Global Effect of PEG-interferon-alpha and ribavirin on gene expression in PBMC in vitro.

 

Milton W. Taylor1, William M. Grosse1, Corneliu Sanda1 , Joel E. Schaley1, Xiaoning Wu 3, Shih-Chang Chien3, Fred Smith3, Thomas G. Wu3, Matthew Stephens2, Mary W. Ferris1, Jeanette N. McClintick2, Ronald E. Jerome2, Howard J. Edenberg2

 

1Department of Biology, Indiana University, Bloomington, Indiana, USA

2Department of Biochemistry and Molecular Biology and Center for Medical Genomics at Indiana University School of Medicine,

 Indianapolis, Indiana 46202-5122, USA

3Infectious Diseases Department, Roche Molecular Systems, Inc., Alameda, California 94501, USA

Correspondence should be addressed to M.W.T.; e-mail: taylor@indiana.edu

 

 

Running  title: Effect of IFN-alpha on gene expression in PBMC.

 

Abstract.

Utilizing oligonucleotide microarrays, we have examined the expression of 22,000 genes in peripheral blood cells treated with pegylated interferon-alpha2b (PEG-IFN-a) and ribavirin. Treatment with ribavirin had very little effect on gene expression, however treatment with PEG-IFN-a had a dramatic effect, modulating the expression of approximately 1000 genes (at P<0.001). In addition to genes previously reported to be induced by type I or type II interferons, many novel genes were found to be unregulated, including transcription factors such as ATF3, ATF4, properdin, a key regulator of the complement pathway, a homeobox gene (HESX1), and an RNA editing enzyme apobec3. Chemokines CXCL10 and 11 were up regulated whereas CXCL5 was down regulated. Cytokines IL-15 and IL-18 were also significantly induced whereas IL-1a and IL-1b were down regulated.  Most other interleukins were not affected.  The results of the microarrays were confirmed by kinetic real time PCR.  These data indicate that interferon treatment results in the up regulation of genes associated with the stress response, apoptosis, and signaling, while an equal number of genes are down regulated, including those associated with protein synthesis specific cytokines and chemokines and other biosynthetic functions.


INTRODUCTION.

 

The combination of pegylated interferon-alpha (PEG-IFN-a) and ribavirin is the current treatment of choice for chronic infections of hepatitis C. As part of a large study involving hepatitis C patients (http://www.edc.gsph.pitt.edu/virahepc/) we shall perform transcriptional microarrays with RNAs isolated from the peripheral blood monocytes (PBMC) of patients at different time points after treatment to discern possible differences in the gene response to IFN-ribavirin treatment between responders and non-responders. As a pilot study we examined the response of PBMC from healthy controls treated with this combination in vitro.  These data will be used in future to compare the response in vivo to the same treatment.  PBMC are being used in this study because of the lack of human liver tissue.

The mechanism whereby the combination of IFN-a and ribavirin results in complete clearance of the virus in some cases but only partial response in others is unknown.  In previous work1,2 we have examined the induction of serum cytokines in hepatitis C patients undergoing interferon/ribavirin treatment.  The only serum cytokines found to be elevated after treatment were the IL-1 agonist IL-1Ra and the pro-inflammatory cytokine IL-6. None of the more common cytokines such as IL-2, IL-4, IL-1, IFNg, or TNF-a were elevated in the serum of treated hepatitis C patients at any time following treatment, as measured by ELISA.  The induction of IL-1Ra and IL-6 was only transient, and in most cases had returned to normal (before treatment) levels by 24/48 hours.

In order to obtain a more thorough picture of the genes induced by the combined treatment of PEG-IFN-a and ribavirin, we treated peripheral blood mononuclear cells (PBMC) from normal individuals in culture for 24 hours with either PEG-IFN-a alone, ribavirin alone or the two in combination. The profile of genes being induced under these conditions was analyzed using Affymetrix GeneChip® HG-U133A microarrays.  These arrays contain oligonucleotides that report on the expression of approximately 22,000 human genes.  Both PEG-IFN-a and  combination treatment result in the induction or down regulation of a large number of genes (approximately 1000) at a significance level of p ≤ 0.001.

Four hundred and thirty nine unique sequences were induced by 24 hours treatment using a P value < 0.001. A subset of IFN related genes were selected for kinetic RT-PCR analysis.  The confirmatory results of kinetic RT-PCR in general matched the results obtained from the microarrays.  Four hundred and forty six unique sequences were down regulated by 24 hours in the presence of PEG-IFN and ribavirin, including a large number of ribosomal proteins and translation factors. Ribavirin by itself had essentially little effect on gene induction or down regulation. In this paper we report the identification of genes of biological importance, many not previously reported to be affected by interferon. An analysis of the response in human PBMC has not been documented before.

 

Methods.

Interferons.

Peg-intron (PEG-IFN-a2/b) was kindly provided by Schering-Plough (Kenilworth, N.J.) and used at 500 international units/ml when added to cultures of PBMC. Ribavirin (Schering-Plough, Kenilworth N.J.) was used at 10 mg/ml in the treatment of PBMC.

RNA purification

Whole blood (60 ml) was collected from healthy volunteers (4 individuals; one on two different days) into heparin-containing collection tubes.  Aliquots of blood were diluted in equal volumes (8 ml) of PBS and carefully layered over 10 ml Ficoll-Paque Plus (Amersham Pharmacia Biotech, Piscataway, NJ) gradients at room temperature.  The total PBMC were collected in PBS, washed once, and incubated at 37° C in the presence of CO2 for 24 hours in RPMI media (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum at 5x106 cells per ml. Samples were treated with either 500 units/ml PEG-IFN-a, 10 mg/ml ribavirin, a combination of both or incubated without interferon or ribavirin as control. RNA was extracted from 5x107 cells using Tri Reagent (Molecular Research Center, Inc., Cincinnati, OH) according to the manufacturer's specifications,  and finally the RNA was purified through an RNeasy column (Qiagen, Valencia, CA).  The RNA was subjected to several quality control measures, including agarose gel electrophoresis to examine for RNA degradation, a spectrophotometer scan from 200 to 350 nm and capillary electrophoresis using an Agilent Bioanalyzer 2100 instrument.

Array hybridization/detection

RNA Labeling, hybridization and scanning were performed as recommended by Affymetrix in their GeneChip® expression Analysis Technical Manual (Affymetrix, Santa Clara, CA).  Fifteen micrograms of biotinylated cRNA was added to a total hybridization cocktail of 300 µl, and 200 µl was hybridized to each Affymetrix Human Genome HG133A GeneChip®.  Hybridization was at 45°C for 17 hours with constant rotation.  The hybridization mixture was then removed and the GeneChips® washed and doubly stained with phycoerythrin-labeled Streptavidin. GeneChips® were then scanned using the dedicated scanner, controlled by Affymetrix Microarray Suite version 5 (MAS5) software. The average intensity on each array was normalized by global scaling to a target intensity of 1000. Microarray Suite calculates a set of absolute metrics for each transcript: a measure of the expression level ("signal"), a detection call for each transcript: Present (P), Absent (A), or Marginal (M) and the level of confidence with which the sequence was detected.

Database and analysis

The data were exported from MAS5 and entered into a custom-designed database (MicroArray Data Portal) at the Center for Medical Genomics at the Indiana University School of Medicine. Data from genes that are not reliably detected in at least one experimental condition (detected on at least half of the individual arrays in at least one experimental condition) are filtered out before further analysis: this avoids calculations on data that primarily represent “noise” at or near background intensities3. Expression levels (signals) were log-transformed to make the data more normal before conducting t-tests; results of t-tests on the untransformed data are very similar, as are results of non-parametric statistical analyses (Mann-Whitney test).  In addition, hierarchical clustering was applied to data after filtering out genes that did not vary with the different conditions 4 by requiring that the ratio of standard deviation to mean across all of the arrays be equal to or greater than a specified level.

