Interleukin-29 uses a type 1 interferon-like program to promote antiviral responses in human hepatocytes

VIRAL HEPATITIS

Interleukin-29 Uses a Type 1 Interferon-Like Program to Promote Antiviral Responses in Human Hepatocytes Sean E. Doyle,1 Heidi Schreckhise,1 Kien Khuu-Duong,1 Katherine Henderson,1 Robert Rosler,1 Harold Storey,1 Lena Yao,1 Hong Liu,1 Fariba Barahmand-pour,1 Pallavur Sivakumar,1 Chung Chan,1 Carl Birks,1 Don Foster,1 Christopher H. Clegg,1 Perdita Wietzke-Braun,2 Sabine Mihm,2 and Kevin M. Klucher1 Interleukin-28A (IL-28A), IL-28B and IL-29 are a family of class II cytokines that stimulate antiviral responses through a heterodimeric receptor that is distinct from the type I interferon (IFN) receptor. To better understand how this newly described family of cytokines regulates the antiviral state, we compared various cellular responses elicited by IL-29 and IFN-␣. Here we show that these cytokines stimulate similar patterns of signal transducer and activator of transcription 1 (STAT-1), -2, -3, and -5 phosphorylation and nearly identical patterns of gene expression when analyzed in two distinct cell types by microarray analysis. Interestingly, the IL-29 receptor is preferentially expressed on primary hepatocytes within normal liver and pegylated forms of IL-29 and IFN-␣ induced equivalent 2ⴕ5ⴕ oligoadenylate synthetase (OAS) and MX1 gene expression in this cell type. Pegylated IL-29 also produced a significant reduction in human hepatitis B and hepatitis C viral load in vitro and reduced the cytopathic effect caused by the fully replicating flavivirus, West Nile virus. In conclusion, IL-29 and IFN-␣ stimulate identical antiviral responses despite their utilization of different receptors. This fact, combined with significant receptor expression in hepatitis virus-infected livers, suggests that IL-29 may have therapeutic value against chronic viral hepatitis in human patients. (HEPATOLOGY 2006;44:896-906.)

T

he superfamily of human Class II cytokines contains interleukin-10 (IL-10), the IL-10 –related interleukins (IL-19, IL-20, IL-22, IL-24, IL-26), the interferons (IFN-␣, -␤ -␴, -␬, -␻, and -␥) and the interferon-like molecules IL-28A, IL-28B, and IL-29 (also referred to as lambda interferons).1 Collectively, these molecules modulate innate and adaptive immune responses to environmental pathogens and protect the host against diseases such as cancer. The best-characterized class II cytokines are the type I interferons, whose expression is tightly regulated by viral infection.2 After Abbreviations: IL, interleukin; IFN, interferon; STAT, signal transducer and activator of transcription; OAS, 2⬘-5⬘-oligoadenylate synthetase; ISG, interferonstimulated gene; HBV, hepatitis B virus; HCV, hepatitis C virus; HPRT, hypoxanthine-guanine phosphoribosyl transferase; RT-PCR, reverse transcriptionpolymerase chain reaction. From 1ZymoGenetics, Inc., Seattle, WA; and the 2Division of Gastroenterology and Endocrinology, Department of Internal Medicine, Georg-August-Universitat, Gottingen, Germany. Received December 30, 2005; accepted June 20, 2006. Address reprint requests to: Kevin M. Klucher, ZymoGenetics, Inc., 1201 Eastlake Avenue E., Seattle, WA 98102. E-mail: [email protected]; fax: 206-442-6608. Copyright © 2006 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/hep.21312 Potential conflict of interest: Some of the authors are employees of ZynoGenetics, Inc., and as such receive salaries and shares of ZynoGenetics stock. 896

binding, these proteins induce a large set of interferonstimulated genes (ISGs) that inhibit viral replication and activate numerous downstream cellular responses involving dendritic cells, lymphocytes, and macrophages.3 In addition to the type I interferons, viral infection also stimulates the rapid production of IL-28 and IL-29, a related, but distinct subset of the class II cytokine superfamily.4,5 These proteins also possess potent antiviral activity; however, in contrast to the type I interferons, they bind a heterodimeric receptor consisting of the IL-28R␣4,5 subunit and the IL-10R␤ subunit, a receptor subunit that is also shared by IL-10, IL-22, and IL-26.6 Chronic viral hepatitis is the leading cause of liver disease and may play a role in the pathogenesis of lesions characteristic of cirrhosis, hepatocellular carcinoma, and end-stage liver failure. The two major causes of chronic viral hepatitis are hepatitis B virus (HBV), a DNA-containing member of the Hepadnaviridae family that infects approximately 350 million people worldwide,7 and hepatitis C virus (HCV), an RNA virus of the Flaviviridae family that infects approximately 170 million individuals worldwide.8 IFN-␣ is an approved treatment for both types of chronic viral hepatitis and has demonstrated considerable clinical success.9-11 However, this cytokine is ineffective for a substantial percentage of infected individ-

HEPATOLOGY, Vol. 44, No. 4, 2006

uals,12-14 and adverse side effects including fatigue, fever, anorexia, depression, and myelosuppression can be a serious issue for patients receiving treatment.15 Given that IL-29 has many of the same antiviral characteristics as IFN-␣ yet binds a different receptor with a distinct expression pattern, its use as an alternative therapy for chronic viral hepatitis patients needs to be examined. Using biochemical and DNA microarray analyses, we show here that IL-29 and IFN-␣ induce direct antiviral activity in human hepatocytes using a conserved signal transduction pathway. In addition, we find that a pegylated form of IL-29 can inhibit the in vitro replication of a broad range of viruses, including human HBV, and viruses of the Flaviviridae family, including human HCV. This, combined with biological activity in primary human hepatocytes and significant receptor expression in virus-infected livers, strongly suggests that IL-29 may be therapeutically useful for the treatment of chronic viral hepatitis in humans.

