Differential Gene Expression of Liver Tissue and Blood From Individuals With Chronic Viral Hepatitis

Overview

The purpose of this research is to study body materials like blood proteins as well as white blood cell and liver cellular RNA in individuals with liver diseases such as chronic viral hepatitis with or without hepatoma and autoimmune liver disease. Presently it is not understood how infection with chronic viral hepatitis or autoimmune liver disease damages the liver. This research study enroll patients with either chronic viral hepatitis with or without hepatoma or autoimmune liver disease. The purpose of this study is to find the genes that are expressed in both the circulating white blood cells and the liver of patients with varying degrees of liver damage of different causes. Genes are biological messengers some of which determine how the body responds to injury. We anticipate that results from Differential Gene Expression (DGE) analysis will allow us to make predictions about likelihood of disease progression and/or response to treatment. In addition we will test the blood for markers of injury. The blood collected will be prepared differently from the liver tissue. We will use technologies to express pure proteins and then we will investigate the functions of these proteins. Nearly all drugs act on proteins, not genes, so understanding proteins is the key to really effective new medicines. Similarly the first signs of ill health appear in changes to the body's blood proteins, making them the most sensitive diagnostic indicators. The studies we plan are called proteomics. We will later correlate the patterns of gene expression in both circulating white blood cells and the liver tissue with clinical outcome and patterns of proteins measured in blood and we hope to gain an understanding of how the disease process occurs, which may in turn help us to make more precise diagnoses and develop new forms of treatment. These techniques that we use are still experimental and so we do not yet know if they will be helpful in monitoring changes which may help us to predict the potential severity of your liver disease or even if they can be used to indicate who will best respond to treatment.

The objective of this study is to identify genes that are specifically up or down regulated in chronic viral hepatitis and autoimmune liver disease and then to examine how these expression patterns relate (if at all) to clinical outcome. The expression pattern of thousands of genes should provide extremely powerful statistical tools to distinguish between different pathophysiological states. It is well recognized that neither hepatitis B or hepatitis C is directly cytotoxic, their effect appears to be mediated via an immune response. Similarly in individuals with autoimmune liver disease e.g. primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC) and autoimmune hepatitis (AIH) immune mediated mechanisms appear to be the cause of their liver disease although the eliciting antigen(s) remain unknown. The pattern of gene expression can give clues as to both the cause and the pathogenesis of disease. For instance, if a disease is caused by an infection, a specific cytokine response is observed which will be different if this infection is viral or bacterial or parasitic. It is also possible that a certain pattern of response would be elicited if the disease is caused as a response to xenobiotic. In patients with autoimmune liver disease, it is possible that an endogenous and/or an exogenous antigen initiates the disease with a subsequent autoimmune response. Finally, it is also possible that a completely unexpected series of cellular events is responsible for pathogenesis, and the microarray experiments will enable us to discover these processes. HCV - The molecular genetics of hepatitis C viral infection HCV molecular biology: HCV is a positive-stranded RNA Flaviviridae virus with a 9.6kB genome. The genome encodes a polyprotein of approximately 3000 amino acids which is subsequently cleaved by host and virus-specific proteases to yield 4 structural and 6 nonstructural polypeptides. The 4 functional proteins include a metalloproteinase (NS5A), serine protease (NS3/NS4A), RNA helicase (NS3), and a RNA-dependent RNA polymerase (NS5B). Six different genotypes of HCV have been described, with genotype 1 being the most prevalent worldwide and genotype 4 being the most prevalent in the Middle East. The various genotypes interact differently with the host immune system, though the molecular basis for this is unclear. Chronic infection by genotypes 1 and 4 is notoriously resistant to treatment with IFN, while genotypes 2 and 3 have relatively better responses. This difference may be in part due to mutations in the NS5A protein. In cellular models NS5A can interact with IFN-induced antiviral enzymes such as 2'5'OAS. Other HCV proteins likely interact with cellular proteins to alter responses to immune and IFN antiviral mechanisms: HCV core protein dampens lymphocytic Th1 responses and influences IRF, Jak/STAT and iNOS pathways. Although these results suggest how the virus may alter known cellular pathways, viral evasion of the immune system is multifactorial and clearly complex. Research Hypothesis: The processes that lead