 

Quantification of mRNA expression levels using kinetic RT-PCR analysis

Kinetic RT-PCR was used to monitor expression of a subset of IFN inducible genes in IFN a treated PBMC.  Total RNA extracted from IFN treated PBMC was quantified using the RiboGreenŇ RNA Quantification kit (Molecular Probes, Eugene, OR).  The expression levels of selected IFN inducible genes based on the GeneChip data were determined using SYBRŇ Green I dye-based kinetic RT-PCR 5.  A 10 ul single-step RT-PCR reaction consists of 50 mM Bicine, 125 mM KOAc, pH 8.0, 8% glycerol, 3 mM Mn(OAc)2, 0.2 X SYBRŇ Green I diluted in DMSO, (Molecular Probes), 0.45 mM ROX (Sigma-Aldrich, St. Louis, MO), 200 mM dATP, 200 mM dGTP, 200 mM dCTP, 400 mM dUTP, 200 nM each primer, 0.2 units uracil N-glycosylase (Applied Biosystems, Foster City, CA), 1 unit rTth DNA polymerase (Applied Biosystems), and 2 ng total RNA from PBMC.  The samples were amplified at 50°C 2 min, 95°C 1 min, 60°C 30 min, followed by 95°C 30 sec, 60°C 30 sec for 50 cycles, held at 72°C using Applied Biosystems Prism 7900HT Sequence Detection System.  Each RNA sample was assayed for each target gene in triplicate together with a negative control.  Reactions with serially diluted run-off RNA transcripts (hTERT, human telomerase catalytic component gene) were included in each experiment as an external reference to monitor inter-experimental variations and to determine the “relative copy number” in each RNA sample.  Copy number is defined as a PCR signal generation unit.

The gene expression level in each sample was initially expressed as “relative copy numbers” determined by an external standard curve included in each PCR plate.  Then, the relative copy numbers were normalized by the relative gene expression levels of housekeeping genes.  A total of nine housekeeping genes (b2M, GAPDH, PBGD, PPP1CA, RNase P, U3, U8, U12 and U13) were assayed with other IFN inducible gene targets.  The final housekeeping genes (b2M, PBGD, PPP1CA, RNase P, U13, and U3) were chosen based on their consistence of the expression levels among all samples using geNorm software. (http://allserv.rug.ac.be-/~jvdesomp/genorm/index.html).  Finally, gene expression fold changes of IFN-a induction was determined by dividing the relative copy numbers of the same RNA sample after the treatment by the relative copy numbers of each RNA sample before the treatment.

 

RESULTS

Many genes were induced and down regulated by Peg-IFN-a, but essentially none were affected by ribavirin at the concentration used (Table 1). The number of genes with apparently altered expression in the presence of ribavirin (at 10 mg/ml, either with or without interferon) is less than that expected by random chance at each P value (Table 1). In contrast, PEG-IFN-a changes the expression of a greater number of genes than expected by chance, particularly at the most stringent P values (Table 1). The relative lack of effect of ribavirin is further demonstrated by hierarchical clustering of the data from the arrays, including in the analysis those genes that varied in expression across the experiment (selected as having a ratio of standard deviation to the mean ≥0.5) (Fig. 1). The array data from a given individual with and without ribavirin cluster together more closely than the arrays from another individual with either of those treatments.. The effects of interferon treatment on gene expression are large and consistent, whereas those of ribavirin are essentially undetectable. This allowed us to group the arrays on the basis of interferon treatment alone, giving a 10 x 10 array comparison that is much more powerful than a 5 x 5 comparison (Table 1). This merged analysis shows a very large number of genes whose expression is changed by interferon at very highly significant levels (supplemental data, also available at http://cmg.iupui.edu/pub /Taylor1    (upon publication of this article).

Genes induced or decreased by  IFN .

Because there was no detectable effect of ribavirin, we describe the results of the merged dataset, which represents genes whose expression was altered by interferon. Even using the stringent criterion of a significant difference from controls at the p ≤0.001 level, 541 genes were up-regulated by interferon (Supplementary data set 1). Of these 541 genes, 347 are unique characterized genes (i.e. appear in GeneCards and/or other databases), and 194 are unknown.

Likewise, at the same stringent criteria, 534 genes were down-regulated; of these, 354 were known genes (Supplementary data set 2).  The largest classes of genes induced, as a percentage of the total genes of that class on the array, were involved in defense/immunity proteins, cell death, transcriptional regulation, stress proteins and responses to external stimuli, including chemokines (Table 2). This is in general agreement with other reports on the classes of interferon stimulated genes 6-8, although our data analyze many more genes. The largest class of down-regulated genes are genes associated with metabolism, macromolecular biosynthesis, and transcriptional regulation.  To date there has been no detailed analysis of down regulated genes.

 Kinetic RT-PCR, is an accurate, sensitive and reliable method for quantitating mRNA levels in mixtures of total cellular RNA over a wide range of relative transcript abundance. A subset of the IFN inducible genes were selected for further kinetic RT-PCR analysis9. A comparison of the fold changes in expression as found by microarray analysis and kinetic RT-PCR is presented in Table 3.  There was good agreement between the data obtained from the microarrays and the data generated by kinetic RT-PCR, although some difference was observed in the magnitude of differential expression detected by the two methods. For example, OAS1 (12 fold vs. 24 fold), CD80 (2 fold vs. 10 fold), MX1 (8 fold vs. 38 fold), MX2 (6 fold vs. 24 fold), OAS3 (12 fold vs. 43 fold), and IFIT4 (12 fold vs. 82 fold). Because of its broad dynamic range, it is not surprising that kinetic RT-PCR detected larger fold changes than microarrays. Again, kinetic RT-PCR did not detect any gene expression changes in PBMC that have been treated by ribavirin alone. Overall, kinetic RT-PCR confirmed that the differential gene expression detected by microarrays indeed reflected the cellular mRNA levels in PBMC upon the treatment with PRG-IFN-a and ribavirin.

 

Novel interferon induced genes.

A large number of genes not previously identified as being induced by interferons were detected in this study (Supplementary data, Tables 1 )). Among these genes were ApoBec3a (20 fold), an RNA editing enzyme which has cytidine and deoxycytidylate deaminase activity10. The companion enzyme ADAR (adenosine deaminase) reported to be induced by IFNs 11was also induced (2 fold). Properdin, a key factor in the regulation of the complement alternative pathway12 was highly induced. We have found that this factor is also induced in cell lines by the combination of IFN-con1 and IFN-gamma (Ms in preparation). A number of transcription factors were induced including ATF3 (4.5 fold), an immediate response gene that is induced in cells exposed to various stress stimuli 13 and ATF4 (1.5 fold). Another presumptive transcription factor induced is HESX1 (20 fold); mutations in this gene affect the development of the pituitary gland14 .Induction of HESX1 is not unique to PBMC since we have found the HESX1 induction in a number of cell lines of non-hematopoietic origin (manuscript in preparation). The up-regulation of CD38 (6.7 fold) would indicate that interferon may play a role in the development of B-cells and B-cell maturation 15 . The Ly6E protein (8.4 fold) also induced is a cell surface protein expressed at high levels on activated peripheral T cells16.This complex is also activated by interferon, indicating an important role for interferon in both B cell and T-cell activation.  One of the genes most highly induced (132 fold), was sialoadhesion; this molecule is known to bind sialic acid molecules on the cell surface. Cig5 (Viperin) (18.7 fold), a protein first identified in HCMV infected and interferon treated cells 17 does not appear in earlier reports of interferon induced genes 6-8.

Cytokine and chemokine genes.

Since we initiated this study with a primary interest in the role of cytokines in the interferon response in hepatitis C patients 1,2, we examined cytokine expression levels in some detail. The induction pattern of the more common interleukins, chemokines and their receptors is presented in Table 4.  The only interleukins significantly induced were IL-1ra, IL-18, IL-15 and TNF-a. The receptors for IL-2, IL-15 and IL-12 were also up-regulated.

The most significantly suppressed cytokine mRNA's were those coding for IL-1a, IL1-b and IL-23. The following interleukins were not significantly induced or down regulated: IL-3, IL-6, IL-8, IL-10, IL-12, and IL-16.  IFN-g was not induced by Peg-IFN-a plus ribavirin in this study, as confirmed by kinetic RT-PCR. Likewise, no changes in expression of receptor genes were found for these interleukins. TNF-a was induced (1.7 fold change) at a marginally significant P value of 0.01. This is in agreement with previous reports from this laboratory 1 Similar results were obtained when PBMC were incubated with PEG-IFN-a alone. Among the chemokines induced were chemokines 1, 8, 10, 11, and 19, MCP-2 (SCYA8), and SCYB10 (IP-10) and among chemokines significantly down-regulated were chemokine ligand 5 (cytokine B5), and chemokines A 20, 22 and 24.