Materials and Methods Cells and Tissue. Cell lines were obtained from ATCC (Manassas, VA). Primary cells were purchased from Cambrex (East Rutherford, NJ). Peripheral blood leukocytes were obtained from in-house donors (ZymoGenetics, Seattle, WA). Liver biopsy specimens were obtained from consecutive outpatients who were under care and control within the Liver Unit of the Department of Gastroenterology, Georg-August-Universita¨t Go¨ttingen, Germany, as part of a routine clinical evaluation.16 Liver tissue from patients with a suspected but later excluded liver disorder was regarded as healthy liver tissue. Chronic HCV infection was diagnosed by the presence of antiHCV antibodies and HCV-specific RNA within the serum, chronic HBV infection by the presence of anti HBc antibodies of the IgG isotype, HBs antigen, and HBVspecific DNA within sera. Ongoing liver disease was confirmed in all cases histopathologically according to established criteria.17 Patients with concomitant non-B or non-C viral infections and those with continued alcohol or drug abuse were excluded. None of these patients was previously treated with an antiviral therapy. Informed consent was obtained from each patient. The study was approved by the local ethical committee and conformed to the ethical guidelines of the 2000 Declaration of Helsinki. Protein Preparation. A human IL-29 expression construct was transfected into CHO cells. Media from selected cells was isolated and used to generate purified IL-29 through use of Poros 50 HS, Poros 50 HQ, Phenyl 650S, and Superdex 75 chromatography. IL-29 was also

DOYLE ET AL.

897

produced by Escherichia coli fermentation using a variant coding sequence in which the cysteine at amino acid 172 was changed to a serine (C172S) to limit intramolecular disulfide bonding. Fermentation generated IL-29C172S as inclusion bodies that were solubilized and then refolded by dilution (1:20 v:v) into 10 mmol/L NaCl, 0.4 mmol/L KCl, 2 mmol/L CaCl2, 2 mmol/L MgCl2, 0.05% (w/v) PEG3350, 50 mmol/L Tris, 0.75 mol/L Arginine, 1 mmol/L dithiothreitol, 0.1 mmol/L cystamine, pH 7.8. The protein was then purified by SP550C and Phenyl 650S chromatography. IL-29C172S was then pegylated with mPEG-propionaldehyde (NOF, Tokyo, Japan). The mono-pegylated IL-29C172S was isolated from the reaction mixture by SP HP chromatography. Peginterferon alpha-2a (Roche, Nutley, NJ) was purchased through the pharmacy. IFN-␣2a was purchased from R&D Systems (Minneapolis, MN). RNA Isolation and Analysis. Total RNA from all cell sources was isolated using an RNeasy Mini Kit as described by the manufacturer (QIAGEN, Valencia, CA). Total cellular RNA from liver tissue samples was prepared by CsCl density-gradient ultracentrifugation using guanidinium isothiocyanate.16 For microarray studies, cells were treated with media, 50 ng/mL IL-29, or 5 ng/mL IFN-␣2a for varying times. After stimulation, RNA was purified as described. Samples were processed by the University of Washington Center for Expression Arrays (Seattle, WA), and fragmented cRNAs were hybridized to Human Genome Focus Arrays (Affymetrix, Santa Clara, CA) and stained according to the manufacturer’s instructions. Raw data were analyzed using GeneSpring 7.0 (Agilent, Palo Alto, CA). Values of less than .01 were transformed to a value of .01. The intensity of each array was normalized to the 50th percentile for all arrays using all values not absent and having a raw value of 50 or greater. Values on a per gene basis were normalized to the median calculated for values with a raw value of 50 or greater on all arrays. Scatter plots were generated using unfiltered data. IL-29 – regulated genes were identified as having a one-way analysis of variance P value of less than or equal to .05, a raw intensity in IL-29 –treated samples of 600 (three times the background) or greater, and a fold change of 2 or greater as compared with the media-treated sample at the corresponding time. For real-time reverse transcription polymerase chain reaction (RT-PCR) assays, 15 to 300 ng RNA was analyzed using a master mix composed of 1X Master Mix, 300 nmol/L each primer and 100 nmol/L probe for each gene [IL-28R␣, IFNAR2, IL-10R␤, OAS, MX1, PRKR, or hypoxanthine-guanine phosphoribosyl transferase (HPRT)], 1X MultiScribe, and RNase Inhibitor Mix

898

DOYLE ET AL.

(Applied Biosystems, Foster City, CA). Plates were run on an ABI Prism HT 7900 sequence detection machine and analyzed using SDS 2.0 software (Applied Biosystems). Analysis of whole liver gene expression was performed essentially as described.16 In brief, 2 ␮g of total cellular RNA was reverse transcribed using random hexamers (6 ␮mol/L) for priming. cDNA corresponding to 16 ng RNA was amplified in an ABI 7000 sequence detection system. In experiments measuring the level of receptor gene expression, the relative level of receptor RNA expression to HPRT RNA expression is expressed as 2⫺ dCt,18 an arithmetic formula comparing the difference in cycle number (Ct) required to reach a given threshold using IL-28R␣, IL-10R␤, and IFNAR2 primers to the cycle number required using HPRT primers. For ISG induction studies, a standard consisting of known amounts of RNA from HepG2 cells induced with IFN-␣ was run in each experiment and used to generate a standard curve. Quantity mean values of each ISG were divided by the HPRT quantity mean values to normalize expression levels between samples. Immunohistochemistry. Liver biopsy specimens and cell pellets were fixed overnight in 10% neutral buffered formalin and embedded in paraffin using standard techniques. Using a previously described IHC process,19 primary antibodies were diluted at concentrations ranging from 400 ng/mL to 2 ␮g/mL and incubated for 60 minutes at room temperature. A species-specific biotinylated antibody followed by a streptavidin:horseradish peroxidase with DAB substrate was used for detection of bound antibody. Specific information about the primary antibodies are: IL-28R␣ utilized an internally generated hybridoma, clone 275.121.10.1, IFN␣/␤RII, rabbit polyclonal antibody (PBL Biomedical Laboratories, Piscataway, NJ), and IL-10R␤, rabbit polyclonal antibody (US Biological, Swampscott, MA). STAT Activation. Cells were serum-starved overnight in media containing 0.5% bovine serum albumin before cytokine treatment for 15 minutes or varying times for kinetic analysis. STAT activation was measured by immunoprecipitating STAT-2 and STAT-5B (BD BioSciences, San Diego, CA) from whole cell lysates, resolving precipitates on a NuPage 4-12% Bis-Tris gel and then blotting with antibodies pY-STAT-2 (Upstate Biotechnology, Lake Placid, NY), STAT-2 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), pY-STAT-5 (BD Biosciences), and STAT-5B (BD BioSciences). Whole cell lysates were resolved as described and then blotted with antibodies STAT-1, pY-STAT-1, STAT-3, and pYSTAT-3 (BD BioSciences). Antibody binding was detected using the ECL Western Blotting Detection Reagents (Amersham Biosciences, Piscataway, NJ) as de-