to persistence of acute HCV infection are related, at a molecular level, to those that drive HCV resistance to IFN therapy. Elucidating the specifics of the host/viral response in both of these contexts will provide novel targets for small-molecule antiviral therapy which can be applied both to acute and chronicHCV. Progress Report: CIHR 106800 We have established a large and growing tissue and RNA registry for the study of liver disease, and have performed a comprehensive gene array analysis of chronic HCV. Some of our most important findings are detailed here. HCV is a complex and progressive disease in which the interaction of the host and virus has profound effects. The combination of clinical and experimental variability can seriously confound the results of a gene expression study and mandates that a large number of patients and samples be examined. The fact that liver pathology can be similar among a number of different diseases also necessitates that many different patients and diseases are included in the analysis. Accordingly, we have analyzed large numbers of patients and controls, and have maintained careful databases of clinical details in order to perform multivariate analyses of the gene array data. At present we have used a 19K human microarray to determine hepatic gene expression in 86 HCV patients and have compared these to 24 normal, 20 HBV and 14 PBC patients. The array data are of high quality: our rates of confirmation by real-time PCR exceed 80%. We have now examined the relative contributions of age, sex, viral genotype (1 vs 2/3), degree of fibrosis and disease activity on the genes most upregulated by infection with HCV.The most important determinant of consistent gene expression profiles in HCV infected livers is viral genotype (genotype 1 vs all others). However, within Genotype 1 samples there is a clear difference in gene expression. This divergent host-virus interaction is reflected in the subsequent response to PegIFN/Rib therapy. All of the HCV liver biopsies we have analyzed to date were taken from patients who went on to treatment with PegIFN/Rib. At present, treatment has been completed in 49 of these patients. In an observation that may have significant clinical utility, we showed that the ultimate response to treatment (sustained viral response versus vs nonresponder/relapser) was reflected in the original gene expression profiles. We identified 18 genes whose expression levels distinguished responders from non-responders/relapsers in with an accuracy of \>90%. Based on this observation we filed a patent to protect the intellectual property and formed a company to commercialize the genetic test. The results have also been submitted for publication. Taken together we believe we have described a set of genes, or perhaps a biological process, that lies at the heart of the host-HCV interaction. Section I. A possible role for the UBP43, ISG15 Pathway in the HCV Interferon While our gene expression profiles have distinguished two types of chronic HCV infection - one which responds to IFN, and the other that does not - our main objective is to achieve a molecular understanding of the roles of specific genes and proteins in the host response to HCV. Though an altered gene expression level does not directly implicate the gene product in the biological pathway, there are compelling biological reasons to suggest that at least some of the genes on our list play an important role in viral-IFN responses. Several of the most up-regulated genes are interferon-sensitive (including OAS, Mx1, ISG15, VIPERIN, IFIT, and GIP2). Polymorphisms of OAS have been weakly linked to self-limited HCV infection, and polymorphisms of Mx1 have been weakly linked to response status. Hepatic mRNA levels for OAS, Mx1, and GIP2 are increased in chronic HCV but none, alone, have been linked to treatment outcome. The genes that are not directly IFN-responsive may play roles in cellular pathways important for IFN responses (PI3AP1, DUSP1), and are involved in inflammatory cell activation and maturation (LAP). ISG15 and UBP43/USP18, which are components of a newly-recognized IFN regulatory pathway, are perhaps the most interesting subset of genes that were found to be differentially expressed in our studies. Both genes are expressed more highly in nonresponder (NR) compared with responder (R) liver tissue, suggesting that they might interfere with the immune response in HCV-infected liver ISG15 is a ubiquitin-like (Ubl) protein which is thought to be important to innate immune functions and is covalently linked to proteins following interferon activation. How this link is accomplished is controversial and may involve some overlap with known E2 Ub enzymes. Interestingly, the E2 Ub enzyme UBCH8 was recently identified as an activator of ISG15,34 and is one of the genes upregulated in response to chronic HCV in our microarray analysis. The conjugation of ISG15 to its target proteins is reversed by a highly specific protease, USP18/UBP43. UBP43 belongs to a family of Ub-specific proteases and is induced by IFN, LPS and viral infection; it is degraded by the Skp2 Ub ligase. Loss of USP18 in mice leads to IFN hypersensitivity. Our data links UBP43 with HCV, and suggests that the pathway is important in a divergent