Genes involved in transcriptional regulation

As expected, STAT1 and STAT2 as well as Jak 2 were induced in PBMC following PEG-IFN-a treatment (Table 4).  Two regulatory genes related to the JAK/STAT pathway, NMI (n-myc and STAT interactor) and SOCS1 (suppressor of cytokine signaling), were also induced. These gene products regulate the STAT induction pathway, NMI in a positive fashion and SOCS1 as a negative regulator 18.  NMI protein interacts with all STATS except for STAT2 and has been reported to increase STAT mediated transcription in response to IL-2 and IFN- g.  SOCS1 binds to JAK kinases and negatively regulates the JAK signaling pathway 19 which has been reported to be induced by IL-3, EPO, CSF/CM-CSF and IFN-g.

 IFN-a treatment also results in the induction of DRAP1, IRF-2 and IRF-7.  DRAP-1 (DR1 associated protein) interacts with TATA binding protein (TBP) of TFIID and prevents the formation of an active complex, by enhancing DR-1 repression 20  IRF-2 is a competitive repressor of IRF-1, which is involved in the induction of Type I interferons as well as binding to the ISRE region of interferon inducible genes 21.  IRF-7 likewise binds to the ISRE and interferes with IRF-1 transcriptional activation.  IRF-7 is uniquely produced by hematopoietic cells 21

Genes involved in apoptosis

It has been suggested that the second stage of response to interferon treatment in hepatitis C patients is due to apoptosis of infected cells.  Thus we specifically examined those genes known to be associated with cell death (Table 4). TNFSF10 (TRAIL), a member of the TNF ligand superfamily that binds to the death signaling receptors DR4 and DR527 22, is induced 7-fold. . TNFRSF6, also known as APO-1, or FAS-1 receptor, was likewise elevated over 2-fold; TNFRSR6 contains a death domain adapter molecule and recruits caspase 8 to the activated receptor involved in the initiation of the caspase cascade 23.  RIPK2, a receptor interacting serine threonine kinase, interacts with CD40 or TNF receptor 24: its c-terminal region is also involved in caspase recruitment and activation. RIPK2 is induced nearly 2-fold. Thus the major pathway of apoptosis induced following PEG-IFN-a treatment of PBMC appears to be through the Fas/FAS ligand TRAIL system, and not through TNF-a (Figure 2). As expected, the expression of the interferon-inducible double-stranded RNA kinase (PRKR) was up-regulated approximately 3-fold by the PEG-IFN-a treatment .

Induction of known interferon-stimulated genes (ISGs) and anti-viral  proteins.

As expected, previously identified interferon related genes were induced (Table 2a,b and supplementary data).  These included genes known to be involved in the activation of the ISG gene family.  Although 7 or 8 interferon responsive factors (IRFs) have been identified, only IRF-1 (1.5 fold, p=.00056), IRF-2 (1.8 fold, p=0.003), IRF-4 (1.6, fold, p=0.003) and IRF-7 (4.0 fold, p=0.00001) were induced in PBMC.  Among other genes altered in expression were ISG-15, MX 1 and MX 2, and 2-5 Oligo-A synthetase isozymes.  The level of these genes is in agreement with those recently reported by Schlaak and co-workers 26

Discussion

In this paper we report the expression profile of genes induced in PBMC following 24 hours of treatment in vitro with PEG-IFN-a, ribavirin, and PEG-IFN-a plus ribavirin. Affymetrix GeneChips® allowed us to examine the expression of more than 22,000 human genes. To confirm the results from microarray analyses, we measured by kinetic RT-PCR 62 genes that were detected as induced in the microarray screen at a P <0.05, and found excellent agreement between the two methods (Table 3).

Ribavirin by itself had essentially no detectable effect on overall cellular gene expression.  IFN on the other hand, had a dramatic effect on global gene expression, inducing and down-regulating similar numbers of genes. We detected many more IFN-regulated genes than have previously been identified 6-8

In this study ribavirin (1-b-D-ribofuranosyl-1, 2, -triazole-3-carboxaminde) was used at what was considered physiological levels (10mg/ml). It is possible that at higher concentrations it would have greater effects in vitro. Ribavirin is a synthetic guanosine analogue with broad antiviral activities against DNA and RNA viruses 26. Although the molecular mechanism of ribavirin action in IFN/ribavirin combination therapy remains unclear, ribavirin is known to have no direct antiviral activities against HCV viral replication when it was used alone 27. Previous reports suggested that ribavirin acts as an immunomodulator and promotes T cell mediated immunity against HCV infection during IFN/ribavirin combination therapy 28,29. Ribavirin also acts as a  RNA mutagen in the HCV replicon system 30. It is possible that the induction of RNA editing enzymes such as apobec3a and ADAR will lead to an even greater mutation frequency leading to an “error catastrophe”. This is under investigation.

Since ribavirin had little effect either by itself or in the combination, we were able to merge the data ± ribavirin from 20 arrays into a 10 X 10 analysis in which the difference was presence or absence of PEG-IFN-a. We had sufficient samples to perform t-tests to give meaningful statistical and biological differences, and therefore did not have to resort to an arbitrary and non-biological cut off (e.g. 2 fold differences). We detected a total of 1061 non-redundant mRNA sequences either up regulated or down regulated following PEG-IFN-a treatment (supplementary material), many more interferon-regulated genes than have previously been identified 6-8, even when restricting our attention to genes that differed at P ≤ 0.001. The data we present were analyzed from matched experiments in which PBMC from an individual are divided and treated in parallel; data from preliminary experiments with samples from other individuals were essentially the same. It is possible that at the 24 hour time point studied some of IFN response reflect secondary events, although similar results have been found following shorter treatments of cell lines with the same IFNs (unpublished data).

We have compared our list of genes up regulated by IFN-a (supplementary material) with those previously published 6-8 as well as with the data base http:// www.lerner.ccf.org/labs/-williams. Almost all the genes identified in these earlier experiments appear in our analysis.  However, our data expand considerably the number of genes modulated by interferon. The largest percentage of genes induced by IFN-a involved cell death, defense and immunity proteins, stress response, signaling, and transcriptional regulation (Table 2).  A number of novel genes were identified in this study that may be important in the interferon response. These include the cytidine and deoxycytidine deaminase proposed to be an editing enzyme, and a large group of transcription factors. ATF4 is activated by phosphorylation of eIF2, and thus may play a major role in the regulation of amino acids and other metabolites 31. Phosphorylation of eIF2 is also a result of PKR activity. Thus these genes may be co-regulated and related to the anti-viral or apoptotic response. An homeobox protein 14 of unknown function, but that has been detected in other cell lines treated with interferon (in preparation), and viperin 17, a previously identified but uncharacterized protein that has inhibitory effects on human cytomegalovirus, were identified. Properdin, a key element of the complement alternative cascade is also induced by Peg-IFN-a.  This was rather surprising since it has been reported that properdin activity is suppressed by IFN-g 12. A possible role for interferons in the induction of the complement cascade has never been studied before and is now under investigation.

There is no published list of genes down regulated by IFN-a isoforms, so this report greatly expands our knowledge of gene suppression by IFNs. Treatment of PBMC with PEG-IFN-a results in a down regulation of overall metabolism and translational regulation, which may reflect the induction of apoptotic related genes (Table 4). A large number of mRNAs encoding ribosomal proteins are  reduced in concentration (supplemental data). If this decrease of mRNA was due to global mRNA degradation, it would be a general phenomenon and not confined to a select number of genes.

Interestingly, we were unable to detect the induction of IFN-g mRNA, although a large number of genes previously reported to be IFN-g inducible were detected. This is in agreement with our previous results from an analysis of serum samples from HCV patients treated with the consensus IFN, in which we were unable to detect increases in IFN-g or most other cytokines 1,2 This lack of IFN-g induction is in agreement with the work of Schlaak and coworkers25. Kinetic RT-PCR confirmed that either very low or non-detectable amounts of IFN-g were present in PBMC treated by PEG- IFN-a, and that the induced genes were likely due the effects of PEG-IFN-a instead of secondary induction.