HEPATOLOGY, October 2006

scribed by the manufacturer. Neutralization experiments were carried out with monoclonal antibodies to IL-10R␤ and IFNAR2 (R&D Systems). Antiviral Assays. Antiviral assays against human HBV were carried out at the Southern Research Institute (Frederick, MD) using HepG2-WT10 cells, a stable cell line producing wild-type HBV adr1 strain virus.20 Assays were performed essentially as described.21 Briefly, cells were treated with complete medium containing varying concentrations of pegylated IL-29 or peginterferon alpha-2a for 6 days. Supernatant was collected, viral DNA was extracted using DNeasy 96-well kits (QIAGEN, Valencia, CA), and a real-time quantitative PCR assay was used to measure HBV DNA levels. Percent reduction in HBV viral genomes in supernatants of cytokine-treated cultures is relative to untreated cultures. Antiviral activity against human HCV was assessed by Dr. Brent Korba (Georgetown University), using the stable HCV RNA-replicating cell line, AVA5, derived by transfection of HuH7 cells with the subgenomic HCV replicon BB7.22 AVA5 cells were treated once daily for 3 days with varying concentrations of pegylated IL-29 or peginterferon alpha-2a. HCV RNA levels were measured 24 hours after the last dose of compound. Intracellular HCV RNA levels were measured using a commercial assay (Versant 3.0, Bayer Diagnostics, Berkeley, CA).23 Antiviral activity against West Nile virus (strain New York) was assessed at the Institute for Antiviral Research at Utah State University (Logan, UT) in an inhibition of cytopathic effect endpoint assay using Vero cells as described.24

Results Hepatitis Virus-Infected Liver Cells Express the Genes Encoding the IL-29 Receptor But IL-28R␣ Is Expressed in a Cell-Type-Specific Fashion. The primary target for infection and spread of HBV and HCV in the liver is the hepatocyte.25,26 To begin asking whether IL-29 can stimulate an antiviral response in this cell type, we measured the expression of the IL-29 receptor subunit genes in primary human hepatocytes and two hepatoma cell lines (HepG2 and HuH7) that are commonly used to study HBV and HCV virus biology.22,27 Total RNA was isolated from these cells and used as a template for a realtime RT-PCR assay employing primers to the IL-28R␣ and IL-10R␤ genes. For comparison, we also determined the expression of the type I interferon receptor gene most similar to IL-28R␣, IFNAR2. As shown in Fig. 1A, both IL-29 receptor subunit genes are expressed in primary hepatoctyes and hepatoma cell lines. Immunohistochem-

HEPATOLOGY, Vol. 44, No. 4, 2006

DOYLE ET AL.

899

Fig. 1. IL-29 receptor expression in human liver and blood cells. (A) Cytokine receptor gene expression was analyzed in RNA from primary human hepatocyte cultures (HuHep, n ⫽ 5), hepatoma cell lines, HepG2 and HuH7, and the non-hepatocyte liver-derived cells, SK-Hep-1, hepatic vein smooth muscle cells (HepSMCV), hepatic artery smooth muscle cells (HepSMCA) and hepatic fibroblasts (HepFIB). Levels of IL-28R␣ (black bars), IL-10R␤ (gray bars), and IFNAR2 (white bars) were measured using a real-time RT-PCR assay to determine the level of receptor gene expression relative to HPRT. The ratio of receptor/HPRT RNA is shown on the y-axis. The cell type analyzed is shown on the x-axis. Where applicable, the mean and standard deviation is shown. The mean expression level of IL-10R␤ in the primary hepatocyte (HuHep) samples is off-scale (4.8 ⫾ 3.3 relative units). (B) Cytokine receptor gene expression was analyzed in RNA from healthy livers of uninfected control individuals (n ⫽ 8) or livers from hepatitis B virus (HBV)- (n ⫽ 11) and hepatitis C virus (HCV)-infected (n ⫽ 43) patients. Primers to IL-28R␣ (black bars) and IL-10R␤ (gray bars) were used to determine the level of receptor gene expression relative to HPRT. The ratio of receptor/HPRT RNA is shown on the y-axis. The liver type analyzed is shown on the x-axis. The mean and standard deviation is shown. (C) Cytokine receptor gene expression was analyzed in RNA derived from peripheral blood leukocytes. CD19⫹ B cells, CD3⫹ T cells, CD14⫹ monocytes and CD34⫹ stem cells were isolated and used as a source for RNA. Primers to IL-28R␣ (black bars), IL-10R␤ (gray bars), and IFNAR2 (white bars) were used to determine the level of receptor gene expression relative to HPRT. The ratio of receptor/HPRT RNA is shown on the y-axis. The cell type analyzed is shown on the x-axis. Where applicable, the mean and standard deviation is shown. For all experiments, receptor expression levels were normalized to an internal housekeeping gene control, HPRT, using the 2⫺dCt method (see Materials and Methods).

istry confirmed expression of IL-28R␣ and IL-10R␤ in hepatoma cell lines at the protein level (data not shown). In contrast to the type I interferon receptor and IL10R␤, which are expressed ubiquitously in mammalian cells, the IL-28R␣ gene appears to have a more restricted expression pattern.4 To help characterize IL-29 receptor expression within liver, we assayed several liver-derived, but non-hepatocyte, cell types, including SK-Hep-1, an adenocarcinoma cell line of endothelial origin derived from the liver,28 primary hepatic vein and artery smooth muscle cells, and hepatic fibroblasts. As shown in Fig. 1A, the IL-28R␣ gene was not expressed or expressed at very low levels in these liver-derived non-hepatocyte cells, whereas IL-10R␤ and IFNAR2 mRNA were readily detected. This, together with the hepatocyte data (above), argues that hepatocytes may be the predominant source of IL-29 receptor gene expression within the liver. To demonstrate that the IL-29 receptor subunit genes are expressed in virus-infected liver cells, we isolated total RNA from patients with chronic hepatitis B, hepatitis C and uninfected controls with healthy livers, and then used this RNA as a template for gene expression analysis as described. As seen in Fig. 1B, the IL-28R␣ and IL-10R␤ genes were expressed in both healthy uninfected livers and virus-infected liver samples, although significant interindividual variability occurred in the expression level be-