host/viral response. Research Objective I. To establish the roles of the 18 genes - in particular, of ISG15 and UBP43 - associated with a divergent host/ virus interaction, using a HCV replicon model. Research Protocols for Section I: DNA microarray studies shed light on the transcription state of the cell but do not necessarily link specific gene products with the biological problem being addressed. Accordingly, it is important to test the roles of individual genes using other approaches. We will use in vitro assays of viral replication to study the roles of genes whose expression we have found to be altered in HCV infected liver. These are not straightforward experiments, since HCV is a difficult virus to study. HCV replicates only in humans and chimpanzees, and only poorly in isolated human peripheral blood cells, though these do act as an extrahepatic viral pool. Subgenomic RNA replicons can be transfected into certain cell lines and maintained, but to date only a few genotypes have been successfully translated into a replicon model (1b, 1a, and 2a). Recently, an elegant HCV model was described in Edmonton in which human hepatocytes are transplanted into SCID-beige mice. This model supports HCV infection and replication; however, since the bulk of HCV effects are due to the immune response generated by the virus, this model as it exists currently may not be ideal for studying how the virus leads to liver damage and evades antiviral immune responses in humans. We have chosen to explore the role of our genes of interest using the full-length HCV genotype 1a replicon assay, in collaboration with Dr. Charles Rice (Rockefeller University, New York). This replicon is a variant of the full-length HCV H77 strain described recently by Dr. Rice, but has higher levels of replication in cell culture than the previous version. The replicon model is ideal for rapidly testing the effects of many genes and has been used extensively to study HCV mechanisms at a cellular level. This approach will be used for all of the genes identified as of potential interest in the studies above, focusing first on the ISG15 and UBP43 genes, second on the 16 genes remaining in our response-discriminatory list, and finally on novel genes identified in the gene expression studies detailed below. Ia. Effect of siRNA knockdown and upregulation (transfection) of individual genes, including UBP43 and ISG15, on HCV replicon in Huh7.5 cells. We will use RNA interference (RNAi) to "knock down" specific genes in order to evaluate their roles on viral replication. Small double-stranded RNA molecules will be used to induce sequence-specific degradation of homologous single-stranded RNA. For most of these studies stock Huh7.5 cells will be compared to Huh7.5 cells stably expressing the full-length Genotype 1a HCV replicon developed in Dr. Charles Rice's laboratory. The methodology detailed below is routine in our laboratory and the laboratories of our collaborators. Control studies: Baseline gene and protein expression, effect of IFN, effect of replicon: The replicon system can be used not only to test the effect of a given gene on viral replication, but also for the effects on the IFN response. Previous studies have demonstrated that HCV 1b genotype RNA replication is sensitive to IFN-a treatment, even at low doses. In fact, the roles of two of the genes on our list of eighteen, MxA and GIP3/IFI6-16, have already been examined. The IFN inhibition of HCV 1b replicon RNA replication is independent of MxA, and transient transfection of GIP3 does not itself inhibit the 1b replicon but does enhance the effect of IFNa. We will now study the roles of the remaining genes in our list using the HCV genotype 1a replicon. The studies described below outline our approach for the ISG15 and UBP43 genes; however, similar strategies will be employed for the other genes of interest. Huh7.5 cells and HCV replicon cells will be incubated for 12 hours with increasing doses of IFNa2b (0, 1,10 and 100U/ml), IFNb (0,10,100 and 1000 U/ml ), or IFNg (0,10,100 and 1000 U/ml), and ISG 15, UBP43, and HCV mRNA levels determined by real-time PCR. ISG15 protein levels will be determined by western blot (mAb the kind gift of Dr. E.C. Borden, Scripps Institute); UBP43 western studies will be performed after we create antibodies or as the antibody becomes available (currently it is being developed in the lab of Dr. D.E. Zhang, Scripps Institute). RNA isolation will be performed by standard Trizol extraction, and protein extraction by lysing the cells in RIPA buffer. These studies will establish mRNA and protein levels for ISG15 and UBP43 at baseline, and in response to IFN, replicon and replicon/IFN. One important issue is confirming that the gene or protein-of-interest is expressed in these cells prior to the knock-down experiments. In preliminary studies we have determined that there is baseline expression of both ISG15 and UBP43 mRNA in both Huh7.5 and replicon cells (real-time PCR) and that IFNa strongly induces protein expression of ISG15 in Huh7.5 cells. These results argue that the ISG15 and UBP43 pathway is relevant to this model. siRNA studies: We hypothesize that elimination of UBP43 will lead to less HCV RNA replication at baseline and will magnify the effect of