We were particularly interested in cytokines induced by the combination drug treatment, since previous work from this lab had detected little cytokine activity in serum samples from hepatitis C patients treated with IFN-con1 1,2. Table 4 presents a list of cytokines and chemokines altered at the mRNA level by this treatment in vitro. One of the cytokines induced by in this study was IL-18, which was originally described as an IFN-g inducing factor (GIF) in mouse spleen cells32; IL-18 has not been previously shown to be induced by type I interferons.  An important function of IL18 is the regulation of functionally distinct subsets of T-helper cells required for cell-mediated immune responses 33  IL-18 also functions as a growth and differentiation factor for Th1 cells, and up regulates Fas ligand mediated cytotoxic activity of murine natural killer cells34 , probably through the induction of IFN-g. IL-18 may have additional roles other than the induction of IFN-g, and in PBMC this may not be its predominant role. 

Il-15, a recently discovered cytokine with T-cell stimulating activity similar to IL-2 35,36 was also induced in this study.  IL-15 is activated in monocyte/macrophages 37.  The IL-15 receptor and IL-2 receptors are both induced by PEG-IFN-a (Table 4). The high affinity receptor for IL-15 involves a complex with the IL-2 receptor 38  Thus, it was of interest that both the IL-15 receptor and IL-2 receptor (but not IL-2) are both up regulated following treatment, suggesting that IL-15 may be the major activator of T-cells following interferon treatment.

In a previous study 39, IFN-a therapy was associated with a reduction in levels of the T-helper type 2 (Th2) cytokines IL-4 and IL-10. Production of IL-1b and TNF-a by peripheral blood mononuclear cells also was found to decrease during IFN-a therapy 40  Data from our analysis confirms that the pro-inflammatory cytokines IL-1a and IL-1b and their receptors are coordinately suppressed by PEG-IFN-a, whereas TNF-a appeared to increase, although at a low level.

Apoptosis, programmed cell death, provides a mechanism for controlling the number and types of active blood cells, as well as for the elimination of viral infected cells. A large number of genes on the apoptotic pathway were induced by PEG-IFN-a  treatment. In mammalian cells, one pathway of activating the caspases is through the activation of receptors by members of the tumor necrosis family. One of the best-characterized receptors of this family is FAS (Table 4; TNFRSF-6), also known as APT-1/CD95 (apo-1) (fig. 2).  This receptor is highly induced by PEG-IFN-a, as is its ligand TNSF10 (TRAIL).  Fas and TRAIL are induced by IFN-a in PBMC of HIV patients 41.  We propose that the binding of TRAIL to FAS results in the synthesis and activation of caspase 1 (ICE-1), caspase 7 and caspase 9. Other genes involved in the regulation or induction of apoptosis also include Bcl-G, a new member of the Bcl-2 family 42.  Over-production of Bcl-G in cells induced apoptosis 43  Both RIPK1 and RIPK2 (table 4) are transducers and integration signals for the immune system 24.  These serine threonine kinase associated receptors are part of the TNF receptor complex and are implicated in activation of NF-kappa B and cell death. TSSC3 (Table 4) is the human homologue of the mouse apoptosis gene TDAG5144.

In conclusion, a large number of genes are induced by the in vitro treatment of PBMC with PEG-IFN-a.  Our list expands considerably the number of genes previously reported to be IFN-a induced, and indicates that PEG-IFN-a also suppresses a large number of genes, particularly those related to overall biosynthesis and metabolism in the cell. Many of these may be secondary effects of primary gene induction. We have also demonstrated that ribavirin, used clinically in combination with PEG-IFN-a, has little effect upon steady-state mRNA levels, and therefore must function by another mechanism. Few interleukins are induced by PEG-IFN-a, although genes involved with cell motility, such as chemokines, are induced.  The finding that PEG-IFN-a also induces large numbers of genes involved in cell death and regulation of apoptosis gives credence to the fact that PEG-IFN-a is not only an antiviral agent but is intrinsically involved in cell regulation and may act as a tumor suppressor

 

Acknowledgements

This work was supported by a grant from the US National Institutes of Health, NIDDK DK60309 (MWT) and a pilot grant from Schering-Plough (MWT). The microarray studies were carried out using the facilities of the Center for Medical Genomics at Indiana University School of Medicine. The Center for Medical Genomics is supported in part by grants from the Indiana 21st Century Research and Technology Fund and the Indiana Genomics Initiative (supported in part by the Lilly Endowment, Inc.).

Competing interests statement:

The authors declare that they have no competing financial interests.

References

  1. Cotler, S.J., Reddy, K.R., McCone, J., Wolfe, D.L., Liu, A., Craft, T.R., Ferris, M., Conrad, A., Albrecht, J., Morrisey, M., Ganger, D., Rosenblate, H., Blatt, L.M., Jensen, D.M., Taylor, M.W.(2001)  An analysis of acute changes in interleukin-6 levels after treatment of hepatitis C with consensus interferon.  J. Interferon and Cytokine Res. 21, 1011-19. 

2. Cotler, S.J., Craft, T., Ferris, M., Morrisey, M., McCone, J., Reddy, R.K., Conrad, A., Jensen, D.M., Albrecht, J., Taylor, M.W. (2002) Induction of IL-1Ra in resistant and responsive hepatitis C patients following treatment with IFN-con1. J. Interferon and Cytokine Res. 22, 549-554. 

3. McClintick, J.N., Jerome, R.E., Nicholson, C.R., Crabb, D.W., Edenberg, H.J. (2003) Reproducibility of oligonucleotide arrays using small samples.  BMC Genomics 30 :4 . 

4. Eisen, M.B., Spellman, P.T., Brown, P.O., Botstein, D (1998), Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci USA  95:14863-8.

5. Higuchi, R., and Watson, R.  Kinetic PCR analysis using a CCD camera and without using oligonucleotide probes.  PCR Applications (eds Innis, M.A., Gelfand, D.H. and Sninsky, J.J.(1999) 263-284 (Academic, San Diego). 

6. de Veer, M.J., Holko, M., Frevel, M., Walker, E., Der, S., Paranjape, J.M., Silverman, R.H.,Williams, B.R.(2001) Functional classification of interferon-stimulated genes identified using microarrays. J Leukoc Biol; 69:912-20.

  1. Der S.D. Zhou A., Williams B.R., Silverman  R.H. (1998) Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays. Proc. Natl. Acad. Sci. USA. (1998). 95:15623-8.

8. Ji, X., Cheung, R., Cooper, S., Li, Q., Greenberg, H.B., He, X.S. (2003) Interferon alfa regulated gene expression in patients initiating interferon treatment for chronic hepatitis C  Hepatology  37, 610-21.

9. Rajeevan M.S., Vernon S.D., Taysavang N., Unger E.R. (2001) Validation of array-based gene expression profiles by real-time (kinetic) RT-PCR, J. Molec. Diagn.  3.26-31.

10.  Jarmuz, A., Chester, A., Bayliss, J., Gisbourne, J., Dunham, I., Scott, J., Navaratnam, N (2002) An anthropoid-specific locus of orphan C to U RNA-editing enzymes on chromosome 22Genomics.  79:  285-96.

11.  Samuel, C.E.  Clin. Microbiol. Rev (2001) Antiviral actions of interferons 14 :778-809

12.  Schwaeble W.J, Reid K.B.M.(1999) Does properdin crosslink the cellular and the humoral immune response?. Immunology Today, 20, 17-21.