tween liver samples, particularly for IL-28R␣. To determine whether virus-infected liver expressed the IL-29 receptor subunits at the protein level, HCV-infected liver was analyzed for receptor protein expression using immunohistochemistry. As shown in Fig. 2, both IL-28R␣ (a) and IL-10R␤ (b) were detected in HCVinfected liver using receptor subunit-specific antibodies. Similar staining was seen for IFNAR2 (c) expression. In addition to hepatocytes, HCV has also been shown to infect B-lymphocytes.29-31 To look at IL-28R␣ and IL-10R␤ expression in this and other blood types, RNA was isolated from CD19⫹ B cells, CD3⫹ T cells, CD14⫹ monocytes, and CD34⫹ stem/progenitor cells and used as a template for real-time RT-PCR. As shown in Fig. 1C, only CD19⫹ B-cells express substantial amounts of IL-28R␣ RNA. T cells, monocytes, and CD34⫹ stem cells express only very low levels of IL28R␣ RNA. In contrast, IL-10R␤ and IFNAR2 RNA are easily detected in all of the blood cells analyzed. Flow cytometry confirmed the expression of IL-10R␤ and IFNAR2 on these same cell types at the protein level (unpublished data). IL-29 and IFN-␣ Share Strikingly Similar Patterns of STAT Phosphorylation and Gene Induction. Although IL-29 and IFN-␣ bind and signal through distinct receptors, they both stimulate antiviral activity and the

900

DOYLE ET AL.

HEPATOLOGY, October 2006

Fig. 3. STAT activation in hepatic cells. Protein extracts from HepG2 (A-D) or HuH7 (E-H) cells treated with media alone (a), 1,000 ng/mL IL-29 (b), 100 ng/mL IL-29 (c), or 100ng/mL IFN-␣ (d) were analyzed for phosphorylated STAT (top panels) or total STAT protein (bottom panels). STAT-1 (A, E), STAT-2 (B, F), STAT-3 (C, G), and STAT-5 (D, H) were analyzed.

Fig. 2. IL-29 receptor protein is expressed on hepatitis C virus (HCV)infected liver. IHC on HCV-infected human liver tissue using an antibody to IL-28R␣ (A), IL-10R␤ (B), IFNAR2 (C), a hepatocyte-specific antigen (D), or a macrophage-specific antigen (E). Specific staining is shown as brown.

transcription of genes such as IRF7, MX1, OAS, and major histocompatibility class I.4,5,32 To address the question of whether these cytokines are functionally identical or whether they share overlapping but distinct activities, we monitored the signaling pathways used by these cytokines as well as the gene expression patterns that they can regulate. As shown in Fig. 3, IL-29 and IFN-␣ stimulated the phosphorylation of STAT-1, -2, and -3 in both HepG2 and HuH7 cells. Similar to IFN-␣, high concentrations of IL-29 also induced STAT-5 phosphorylation in HepG2 cells, although this was not observed in HuH7 cells. Because IFN-␣ treatment of the hepatoma cell lines produced higher levels of phosphorylated STAT proteins than IL-29 treatment in these experiments, we asked whether different kinetics of STAT phosphorylation might explain these quantitative differences. As shown in Fig. 4A, IL-29 –mediated STAT-1 phosphorylation in HuH7 cells increased between 5 and 30 minutes of cytokine treatment, before decreasing by the 60-minute time point. At no time during this course of treatment was IL-29 –mediated STAT-1 phosphorylation equivalent to IFN-␣ treatment. Similar results were seen with STAT-2 and -3 phosphorylation (data not shown). No IL-29 –

HEPATOLOGY, Vol. 44, No. 4, 2006

Fig. 4. Kinetic analysis and receptor neutralization of STAT activation in hepatic cells. (A) Protein extracts from HuH7 cells treated for varying times (shown at right) with media alone (a, e), 1000 ng/mL IL-29 (b, f), 100 ng/mL IL-29 (c, g), or 50 ng/mL interferon alpha (IFN-␣) (d, h) were analyzed for phosphorylated STAT-1 (a-d) or total STAT-1 protein (e-h). (B) Protein extracts from HuH7 cells treated with media alone (a-c), 50 ng/mL IL-29 (d-f), or 50 ng/mL IFN-␣ (g-i) in the absence (a, d, g) or presence of neutralizing antibodies to IL-10R␤ (b, e, h) or IFNAR2 (c, f, i) were analyzed for phosphorylated STAT-1 (top panels) or total STAT-1 protein (bottom panels).

mediated STAT-5 phosphorylation was seen in HuH7 cells at any time analyzed (data not shown). Finally, to show that IL-29 and IFN-␣–mediated STAT phosphorylation was mediated by distinct receptors, STAT-1 phosphorylation was analyzed in the presence or absence of neutralizing antibodies to either IL-10R␤ or IFNAR2. As shown in Fig. 4B, IL-29 –mediated STAT-1 phosphorylation was reduced in the presence of the IL-10R␤ antibody but not the IFNAR2 antibody. Similarly, IFN-␣– mediated STAT-1 phosphorylation was reduced only in the presence of the IFNAR2 antibody. To profile gene expression, HepG2 cells were treated with IL-29 or IFN-␣ for 1, 6, and 24 hours. Although we attempted to use IL-29 and IFN-␣ at concentrations that produce roughly equivalent STAT-1 and STAT-2 phosphorylation, IFN-␣ activity was still higher than IL-29 activity in these experiments (data not shown). Total RNA was then isolated from cells and analyzed using DNA microarray technology. As indicated in Fig. 5, these cytokines induced nearly identical gene expression profiles. Approximately 600 genes were induced at least twofold by both IFN-␣ (Fig. 5A) and IL-29 (Fig. 5B) as compared with the media-treated control cells. The degree to which these genes were induced peaked at 6 hours

DOYLE ET AL.