IFN, and that elimination of ISG15 will increase baseline HCV RNA and decrease the effect of IFN. Silencing and control siRNAs will be obtained from Ambion. These will be transfected using GenePorter according to the manufacturer's instructions; transfection conditions will be optimized using the SV40 promoter/enhancer luciferase control plasmid, pGL2-control, to ensure that increasing amounts of templates results in proportional increases in luciferase activity. Forty-eight hours after the transfection, total RNA or protein samples will be prepared as before. Real-time PCR will be performed to determine mRNA expression in the presence or absence of siRNA, and protein levels will be determined by Western blot. After these control studies have been performed, the effect of siRNA inhibition on baseline and IFN-treated HCV RNA replication will be tested. Transfection studies: We hypothesize that upregulation of UBP43 will increase in vitro HCV RNA replication at baseline and will decrease the effect of IFN, and that upregulation of ISG15 will decrease baseline HCV RNA and increase the effect of IFN. UBP43 expression will be upregulated using a UBP43 expression plasmid pcDNA6-UBP43; ISG15 expression plasmid will be obtained by subcloning ISG15 gene into pcDNA6 vector. As before, HCV RNA levels will be determined in the presence and absence of IFNa using real-time PCR. Taken together these studies will define the roles of the genes suggested to play an important role in the distinction between clinical PegIFN/Rib treatment responders and nonresponders. Ib. Mechanism of action of the ISG15/UBP43 pathway: If we determine that ISG15 and UBP43 play a role in HCV RNA replication in the genotype 1 replicon model we will examine how this effect is mediated. We will determine whether the effect of UBP43 is dependent on its protease function or on its association with another cellular protein. We will also identify which ISG15 protein targets might mediate its effect. Role of UBP43 protease activity: If elimination of UBP43 is shown to reduce viral replication or magnify the interferon response, we hypothesize that inhibition of UBP43 will represent a novel means of treatment of HCV. UBP43 would be of particular interest because it has protease activity: proteases are known and validated targets for small molecule therapies. Thus, it would be critical to establish whether the protease activity of UBP43 is important for its effect on HCV replication, and not, for instance, its interaction with another cellular protein. To address this question we will over-express UBP43 protease inactive constructs. An inactive form of UBP43 will be created by mutation of a critical cysteine residue (Cys61) into serine using site-directed mutagenesis of pBK/CMV-UBP43 plasmid as described previously. In order to characterize the protease active and inactive forms of UBP43, wild type and mutant UBP43 will be expressed as GST fusion proteins in the expression vector pGEX-4T-3 (pGEX-4T-3-UBP43), and transformed into E.Coli BL21 (DE3) with a plasmid, pET-ISG15-UBP43-H, that expresses ISG15-UBP43 fusion protein.35 ISG15 cleavage from the fusion protein will be shown by Western blot. Once we have confirmed that the protease function is eliminated by site-directed mutagenesis, wild-type and mutated UBP43 will be overexpressed in Huh7.5 cells with the HCV replicon and exposed to increasing doses of IFNa. If the protease function is critical overexpression of the inactive (in a dominant negative fashion) will increase responsiveness to IFN. Protein targets of ISG15: Previous work using thymic tissue has identified several targets for ISGylation by ISG15, including phospholipase Cg1, Jak1, and ERK1. Whether these are targets in Huh7.5 cells is not known, though ERK and JAK have been previously linked to antiviral effects in the HCV replicon model. If we find that ISG15 is implicated in HCV replication, we would seek to identify substrates for ISG15 modification. We will first perform a series of immunoprecipitation studies directed specifically at the proteins previously shown found to be modified by ISG15, and explore the differences between infected and uninfected cells. In brief, following incubation of Huh7.5 cells +/- replicon with IFNa for 12 hours, cells will be lysed in RIPA buffer and immunoprecipitation of ISG15 conjugates performed with the anti-ISG15 mAb used above. After resolution on a reducing gel, western blots will be performed to determine if the three targets above are conjugated by ISG15, using commercially-available mAbs. If any or all of the targets are conjugated, further studies will be performed to elucidate the consequences of ISGylation. To more fully explore which proteins are modified with ISG15 in cell culture, we will undertake a more comprehensive screen using mass-spectroscopy - applying a strategy previously used to identify targets of the ubiquitin-like SUMO protein. TAP-tagging approaches coupled with MALDI-TOF mass spectrometry will be used. We will transfect cells with TAP-tagged or SPA-tagged ISG15 and purify the protein and protein complexes from cells treated or untreated with IFN. Using established procedures, we will then purify and identify the substrates using gel electrophoresis and mass spectrometry. Gel