  1. Zhang, C., Gao, C., Kawauchi, J., Hashimoto, Y., Tsuchida, N., Kitajima, S. (2002) Transcriptional activation of the human stress-inducible transcriptional repressor ATF3 gene promoter  p53  Biochem. Biophys. Res. Commun., 297:1302-10. 
  2. Tajima T, Hattorri T, Nakajima T, Okuhara K, Sato K, Abe S, Nakae J, Fujieda K. (2003) Sporadic heterozygous frameshift mutation of HESX1 causing pituitary and optic nerve hypoplasia and combined pituitary hormone deficiency in a Japanese patient.J Clin  Endocrinol Metab  88:45-50
  3. D'Arena G, Di Renzo N, Brugiatelli M, Vigliotti ML, Keating MJ. (2003) Biological and clinical heterogeneity of B-cell chronic lymphocytic leukemia. Leuk Lymphoma  44; 223-8
  4. Khodadoust., M.M., Khan, K.D., Bothwell, A.L (1999) Complex regulation of Ly-6E gene transcription in T cells by IFNs.   J. Immunol.; 163: 811-9. 
  5. Chin,K.C. and Cresswell,P (2001) Viperin (cig5), an IFN-inducible antiviral protein directly induced by human cytomegalovirus Proc. Natl. Acad. Sci. U.S.A. 98 15125-15130.
  6. Shuai, K. (2000). Modulation of STAT signaling by STAT-interacting proteins .Oncogene 19. 2638-44. 
  7. O'Shea, J.J., Gadine, M., and Schreiber, R.D (2002) Cytokine signaling in 2002. New surprises in the JAK/STAT pathway.  Cell.  109.  121-131.
  8. Kim, S., Na, J.G., Hampsey, M., Reinberg, D (1997)The DR1/DRAM heterodimer is a global repressor of transcription in vivo.  Proc. Natl. Acad. Sci. USA. 94. 820-825. 
  9. Taniguchi T . (1997)  Transcription factors IRF-1 and IRF-2 linking the immune response and tumor suppressors  J. Cell Physiol.  173: 128-130. 
  10. Baetu, T.M., and Hiscott, J. (2002) .  On the TRAIL to apoptosis .Cytokine Growth Factor Rev. 13, 199-207. 
  11. Hymowitz, S.G., Christinger,. H.W., Fuh, G., Ultsch, M., O'Connell, M., Kelley, R.F., Ashkenazi, A., de Vos, A.M. (1999) Triggering cell death: the crystal structure of Apo2L/TRAIL in a complex with death receptor 5 . Mol. Cell  4: 563-71., 
  12. Kobayashi, K., Inohara, N., Hernandez, L.D., Galan, J.E., Nunez, G., Janeway, C.A.,  Medzhitov, R., Flavell, R.A (2002). RICK/Rip2/CARDIAK mediates signaling for receptors of the innate and adaptive immune systems. Nature  416:194-9.
  13. Schlaak, J.F., Hilken, C.M., Costa-Pereira, A.P., Strobl B., Aberger F., Frischauf A.M., Kerr I.M. (2002) Cell type and donor specific transcriptional responses to interferon alpha: Use of customized gene arrays  Jour. Biol. Chem. 277, 49428-37. 
  14. Sidwell, R.W., Huffman, J.H., Khare, G.P., Allen, L.B., Witkowski, J.T., Robins, R.K (1972) Broad-spectrum antiviral activity of Virazole: 1-beta-D-ribofuranosyl-1,2,4-triazole-3-carboxamide . Science; 177: 705-6. 
  15. Lee, J.H., von Wagner, M., Roth, W.K., Teuber, G., Sarrazin, C., Zeuzem, S.(1998) Effect of ribavirin on virus load and quasispecies distribution in patients infected with hepatitis C virus.  J. Hepatol. 29 : 29-35. 
  16. Martin, J., Navas, S., Quiroga, J.A., Pardo, M., Carreno, V (1998) Effects of the ribavirin-interferon alpha combination on cultured peripheral blood mononuclear cells from chronic hepatitis C patients. Cytokine. 1998 10, 635-44.
  17. Tam, R.C., Pai, B., Bard, J., Lim, C., Averett, D.R., Phan, U.T., Milovanovic, T  (1999) Ribavirin polarizes human T cell responses towards a Type 1 cytokine profile. J. Hepatol.  30:376-82. 
  18. Contreras A.M., Hiasa Y., He W., Terella A., Schmidt E.V., Chung R.T. (2002) Viral RNA mutations are region specific and increased by ribavirin in a full-length hepatitis C virus replication system J. Vir. 76, 8505-17,
  19. Rutkowski, D.T., Kaufman, R.J.(2003) All roads lead to ATF4 (2003)  Dev. Cell ;4. 442-4. 
  20. Nakanishi, K., Yoshimoto, T., Tsutsui, H., Okamura, H . (2001).  Interleukin-18 regulates both Th1 and Th2 responses (2001) Annu. Rev. Immunol.  19:423-74. 
  21. Dao, T., Ohashi, K., Kayano, T., Kurimoto, M., Okamura, H. (1996) Interferon-gamma-inducing factor, a novel cytokine, enhances Fas ligand-mediated cytotoxicity of murine T helper 1 cell   Immunol.  173:230-5. 
  22. Tsutsui, H., Nakanishi, K., Matsui, K., Higashino, K., Okamura, H., Miyazawa, Y., Kaneda, K, (1996)  IFN-gamma-inducing factor up-regulates Fas ligand-mediated cytotoxic activity of murine natural killer cell clones (1996). J. Immunol.  157 :3967-73. 
  23. Trentin L, Zambello R, Facco M, Sancetta R, Agostini C, Semenzato G. (1997) .  Interleukin-15: a novel cytokine with regulatory properties on normal and neoplastic B lymphocytes  Leuk. Lymph.  27: 35-42. 
  24. Waldmann,T., Tagaya, Y., and Bamford, R (1998). Interleukin-2, interleukin-15, and their receptors.  Int. Rev. Immunol.. 16: 205-226.
  25. Grabstein KH, Eisenman J, Shanebeck K, Rauch C, Srinivasan S, Fung V, Beers C, Richardson J, Schoenborn MA, Ahdieh M, et al. (1994) Cloning of a T-cell growth factor that interacts with the beta chain of the interleukin-2 receptor.   Science.  264: 965-968. 
  26. Burton JD, Bamford RN, Peters C, Grant AJ, Kurys G, Goldman CK, Brennan J, Roessler E, Waldmann TA. (1994)  A lymphokine, provisionally designated interleukin T and produced by a human adult T-cell leukemia line, stimulates T-cell proliferation and the induction of lymphokine-activated killer cells. Proc. Natl. Acad. Sci.  USA  91: 4935-4939.
  27. Cacciarelli TV, Martinez O.M., Gish R.G., Villanueva J.C., Krams S.M. (1996) Immunoregulatory cytokines in chronic hepatitis C virus infection: pre- and posttreatment with interferon alfa. Hepatology. 24, 6-9.
  28. Kishihara, Y., Hayashi, J., Yoshimura, E., Yamaji, K., Nakashima, K., Kashiwagi, S. (1996) IL-1 beta and TNF-alpha produced by peripheral blood mononuclear cells before and during interferon therapy in patients with chronic hepatitis C.   Dig. Dis. Sci; 41, 315-21.
  29. Stylianou, E., Aukrust, P., Bendtzen, K., Muller, F., Froland, S.S (2000).  Interferons and interferon (IFN)-inducible protein 10 during highly active anti-retroviral therapy (HAART)-possible immunosuppressive role of IFN-alpha in HIV infection.  Clin. Exp. Immunol.   119:479-85. 

 

  1. Guo, B., Godzik, A., Reed, J.C. (2001) Bcl-G, a novel pro-apoptotic member of the Bcl-2 family. J. Biol. Chem.  276: 2780-5. 

  

  1. Lee, M.P., Feinberg, A.P. (1998)  Genomic imprinting of a human apoptosis gene homologue, TSSC3.  Cancer Res.  58:1052-6. 

 

 

TABLES

 

Table 1. Number of genes modulated by PEG-IFN-a and ribavirin and the combination at different P values.

 

Table 2. Classes of genes induced or down regulated by Peg-IFN-a

 

Table 3. Comparison  of micro-array levels and kinetic real time PCR of genes modulated by Peg-IFN-a

 

Table 4. Induction or down regulation of specific classes of genes related to cytokines, chemokines, transcription factors, inflammation and apoptosis.