901

and declined thereafter (Fig. 5D). Interestingly, neither cytokine significantly down-regulated gene expression in these cells. Subsequent analysis using a more stringent filtering criteria identified approximately 35 genes that were highly expressed after IL-29 treatment and whose induction was statistically significant. All of these genes are known ISGs encoding proteins involved in antiviral responses (e.g., OAS genes, MX1 genes, and PRKR),33 regulation of proliferation (e.g., IFITM1, IFITM3),34,35 apoptosis (e.g., TNFSF10),36 and signal transduction (e.g., NMI, STAT1, ISGF3).37-39 IFN-␣ induced these same genes in HepG2 cells but to higher levels, consistent with the higher level of STAT phosphorylation observed in HepG2 cells after 5 ng/mL IFN-␣ treatment as compared with 50 ng/mL IL-29 treatment (data not shown). The ability of IL-29 and IFN-␣ to induce these genes, as well as the kinetics of this induction, was confirmed after measuring the induction of PRKR and MX1 using a realtime RT-PCR assay employing gene-specific primers (Fig. 5E). Because HCV also infects B cells,29-31 we compared the gene expression profile induced by IL-29 and IFN-␣ in the U266 B-cell line (Fig. 6). More than 400 of the same genes were induced at least two-fold 6 hours after exposure to both IL-29 and IFN-␣ (data not shown). Application of stringent filtering criteria to the data identified over 80 genes regulated by IL-29 at the 6-hour time point and approximately 150 genes at the 24-hour time point (Fig. 6A). All of the genes regulated by IL-29 in HepG2 cells were also regulated by IL-29 in U266 cells. These data indicate that although IL-29 regulates an overlapping set of genes in these two cell lines, the kinetics of gene regulation by IL-29 and IFN-␣ are more extended in U266 cells as compared with HepG2 cells. Proximal signaling events do not seem responsible for the kinetic difference in gene regulation, because analysis of the phosphorylation state of STAT proteins after IL-29 treatment did not show any significant difference between U266 and HepG2 cells. For both cell lines, STAT activation peaked after 20 to 30 minutes of stimulation with IL-29 and began to decline by the 1- to 2-hour time point (Fig. 4A, unpublished data). Gene induction and extended kinetics were confirmed for a subset of these genes using real-time RT-PCR (Fig. 6B). In contrast to the lack of negative gene regulation obtained with HepG2 cells, IL-29 and IFN-␣ inhibited the expression of more than 30 genes in U266 cells. Many of these genes, including S100A4 and TFRC, have previously been associated with transformation and metastatic potential.40,41 This, combined with the upregulation of proapoptotic and anti-proliferative genes, such as TRAIL and IFITM3, respectively, suggests that IL-29, like

902

DOYLE ET AL.

HEPATOLOGY, October 2006

Fig. 5. IL-29 induces interferon-stimulated genes in HepG2 cells. HepG2 cells were stimulated in triplicate with media, IL-29 (50 ng/mL) or IFN-␣ (5 ng/mL) for 1, 6, or 24 hours. Total RNA was collected and subjected to a microarray analysis as described in Materials and Methods. (A-C) Logarithmic scatter plots compare all data points at the 6-hour time point. Raw expression values for individual genes (each represented by a dot) are compared for the samples indicated on each axis. Diagonal lines represent a 0- or two-fold change. IFN-␣ (y-axis) compared with media (x-axis) (A), IL-29 (y-axis) compared with media (x-axis) (B), and IL-29 (y-axis) compared with IFN-␣ (x-axis) (C) are shown. (D) Hierarchical clustering of genes identified as upregulated by IL-29 using filtering criteria described in Materials and Methods. Red represents high expression, yellow equals moderate expression, and blue signifies low expression. (E) Induction of antiviral genes MX1 and PRKR was analyzed by real-time relative RT-PCR. Ratio of ISG/HPRT RNA is shown on the y-axis, and time of induction is shown on the x-axis. Response of IL-29 –treated (blue, open square), IFN-␣–treated (red, open square), and media-treated cells (green, filled square) is shown.

HEPATOLOGY, Vol. 44, No. 4, 2006

DOYLE ET AL.

903

Fig. 6. IL-29 mediates antiviral gene induction in the U266 B-cell line. U266 cells were stimulated in triplicate with media, IL-29 (50 ng/ mL) or IFN-␣ (5 ng/mL), total RNA harvested after 1, 6, or 24 hours, and transcriptional regulation assessed via microarray analysis. (A) Hierarchical clustering of IL-29 –regulated genes identified using filtering criteria described in Materials and Methods. Red represents high expression, yellow equals moderate expression, and blue signifies low expression. (B) Induction of antiviral genes MX1 and PRKR in U266 cells was confirmed by real-time relativeRT-PCR. Ratio of ISG/HPRT RNA is shown on the y-axis, and time of induction is shown on the x-axis. Response of IL-29 –treated (blue, filled square), interferon alpha (IFN␣)–treated (red, open square), and media-treated cells (green, filled square) is shown.

IFN-␣ may possess anti-tumor activity.42,43 Collectively, these data convincingly demonstrate that despite their utilization of different receptors, IL-29 and IFN-␣ regulate nearly identical patterns of gene expression within responsive cells. Pegylated IL-29 Induces Antiviral Gene Expression and Represses Hepatitis B and C Replication In Vitro. The close functional relationship that IL-29 and IFN-␣ share with respect to gene regulation and anti-viral activity in hepatocytes strongly suggests that IL-29 could be used to treat human hepatitis infections. To begin testing this concept, we incubated primary human hepatocytes

with a pegylated form of IL-29 for 20 hours, then assayed for the induction of the antiviral ISGs, OAS, and MX1,32 by real-time relative RT-PCR. As shown in Fig. 7, this pegylated form of IL-29 stimulated a 20-fold increase in antiviral gene expression. For comparison purposes, we also treated primary human hepatocytes with the current standard of care for chronic hepatitis C patients, pegylated IFN-␣ (peginterferon alpha-2a44). Although pegylation has been shown to reduce the specific activity of IFN-␣, pharmacological advantages of the pegylated protein provide increased antiviral efficacy in the clinic. As indicated, pegylated IFN-␣ stimulated, respectively, a 20-

Fig. 7. IL-29 activity in primary human hepatocytes. (A-B) Primary human hepatocyte cultures were treated with increasing concentrations of pegylated IL-29 (black square, solid line) or pegylated IFN-␣ (peginterferon alpha-2a, open circle, dashed line) for 20 hours before RNA isolation. Known antiviral genes OAS (A) and MX1 (B) were analyzed using a real-time reverse transcription polymerase chain reaction (RT-PCR) assay (normalized to HPRT). Fold induction relative to untreated cells is shown on the y-axis. Cytokine concentration used is shown on the x-axis.

