig2.tif||680K||Supplemental Figure 2.|
|HEP_22988_sm_SupFig3.tif||538K||Supplemental Figure 3.|
|HEP_22988_sm_SupFig4.tif||76K||Supplemental Figure 4.|
|HEP_22988_sm_SupFig5.tif||177K||Supplemental Figure 5.|
|HEP_22988_sm_SupFig6.tif||62K||Supplemental Figure 6.|
|HEP_22988_sm_SupFig7.tif||728K||Supplemental Figure 7.|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.