 

 

 

Figures

 

 

Figure 1. Two way hierarchical clustering (using CLUSFLAVOR 6.0) shows that there is a major difference in overall gene expression between samples treated with or without interferon, but that samples from the same blood draw (indicated by common first and last character; middle character indicates treatment group, with 2 = no treatment, 3 = PEG-IFN, 4 = ribavirin, 5 = PEG-IFN + Ribavirin) cluster together whether or not treated with ribavirin (e.g. D41 and D21 without interferon, D31 and D51 with interferon). The region shown was selected from the full dataset because it shows several interferon induced genes (labeled at right), and demonstrates clear differences in their pattern of expression

 

Figure 2 . Apoptotic pathway. Genes inside squares are induced by PEG-IFN-a

 

 

Table 1. Number of genes modulated by PEG-IFN-a and ribavirin and the combination at different P values.

 

 

Random Probability†

No drug vs. Ribavirin

No drug vs. Interferon

IFN vs IFN + Ribavirin

No drug vs combined treatment.

Arrays compared

 

5 x 5

5 x 5

5 x 5

10 x 10

Genes detected*

10,709

10,637

10,823

10,745

10,709

P ≤ 0.05

535

254

1,437

478

3,372

P ≤ 0.01

107

48

546

85

2,043

P ≤ 0.001

10

3

162

2

1,061

P ≤ 0.0001

1

0

52

0

563

 

 

 

 

 

 

*Present in at least half of the arrays in at least one experimental condition

†Number of genes expected to meet the criterion by chance, based on the normal distribution.

#The increase in numbers is not due to an effect of the drug combination, but rather to the great increase in statistical power due to the larger number of arrays compared.

 

 

Table 2. Classes of genes induced or down regulated by Peg-IFN-a

 

 

Gene Ontology

# on array

Induced

% 

Repressed

% 

Total changed

%

Ontology total (Database)

8627

780

9.0%

863

10.0%

1643

19.0%

apoptosis regulator

25

3

12.0%

1

4.0%

4

16.0%

biosynthesis

184

8

4.3%

61

33.2%

69

37.5%

carbohydrate metabolism

65

5

7.7%

10

15.4%

15

23.1%

catabolism

120

11

9.2%

11

9.2%

22

18.3%

cell cycle

153

6

3.9%

6

3.9%

12

7.8%

cell death

147

31

21.1%

7

4.8%

38

25.9%

Cell motlity

94

14

14.9%

11

11.7%

25

26.6%

cell organization and biogenesis

119

6

5.0%

7

5.9%

13

10.9%

cell proliferation

129

8

6.2%

11

8.5%

19

14.7%

cell shape and cell size control

68

5

7.4%

5

7.4%

10

14.7%

cell surface linked signal transduction

177

21

11.9%

13

7.3%

34

19.2%

cell-cell signaling

115

14

12.2%

7

6.1%

21

18.3%

defense/immunity protein

47

10

21.3%

2

4.3%

12

25.5%

enzyme regulators

130

14

10.8%

4

3.1%

18

13.8%

enzymes

1210

68

5.6%

66

5.5%

134

11.1%

Intracellular signaling cascade

187

20

10.7%

11

5.9%

31

16.6%

ligand binding

1184

107

9.0%

140

11.8%

247

20.9%

nucleic acid metabolism

489

19

3.9%

19

3.9%

38

7.8%

protein biosynthesis

184

4

2.2%

59

32.1%

63

34.2%

protein degradation

82

9

11.0%

8

9.8%

17

20.7%

protein modification

209

11

5.3%

9

4.3%

20

9.6%

protein targeting

53

2

3.8%

4

7.5%

6

11.3%

response to external stimulus

400

72

18.0%

41

10.3%

113

28.3%

signal transducers

492

48

9.8%

43

8.7%

91

18.5%

stress reponse

296

42

14.2%

27

9.1%

69

23.3%

transcription regulator

193

35

18.1%

19

9.8%

54

28.0%

translation regulators

44

0

0.0%

8

18.2%

8

18.2%

transport

326

14

4.3%

21

6.4%

35

10.7%

Total

6922

607

8.8%

631

9.1%

1238

17.9%

 

 

Table 3. Comparison  of micro-array levels and kinetic real time PCR of genes modulated by Peg-IFN-a

 

 

 

GeneSymbol

GenBank Acc. No.

Microarrays1

RTQPCR2

 

GeneSymbol

GenBank Acc. No.

Microarrays1

RTQPCR2

ADAR

NM_001111.2

1.9

2.7

 

IRF1

NM_002198.1

1.5

1.9

AIM1

U83115.1

1.2

0.3

 

IRF2

NM_002199.2

2.1

1.7

AIM2

NM_004833.1

1.8

4.6

 

IRF4

NM_002460.1

1.7

1.8

ATF3

NM_001674.1

4.5

9.3

 

ISG20

U88964

4.7

9.7

BAG1

NM_004323.2

1.7

2.3

 

ISGF3G

NM_006084.1

1.6

1.6

BCL2

NM_000633.1

1.3

1.7

 

JAK2

NM_004972.2

2.7

4.0

BST2

NM_004335.2

4.1

4.0

 

LOC55893

NM_018660.1

0.8

10.0

CASP1

AI719655

2.1

2.3

 

MMP9

NM_004994.1

0.4

0.3

CASP3

NM_004346.1

1.2

2.5

 

MNDA

NM_002432.1

6.2

6.2

CCR2

NM_000647.2

0.6

0.3

 

MT2A

NM_005953.1

1.8

10.6

CD80

NM_005191.1

2.4

9.4

 

MX1

NM_002462.1

7.3

37.4

EEF1A1

NM_001402.1

0.8

0.7

 

MX2

NM_002463.1

5.7

22.9

EIF2B4

AF112207.1

0.8

0.6

 

NFKB1

M55643.1

1.3

1.3

EIF2S1

NM_004094.1

1.2

0.8

 

NMI

NM_004688.1

2.8

2.4

G1P3

NM_022873.1

4.4

27.7

 

OAS1

NM_016816.1

11.7

23.2

GAPDH

NULL

0.6

0.5

 

OAS2

NM_016817.1

5.1

10.9

GBP1

BC002666.1

3.9

6.3

 

OAS3

NM_006187.1

11.9

42.9

GBP2

NM_004120.2

1.6

2.6

 

PLSCR1

NM_021105.1

3.2

3.1

GCH1

NM_000161.1

3.0

4.0

 

PML

NM_002675.1

3.9

4.9

HIF1A

NM_001530.1

0.8

0.7

 

PRKR

NM_002759.1

3.3

8.7

HLA-C

U62824.1

1.4

1.9

 

PRKRIR

AF081567.1

0.8

0.9

ICSBP1

AI073984

1.4

1.0

 

PSMB8

U17496.1

2.0

0.4

IFI16

NM_005531.1

2.3

3.0

 

PSMB9

NM_002800.1

2.1

2.1

IFI27

NM_005532.1

36.8

356.1

 

RELA

M62399.1

1.3

1.5

IFI35

BC001356.1

5.6

6.1

 

SP110

NM_004509.1

2.4

3.0

IFIT4

NM_001549.1

82.2

11.7

 

SSA1

NM_003141.1

2.0

3.1

IFITM2

NM_006435.1

2.8

6.2

 

STAT1

NM_007315.1

2.2

3.6

IL12RB2

NM_001559.1

2.1

2.9

 

STAT2

NM_005419.1

2.8

4.4

IL15

NM_000585.1

2.1

2.5

 

TIMP1

NM_003254.1

0.2

0.4

IL16

NM_004513.1

0.9

1.1

 

TNFSF10

U57059.1

9.0

14.3

IL1RN

AW083357

6.8

17.1

 

TRIM22

AA083478

2.7

5.3

IL8

NM_000584.1

0.7

0.8

 

TUBB2

BC004188.1

1.1

1.0

INDO

M34455.1

4.5

9.3

 

WARS

M61715.1

2.3

3.0

 

 

 

 

 

 

 

 

 

1PValue≤ 0.05

 

 

 

 

 

 

 

 

2Real-Time RT-PCR

 

 

 

 

 

 

 

 

 

 

Table 4. Induction or down regulation of specific classes of genes related to cytokines, chemokines, transcription factors, inflammation and apoptosis.