904

DOYLE ET AL.

HEPATOLOGY, October 2006

Fig. 8. Antiviral activity of IL-29. (A) Pegylated IL-29 (solid bar) and pegylated interferon alpha (IFN-␣; peginterferon alpha-2a, open bar) were tested for antiviral activity against human HBV at varying concentrations. Percent reduction of hepatitis B virus (HBV) viral genomes from the supernatant of the untreated cultures is indicated on the y-axis (% of virus control). The percent reduction and standard deviation are shown (n ⫽ 5). Cytokine concentration used is shown on the x-axis. (B) Pegylated IL-29 (solid bar) and pegylated IFN-␣ (peginterferon alpha-2a, open bar) were tested for antiviral activity against HCV in the HCV replicon model at varying concentrations. Hepatitis C virus (HCV) genomic RNA copy number in the cell lysates is indicated on the y-axis (HCV genomic copies/mL). The mean copy number and standard deviation are shown (n ⫽ 2). Cytokine concentration used is shown on the x-axis. (C) Pegylated IL-29 (black bar) was tested for antiviral activity against West Nile virus at varying concentrations. The percent cytopathic effect (%CPE) is indicated on the y-axis. Cytokine concentration used is shown on the x-axis.

fold induction of the MX1 gene and a 30-fold induction of the OAS gene at 100 ng/mL. These data indicate that a pegylated form of IL-29 is able to induce antiviral gene transcription in primary human hepatocytes to levels comparable to those induced by pegylated IFN-␣. To extend these findings further, we tested whether IL-29 could stimulate an antiviral response against human hepatitis viruses. Accordingly, HepG2 cells, constitutively expressing human HBV, were incubated with increasing concentrations of either pegylated IL-29 or pegylated IFN-␣. Viral load was measured in the culture supernatants 6 days later. As seen in Fig. 8A, both proteins reduced HBV copy number to approximately 50% of that observed in the untreated control cultures. Separately, HuH7 cells, constitutively expressing a subgenomic HCV viral RNA genome, were also treated with pegylated IL-29 or pegylated IFN-␣.22 Similar to results reported previously by Robek et al.,45 pegylated IL-29 was capable of reducing HCV viral load in a dose-dependent manner with maximum suppression equivalent to pegylated IFN-␣ (Fig. 8B). Finally, to determine whether IL-29 could reduce the cytopathic effect of a related flavivirus, Vero cell cultures were treated with pegylated IL-29 before infection with replication-competent West Nile virus. As seen in Fig. 8C, pegylated IL-29 effectively suppressed West Nile virus–induced cytolysis at a concentration of 10 ng/mL. Collectively, these data demonstrate that a pegylated form of IL-29 effectively reduces HBV and HCV replication in human hepatocyte– derived cell lines in a manner similar to that of pegylated IFN-␣ and that in addition to inhibiting flavivirus replication, IL-29 can prevent the cytopathic effects caused by virus infection.

Discussion Type I interferons play a critical role in viral immunity and are a valuable therapeutic against virus-induced hepatitis. However, the adverse effects associated with IFN-␣ treatment are common and can be debilitating, making the identification of alternate therapies important. Like type I interferons, expression of the IL-29 family of cytokines is elevated in the peripheral blood of chronic hepatitis C patients, suggesting a potential physiological role for IL-29 in hepatitis virus immunity.16 The studies presented here show the potential for IL-29 to fulfill a role as a treatment for virus-induced hepatitis. Previously we had shown that IL-29, like IFN-␣, was able to protect a hepatoma cell line from cytopathic effects induced by encephalomyocarditis virus.4 In our current studies, biochemical and molecular analysis of the hepatocyte response to IL-29 indicated that an IFN-␣–like response was initiated on binding with its receptor. Like the type I interferons, IL-29 induced rapid phosphorylation of STAT-1 and STAT-2, although IFN-␣ had a higher specific activity. In addition, both STAT-3, and in the case of HepG2 cells, STAT-5, were phosphorylated on IL-29 stimulation. Whether the lack of STAT-5 phosphorylation in IL-29 –treated HuH7 cells is physiologically relevant is unclear. This does not appear to be specific to hepatocyte-derived cells, because the HepG2 cells do have phosphorylated STAT-5 in response to IL-29 treatment. We and others have also seen IL-29 – mediated STAT-5 phosphorylation in cells overexpressing the IL-29 receptor5 (data not shown). Analysis of gene expression using DNA microarrays indicated that the similarity between the IL-29 and IFN-␣ signal transduction pathway continues through to