 

 

 

Gene

Fold Change

P-value

Description

Gene Family

 

 

 

 

Interleukins

Induced

 

 

 

 

IL-1Ra

7.5

<0.0001

interleukin 1 receptor antagonist

 

IL-1 member 9

3.3

0.02

interleukin 1-related protein 2

 

IL-18

3

0.0018

interleukin 18 (interferon-gamma-inducing factor)

 

IL-15

1.98

<0.0001

interleukin 15

 

TNF-a

1.69

0.01

tumor necrosis factor alpha

 

Suppressed

 

 

 

 

Il-1a

-4.11

0.008

interleukin 1 alpha

 

IL-1b

-2.1

0.02

interleukin 1 beta

 

IL-23

-1.69

0.0007

interleukin 23, alpha subunit p19

 

Induced

 

 

 

Interleukin Receptors

IL-15

3.22

<0.0001

interleukin 15 receptor

 

IL-2

2.26

0.0001

interleukin 2 receptor

 

IL-12

2.2

0.0001

interleukin 12 receptor

 

Suppressed

 

 

 

 

IL-13RA

-1.54

0.0036

interleukin 13 receptor alpha

 

IL-17R

-1.49

0.003

interleukin 17 receptor

 

IL-21R

-1.64

0.0008

interleukin 21 receptor

 

IL-4R

-1.42

0.0002

interleukin 4 receptor

 

IL-1R

-1.92

0.0007

interleukin 1 receptor

 

Induced

 

 

 

 

CXCL11

48.37

0.003

chemokine (C-X-C motif) ligand 11

Chemokines

CXCL10

8.76

0.0002

chemokine (C-X-C motif) ligand 10

 

CCL8

5.1

<0.0001

chemokine (C-C motif) ligand 8

 

CCL19

2.56

0.002

chemokine (C-C motif) ligand 19

 

XCL1

1.33

0.011

chemokine (C motif) ligand 1

 

Suppressed

 

 

 

 

CKLF1

-1.52

0.009

chemokine-like factor 1

 

CCL22

-1.69

0.0006

chemokine (C-C motif) ligand 22

 

CCL20

-2.58

0.0177

chemokine (C-C motif) ligand 20

 

CCL24

-2.72

0.0024

chemokine (C-C motif) ligand 24

 

CXCL1

-2.89

0.0046

chemokine (C-X-C motif) ligand 1

 

CXCL5

-4.58

0.0054

chemokine (C-X-C motif) ligand 5

Chemokine Receptors

Induced

 

 

 

 

CMKLR1

1.79

0.0038

chemokine-like receptor 1

 

CCR1

1.77

0.0007

chemokine (C-C motif) receptor 1

 

CCR5

1.5

0.002

chemokine (C-C motif) receptor 5

Pathways

 

 

 

 

Transcription Regulation

Induced

 

 

 

 

PML

4.35

4E-05

promyelocytic leukemia

 

ATF3

4.2

0.0001

activating transcription factor 3

 

IFI35

3.48

2E-05

interferon-induced protein 35

 

STAT2

3.4

9E-05

Stat2 type a

 

TFEC

3.17

0.0016

transcription factor EC

 

IRF7

2.95

2E-05

interferon regulatory factor 7

 

PRKR

2.95

1E-05

protein kinase,IFN-inducible double stranded RNA dependent

 

STAT1

2.46

0.0002

signal transducer and activator of transcription 1, 91kDa

 

ZNF147

2.23

2E-05

zinc finger protein 147 (estrogen-responsive finger protein)

 

STAT2

2.2

0.0002

signal transducer and activator of transcription 2, 113kDa

 

SP100

2.13

1E-05

nuclear antigen Sp100

 

NMI

2.11

5E-05

N-myc (and STAT) interactor

 

TRIM22

2.02

2E-05

tripartite motif-containing 22

 

NFE2L3

1.93

0.0012

nuclear factor (erythroid-derived 2)-like 3

 

KLF5

1.87

0.0005

Kruppel-like factor 5 (intestinal)

 

HIRA

1.83

0.0024

HIR histone cell cycle regulation defective homolog A (S.cerevisiae)

 

TEL2

1.79

0.0003

transcription factor ets

 

IRF2

1.74

0.0046

interferon regulatory factor 2

 

TARBP1

1.69

0.0046

TAR (HIV) RNA binding protein 1

 

CREG

1.6

0.0047

cellular repressor of E1A-stimulated genes

 

SP140

1.6

0.0002

SP140 nuclear body protein

 

IRF4

1.58

0.004

interferon regulatory factor 4

 

SUPT3H

1.57

0.0002

Transcription factor SUPT3H (SUPT3H) mRNA, complete cds

 

DRAP1

1.56

2E-05

DR1-associated protein 1 (negative cofactor 2 alpha)

 

ELF1

1.55

7E-05

E74-like factor 1 (ets domain transcription factor)

 

TCF4

1.55

0.0011

transcription factor 4

Apoptosis

Induced

 

 

 

 

TNFSF10

7.17

0.0001

tumor necrosis factor (ligand) superfamily, member 10

 

CD38

5.25

0.0007

CD38 antigen (p45)

 

MX1

4.08

1E-05

myxovirus (influenza virus) resistance 1, IFN-inducible protein p78

 

BCLG

3.8

6E-05

apoptosis regulator BCL-G

 

PRKR

2.95

1E-05

protein kinase, IFN-inducible double stranded RNA dependent

 

TNFRSF6

2.29

0.0004

tumor necrosis factor receptor superfamily, member 6

 

CASP1

1.97

3E-05

caspase 1, apoptosis-related cysteine protease

 

CASP10

1.88

0.0003

caspase 10, apoptosis-related cysteine protease

 

RIPK2

1.86

0.0004

receptor-interacting serine-threonine kinase 2

 

TSSC3

1.81

0.0002

tumor suppressing subtransferable candidate 3

 

CFLAR

1.74

5E-05

CASP8 and FADD-like apoptosis regulator

 

RIPK1

1.72

1E-05

receptor (TNFRSF)-interacting serine-threonine kinase 1

 

BAG1

1.64

0.0004

BCL2-associated athanogene

 

CASP7

1.55

0.0003

caspase 7, apoptosis-related cysteine protease

 

TIA1

1.55

0.0006

TIA1 cytotoxic granule-associated RNA binding protein

 

GADD45B

1.55

0.0009

growth arrest and DNA-damage-inducible, beta

 

TNFRSF9

1.52

0.0031

tumor necrosis factor receptor superfamily, member 9

 

CUL1

1.51

0.007

cullin 1

 

Suppressed

 

 

 

 

DEFCAP

-1.53

0.0001

death effector filament-forming Ced-4-like apoptosis protein

 

LGALS1

-1.65

0.0024

lectin, galactoside-binding, soluble, 1 (galectin 1)

 

CD14

-2.7

3E-05

CD14 antigen

Inflammatory Response

Induced

 

 

 

 

CXCL10

8.76

0.0002

chemokine (C-X-C motif) ligand 10

 

IL1RN

5.96

2E-05

interleukin 1 receptor antagonist

 

CCL8

5.1

1E-05

chemokine (C-C motif) ligand 8

 

NMI

2.42

<0.00001

N-myc (and STAT) interactor

 

APOL3

2.16

8E-05

apolipoprotein L, 3

 

RIPK2

2.04

1E-05

receptor-interacting serine-threonine kinase 2

 

CCR1

1.77

0.0007

chemokine (C-C motif) receptor 1

 

BLNK

1.63

0.001

B-cell linker

 

Suppressed

 

 

 

 

CCL22

-1.69

0.0007

chemokine (C-C motif) ligand 22

 

IL1R1

-1.92

0.0007

interleukin 1 receptor, type I

 

LTA4H

-2.56

0.0001

leukotriene A4 hydrolase

 

S100A8

-2.61

1E-05

S100 calcium binding protein A8 (calgranulin A)

 

PROCR

-2.72

0.0005

protein C receptor, endothelial (EPCR)

 

S100A9

-3.23

<0.00001

S100 calcium binding protein A9 (calgranulin B)

 

S100A12

-3.74

1E-05

S100 calcium binding protein A12 (calgranulin C)

 

CD14

-4.15

<0.00001

CD14 antigen

 

FPR1

-4.34

0.0008

formyl peptide receptor 1

 

 

 

 

 

 

 
 

 

Figures

Figure 1. Two way hierarchical clustering (using CLUSFLAVOR 6.0) shows that there is a major difference in overall gene expression between samples treated with or without interferon, but that samples from the same blood draw (indicated by common first and last character; middle character indicates treatment group, with 2 = no treatment, 3 = PEG-IFN, 4 = ribavirin, 5 = PEG-IFN + Ribavirin) cluster together whether or not treated with ribavirin (e.g. D41 and D21 without interferon, D31 and D51 with interferon). The region shown was selected from the full dataset because it shows several interferon induced genes (labeled at right), and demonstrates clear differences in their pattern of expression

 

Figure 2 . Apoptotic pathway. Genes inside squares are induced by PEG-IFN-a.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Supplementary table 1 ( up-regulated genes)

 

GeneSymbol

Genbank

Fold Change

Description

ACK1

NM_005781.2

1.68

activated p21cdc42Hs kinase

ADAM19

Y13786.2

1.95

a disintegrin and metalloproteinase domain 19

ADAMDEC1

NM_014479.1

2.18

ADAM-like, decysin 1

ADAR

NM_001111.2

1.91

adenosine deaminase, RNA-specific

ADPRTL3

AF083068.1

1.68

ADP-ribosyltransferase

AGRN

AI424797

1.84

agrin

AIM2

NM_004833.1

1.8

absent in melanoma 2

AKAP2

NM_007203.1

2.15

A kinase (PRKA) anchor protein 2

ANKHZN

NM_016376.1

2.05

ANKHZN protein

APOBEC3A

U03891.2

19.16

apolipoprotein B mRNA editing enzyme,

APOL1

AF323540.1

2.97

apolipoprotein L, 1

APOL3

NM_014349.1

2.18

apolipoprotein L, 3

APOL6

NM_030641.1

3.18

apolipoprotein L, 6

ARHGAP8

AA533284

1.5

Rho GTPase activating protein 8

ARHGEF11

NM_014784.1

1.67

Rho guanine nucleotide exchange factor (GEF) 11

ARHGEF3

NM_019555.1

1.77

Rho guanine nucleotide exchange factor (GEF) 3

ATF3

NM_001674.1

4.53

activating transcription factor 3

ATF4

NM_001675.1

1.31

activating transcription factor 4 (tax-responsive enhancer element B67)

ATOX1

NM_004045.1

1.68

ATX1 antioxidant protein 1 homolog (yeast)

ATP10A

N35112

2.68

ATPase, Class V, type 10A

B4GALT5

BF691447

2.06

UDP-Gal:betaGlcNAc beta 1,4- galactosyltransferase, polypeptide 5

BAG1

AF116273.1

1.89

BCL2-associated athanogene

BAZ1A

NM_013448.1

1.65

bromodomain adjacent to zinc finger domain, 1A

BCLG

NM_030766.1

4.17

apoptosis regulator BCL-G

BCRP1

AV755522

3.82

clone IMAGE:4074138, mRNA, mRNA sequence

BF

NM_001710.1

7.85

B-factor, properdin

BLNK

NM_013314.1

1.88

B-cell linker

BLVRA

AA740186

2.78

biliverdin reductase A

BRD2

D42040.1

1.42

bromodomain containing 2

BRDG1

NM_012108.1

2.29

BCR downstream signaling 1

BST2

NM_004335.2

4.09

bone marrow stromal cell antigen 2

BTN2A1

NM_007049.1

1.41

butyrophilin, subfamily 2, member A1

BTN3A1

NM_007048.1

1.58

butyrophilin, subfamily 3, member A1

BTN3A3

NM_006994.2

1.61

butyrophilin, subfamily 3, member A3

C12orf6

NM_020367.1

2.19

chromosome 12 open reading frame 6

C14orf3

NM_012111.1

1.26

chromosome 14 open reading frame 3

C1GALT1

NM_020156.1

1.73

core 1 UDP-galactose:N-acetylgalactosamine-alpha-R beta 1,3-galactosyltransferase

C1orf28

NM_024529.1

1.74

chromosome 1 open reading frame 28

C1orf29

NM_006820.1

4.72

chromosome 1 open reading frame 29

C20orf18

BE788439

1.78

chromosome 20 open reading frame 18

C4S-2

NM_018641.1

2.63

chondroitin 4-O-sulfotransferase 2

C6orf37

AW246673

3.77

chromosome 6 open reading frame 37

C7orf14

AW008531

1.46

chromosome 7 open reading frame 14

CACNA1A

AA769818

2.78

calcium channel, voltage-dependent, P/Q type, alpha 1A subunit

CAPN2

M23254.1

1.56

calpain 2, (m/II) large subunit

CASP1

U13699.1

2.42

caspase 1, apoptosis-related cysteine protease

CASP10

NM_001230.1

2.18

caspase 10, apoptosis-related cysteine protease

CASP4

AL050391.1

1.55

caspase 4, apoptosis-related cysteine protease

CASP7

NM_001227.1

1.61

caspase 7, apoptosis-related cysteine protease

CAST

AF327443.1

1.59

calpastatin

CBR1

BC002511.1

2.45

carbonyl reductase 1

CCL8

AI984980

14.49

chemokine (C-C motif) ligand 8

CCND3

NM_001760.1

1.36

cyclin D3

CCR1

AI421071

2.01

chemokine (C-C motif) receptor 1

CCR5

NM_000579.1

1.57

chemokine (C-C motif) receptor 5

CD164

AF263279.1

2

CD164 antigen, sialomucin

CD2AP

NM_012120.1

2.28

CD2-associated protein

CD38

NM_001775.1

6.69

CD38 antigen (p45)

CD47

BG230614

1.27

CD47 antigen (integrin-associated signal transducer)

CD69

L07555.1

1.81

CD69 antigen (p60, early T-cell activation antigen)

CD80

NM_005191.1

2.42

CD80 antigen (CD28 antigen ligand 1, B7-1 antigen)

CD83

NM_004233.1

2.03

CD83 antigen

CEB1

NM_016323.1

9.6

cyclin-E binding protein 1

CECR1

NM_017424.1

2.1

cat eye syndrome chromosome region

CENTD1

AB011152.1

1.41

centaurin, delta 1

CFLAR

AF041461.1

1.79

CASP8 and FADD-like apoptosis regulator

CHSY1

NM_014918.1

1.43

carbohydrate (chondroitin) synthase 1

cig5

AI337069

18.65

vipirin

CKS1B

NM_001826.1

1.37

CDC28 protein kinase regulatory subunit 1B

CMT2

NM_014628.1

1.33

gene predicted from cDNA with a complete coding sequence

CNP

BC001362.1

1.75

2`,3`-cyclic nucleotide 3` phosphodiesterase

COPEB

BE675435

2.16

core promoter element binding protein

COX17

NM_005694.1

1.42

COX17 homolog, cytochrome c oxidase assembly protein (yeast)

CRACC

NM_021181.2

3.98

19A24 protein

CREG

NM_003851.1

1.78

cellular repressor of E1A-stimulated genes

CRFG

NM_012341.1

1.33

G protein-binding protein CRFG

CRSP6

AF105421.1

1.36

cofactor required for Sp1 transcriptional activation, subunit 6,

CXCL10

NM_001565.1

30.6

chemokine (C-X-C motif) ligand 10

CXCL11

AF002985.1

56.27

chemokine (C-X-C motif) ligand 11

CXCL9

NM_002416.1

2.93

chemokine (C-X-C motif) ligand 9

CYCS

BC005299.1

1.29

cytochrome c, somatic

D13S106E

NM_005800.1

1.33

highly charged protein

DAPP1

NM_014395.1

1.66

dual adaptor of phosphotyrosine and 3-phosphoinositides

DC13

NM_020188.1

1.37

DC13 protein

DCK

NM_000788.1

1.74

deoxycytidine kinase

DDX1

NM_004939.1

1.21

DEAD/H (Asp-Glu-Ala-Asp/His) box polypeptide 1

DEFB1