HEPATOLOGY, Vol. 44, No. 4, 2006

the target gene level, with both cytokines inducing known ISGs in hepatocyte and B-cell lines. Subtle differences in the kinetics or level of induction of specific gene expression were observed between IL-29 and IFN-␣, although how this may affect cytokine-specific functions is unknown. However, the nearly identical gene induction pattern seen with IL-29 and IFN-␣ suggests that both cytokines should have very similar activity in these cell types, including induction of an antiviral state. Importantly, the ability of IL-29 to elicit antiviral responses in both hepatocytes and B cells implies that IL-29 can effectively target major cellular reservoirs of HCV. The ability to promote an antiviral state in B cells is also important because HCV infection is strongly associated with B cell lymphoproliferative disorders, and induction of antiviral and growth suppressive ISGs may aid in the treatment or prevention of such lymphoproliferative states.46 The similarity in the signal transduction pathway used by IL-29 and IFN-␣ is intriguing, given their use of different receptors. In structure, the IL-29 receptor is quite distinct from the type I interferon receptor. For instance, IL-28R␣ is most closely related at the sequence level with the soluble class II cytokine receptor IL-22R␣2,47 whereas the second IL-29 receptor subunit, IL-10R␤, is also used by IL-10, IL-22, and IL-26. All of these cytokines can stimulate STAT-3 phosphorylation, and IL-22 and IL-26 have been shown to phosphorylate STAT-1 (for review see Donnelly et al.6), whereas only the IL-29 cytokine family appears to phosphorylate STAT-2. The ability to phosphorylate STAT-1, -2, and -3 is common to the IL-29 cytokine family and the type I interferons, and the similarity in their activity is likely attributable to the ability to form the ISGF3 complex, containing both STAT-1 and STAT-2.32 One of the key differences we have seen between the IL-29 cytokine/receptor family and the type I interferon/receptor family is in expression of their respective receptor subunits. Using a real-time RT-PCR assay to measure the expression of the RNA for the heterodimeric receptor for IL-29, we determined that both chains of the receptor were expressed in uninfected, HBV-, and HCV-infected livers. Similarly, using immunohistochemistry with specific antibodies to the IL-29 receptor subunits, we determined that the IL-29 receptor is expressed in HCV-infected livers at the protein level. When we analyzed individual cell types within the liver for receptor expression, we found that although the IL-10R␤ gene, like IFNAR2, was expressed in all cell types analyzed, the IL-28R␣ gene was expressed only in the hepatocytes and not in the hepatic vein or artery smooth muscle cells, or in the hepatic fibroblast cells analyzed. This more restricted

DOYLE ET AL.

905

expression of the IL-28R␣ RNA was also seen in the peripheral blood, with high expression of IL-28R␣ mRNA being observed only in the B-cell lineage, whereas in T cells, monocytes, and CD34⫹-stem cells, IL-28R␣ transcript was expressed at very low levels. In contrast, both IL-10R␤ and IFNAR2 were robustly expressed in all of these peripheral blood cell types. These data suggest that although IL-29 might have antiviral activity directly in hepatocytes from virusinfected patients, it likely would have a less systemic effect than IFN-␣ given the lack of the IL-29 receptor on a number of non-hepatocyte cell types. To support this, we have shown that, in contrast to IFN-␣, IL-29 does not inhibit myeloid or erythroid colony formation from human bone marrow cultures (unpublished data). Given the nearly identical patterns in signal transduction and gene induction in hepatocytes, it is perhaps not surprising that IL-29, like IFN-␣, has antiviral activity against a broad range of viruses, including human HBV and viruses of the Flaviviridae family, including human HCV. Importantly, the ability of a pegylated form of IL-29 to reduce HBV and HCV viral load to a similar extent as pegylated IFN-␣, the current gold standard therapy for HCV, shows the potential for IL-29 as a therapeutic against chronic viral hepatitis in human patients. Further studies will provide a better understanding of whether subtle differences in gene expression induced by IL-29 relative to IFN-␣ may reduce the adverse side effects and/or increase the efficacy typically seen in IFN-␣ therapy. Likewise, differences in the expression level and tissue distribution of the IL-29 receptor relative to the type I interferon receptor may directly impact overall efficacy or tolerability of therapy.

References 1. Pestka S, Krause CD, Walter MR. Interferons, interferon-like cytokines, and their receptors. Immunol Rev 2004;202:8-32. 2. Vilcek J, .Sen GC. Interferons and other cytokines. In: Fields BN, Knipe DM, Howley PM, eds. Fundamental Virology. 3rd edition. New York: Lippincott-Raven Publishers, 1996:341-365. 3. Biron CA. Role of early cytokines, including alpha and beta interferons (IFN-alpha/beta), in innate and adaptive immune responses to viral infections. Semin Immunol 1998;10:383-390. 4. Sheppard P, Kindsvogel W, Xu W, Henderson K, Schlutsmeyer S, Whitmore TE, et al. IL-28, IL-29 and their class II cytokine receptor IL-28R. Nat Immunol 2003;4:63-68. 5. Kotenko SV, Gallagher G, Baurin VV, Lewis-Antes A, Shen M, Shah NK, et al. IFN-lambdas mediate antiviral protection through a distinct class II cytokine receptor complex. Nat Immunol 2003;4:69-77. 6. Donnelly RP, Sheikh F, Kotenko SV, Dickensheets H. The expanded family of class II cytokines that share the IL-10 receptor-2 (IL-10R2) chain. J Leukoc Biol 2004;76:314-321. 7. Lok AS. Chronic hepatitis B. N Engl J Med 2002;346:1682-1683.

906

DOYLE ET AL.

8. Lauer GM, Walker BD. Hepatitis C virus infection. N Engl J Med 2001; 345:41-52. 9. Cooksley WG. Treatment with interferons (including pegylated interferons) in patients with hepatitis B. Semin Liver Dis 2004;24(Suppl 1):45-53. 10. Perrillo RP, Schiff ER, Davis GL, Bodenheimer HC Jr, Lindsay K, Payne J, et al. A randomized, controlled trial of interferon alfa-2b alone and after prednisone withdrawal for the treatment of chronic hepatitis B. The Hepatitis Interventional Therapy Group. N Engl J Med 1990;323:295-301. 11. Patel K, McHutchison JG. Initial treatment for chronic hepatitis C: current therapies and their optimal dosing and duration. Cleve Clin J Med 2004;71 (Suppl 3):S8-S12. 12. Marcellin P, Lau GK, Bonino F, Farci P, Hadziyannis S, Jin R, et al. Peginterferon alfa-2a alone, lamivudine alone, and the two in combination in patients with HBeAg-negative chronic hepatitis B. N Engl J Med 2004; 351:1206-1217. 13. Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos G, Goncales FL Jr, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002;347:975-982. 14. McHutchison JG, Gordon SC, Schiff ER, Shiffman ML, Lee WM, Rustgi VK, et al. Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. Hepatitis Interventional Therapy Group. N Engl J Med 1998;339:1485-1492. 15. Dusheiko G. Side effects of alpha interferon in chronic hepatitis C. HEPATOLOGY 1997;26(Suppl):112S-121S. 16. Mihm S, Frese M, Meier V, Wietzke-Braun P, Scharf JG, Bartenschlager R, et al. Interferon type I gene expression in chronic hepatitis C. Lab Invest 2004;84:1148-1159. 17. Mihm S, Fayyazi A, Hartmann H, Ramadori G. Analysis of histopathological manifestations of chronic hepatitis C virus infection with respect to virus genotype. HEPATOLOGY 1997;25:735-739. 18. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001;25:402-408. 19. Bilsborough J, Leung DY, Maurer M, Howell M, Boguniewcz M, Yao L, et al. IL-31 is associated with cutaneous lymphocyte antigen-positive skin homing T cells in patients with atopic dermatitis. J Allergy Clin Immunol 2006;117:418-425. 20. Fu L, Cheng YC. Characterization of novel human hepatoma cell lines with stable hepatitis B virus secretion for evaluating new compounds against lamivudine- and penciclovir-resistant virus. Antimicrob Agents Chemother 2000;44:3402-3407. 21. Buckwold VE, Wilson RJ, Nalca A, Beer BB, Voss TG, Turpin JA, et al. Antiviral activity of hop constituents against a series of DNA and RNA viruses. Antiviral Res 2004;61:57-62. 22. Blight KJ, Kolykhalov AA, Rice CM. Efficient initiation of HCV RNA replication in cell culture. Science 2000;290:1972-1974. 23. Ross RS, Viazov S, Sarr S, Hoffmann S, Kramer A, Roggendorf M. Quantitation of hepatitis C virus RNA by third generation branched DNAbased signal amplification assay. J Virol Methods 2002;101:159-168. 24. Morrey JD, Smee DF, Sidwell RW, Tseng C. Identification of active antiviral compounds against a New York isolate of West Nile virus. Antiviral Res 2002;55:107-116. 25. Sansonno D, Iacobelli AR, Cornacchiulo V, Distasi M, Dammacco F. Immunohistochemical detection of hepatitis C virus-related proteins in liver tissue. Clin Exp Rheumatol 1995;13(Suppl 13):S29-S32. 26. Hadziyannis SJ, Lieberman HM, Karvountzis GG, Shafritz DA. Analysis of liver disease, nuclear HBcAg, viral replication, and hepatitis B virus DNA in liver and serum of HBeAg Vs. anti-HBe positive carriers of hepatitis B virus. HEPATOLOGY 1983;3:656-662.

View publication stats

HEPATOLOGY, October 2006

27. Korba BE, Gerin JL. Use of a standardized cell culture assay to assess activities of nucleoside analogs against hepatitis B virus replication. Antiviral Res 1992;19:55-70. 28. Heffelfinger SC, Hawkins HH, Barrish J, Taylor L, Darlington GJ. SK HEP-1: a human cell line of endothelial origin. In Vitro Cell Dev Biol 1992;28A:136-142. 29. Zehender G, Meroni L, De Maddalena C, Varchetta S, Monti G, Galli M. Detection of hepatitis C virus RNA in CD19 peripheral blood mononuclear cells of chronically infected patients. J Infect Dis 1997;176:12091214. 30. Morsica G, Tambussi G, Sitia G, Novati R, Lazzarin A, Lopalco L, et al. Replication of hepatitis C virus in B lymphocytes (CD19⫹). Blood 1999; 94:1138-1139. 31. Vallat L, Benhamou Y, Gutierrez M, Ghillani P, Hercher C, Thibault V, et al. Clonal B cell populations in the blood and liver of patients with chronic hepatitis C virus infection. Arthritis Rheum 2004;50:3668-3678. 32. Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD. How cells respond to interferons. Annu Rev Biochem 1998;67:227-264. 33. Samuel CE. Antiviral actions of interferons. Clin Microbiol Rev 2001;14: 778-809. 34. Deblandre GA, Marinx OP, Evans SS, Majjaj S, Leo O, Caput D, et al. Expression cloning of an interferon-inducible 17-kDa membrane protein implicated in the control of cell growth. J Biol Chem 1995;270:2386023866. 35. Brem R, Oraszlan-Szovik K, Foser S, Bohrmann B, Certa U. Inhibition of proliferation by 1-8U in interferon-alpha-responsive and non-responsive cell lines. Cell Mol Life Sci 2003;60:1235-1248. 36. Fiorucci G, Vannucchi S, Chiantore MV, Percario ZA, Affabris E, Romeo G. TNF-related apoptosis-inducing ligand (TRAIL) as a pro-apoptotic signal transducer with cancer therapeutic potential. Curr Pharm Des 2005; 11:933-944. 37. Bluyssen AR, Durbin JE, Levy DE. ISGF3 gamma p48, a specificity switch for interferon activated transcription factors. Cytokine Growth Factor Rev 1996;7:11-17. 38. Shuai K. Modulation of STAT signaling by STAT-interacting proteins. Oncogene 2000;19:2638-2644. 39. Darnell JE Jr. STATs and gene regulation. Science 1997;277:1630-1635. 40. Helfman DM, Kim EJ, Lukanidin E, Grigorian M. The metastasis associated protein S100A4: role in tumour progression and metastasis. Br J Cancer 2005;92:1955-1958. 41. Kwok JC, Richardson DR. The iron metabolism of neoplastic cells: alterations that facilitate proliferation? Crit Rev Oncol Hematol 2002;42:6578. 42. Emberley ED, Murphy LC, Watson PH. S100 proteins and their influence on pro-survival pathways in cancer. Biochem Cell Biol 2004;82:508-515. 43. Nagata T, Takahashi Y, Ishii Y, Asai S, Nishida Y, Murata A, et al. Transcriptional profiling in hepatoblastomas using high-density oligonucleotide DNA array. Cancer Genet Cytogenet 2003;145:152-160. 44. Zeuzem S, Feinman SV, Rasenack J, Heathcote EJ, Lai MY, Gane E, et al. Peginterferon alfa-2a in patients with chronic hepatitis C. N Engl J Med 2000;343:1666-1672. 45. Robek MD, Boyd BS, Chisari FV. Lambda interferon inhibits hepatitis B and C virus replication. J Virol 2005;79:3851-3854. 46. Weng WK, Levy S. Hepatitis C virus (HCV) and lymphomagenesis. Leuk Lymphoma 2003;44:1113-1120. 47. Xu W, Presnell SR, Parrish-Novak J, Kindsvogel W, Jaspers S, Chen Z, et al. A soluble class II cytokine receptor, IL-22RA2, is a naturally occurring IL-22 antagonist. Proc Natl Acad Sci U S A 2001;98:95119516.