slices containing the proteins will be reductively alkylated and subjected to trypsinolysis. The peptides will be purified and analyzed by MALDI-TOF mass spectrometry using a cyano-4-hydroxycin-namic acid matrix on a Voyager DE-STR instrument (Applied Biosystems). Identification of the proteins using these mass fingerprinting data will be carried out using the ProFound software . Since MS cannot be used to quantify protein amounts, we will identify proteins of interest by comparing lists of proteins ISGylated under the various conditions and then perform Western blot studies to compare protein amounts. In doing these experiments we will use the MS equipment and expertise readily available at the Best Institute. Conclusions I: Taken together these studies will clearly define the roles of the ISG15 and UBP43 pathway in the HCV replicon response to IFN, and will suggest how the effect is mediated. Importantly, the methodologies described above, though detailed for ISG15 and UBP43, are generic and can be applied to other candidate genes of interest. Section II: Gene Expression Profiling of Acute Hepatitis C Viral Infection: There are similarities in the pathways that we have identified as important in the chronic disease with those that may be important in acute infection. For example, polymorphisms of 2'5'OAS - a gene found on our list - alter the risk of progressing from acute to chronic infection.20 Therefore we hypothesize that this same set of genes is likely to play an important role in determining failure to clear acute infection. We will test this hypothesis using microarray analysis. We plan to examine the role of these and other genes in acute HCV infection in both the systemic and local host/virus interaction. Acutely-infected liver tissue is difficult to obtain because most acute HCV infections are not identified, and because it is inappropriate to perform liver biopsies in these cases. Thus we will use two different approaches. First, we will study the gene expression patterns in peripheral blood mononuclear cells (PBMCs) collected from Egyptian health care workers with acute HCV (provided by Dr. Sanaa Kamal) and compare these to PBMCs from patients with genotype-matched chronic HCV and to healthy volunteers. We will develop predictive models to test which gene subsets are most associated with clinical outcomes such as the establishment of chronic infection. Second, we will use the recurrence of HCV post transplantation as a model of acute hepatitis in vivo. We will take biopsies of the donor liver PRIOR to transplantation and then follow gene expression levels over time as the virus infects the graft. After transplantation the HCV-naïve liver rapidly and universally becomes infected with HCV, though only 70% of patients develop histological evidence of recurrence on liver biopsies taken 3 and/or 6 months post-transplant. It is these patients who we currently treat for recurrent HCV using PegIFN/Rib. As before, we will develop predictive models to associate gene subsets with clinical outcomes. We have achieved more than a 60% response rate by encouraging treatment adherence despite the even more grueling side effects after transplantation. Thus, as for the treatment of HCV prior to transplantation a significant fraction of patients do not respond to therapy and are needlessly exposed to morbid side effects: the key clinical outcome we will consider is response to therapy with PegIFN/Rib. Another outcome of interest is the rapid development of cirrhosis within 5 years in 25-30% of patients, and the development of an aggressive fibrosing cholestatic state in 7-9% within 2 years of transplant. Together these data will point to the genes important for the immunological response that results in disease persistence or elimination (acute HCV in health care workers) and progression of liver disease (post-transplantation patients). Highly up- and down-regulated genes will be further studied in the HCV replicon model. Research Objective 2. To determine the gene expression profiles which are most highly associated with clinical outcomes in acute hepatitis, both in circulating immune cells and in the liver. Research Protocols for Section II: We propose three major analyses. First, we will perform gene expression profiling in PBMC preparations provided by Dr. Kamal from patients with acute HCV. These PBMC preparations are being performed as part of ongoing studies headed by Dr. Kamal. Second, we will use microarray analysis to describe the hepatic genes which are altered by HCV infection in the post-transplant liver graft in response to infection by HCV. For this we will use portions of liver biopsies already being taken as part of routine post-transplant clinical care. Third, we will prepare PBMCs from patients with chronic HCV and from these determine gene expression profiles. Not only are these a critical comparison to acute HCV, but they will also act as the basis for predicting outcomes of treatment in chronic HCV. IIa. Patient recruitment, study population and collection of samples and data: In all cases, patients will be identified by clinical staff as potential candidates for the study and approached by a study nurse. Consent will be obtained for the research use of tissue and blood samples, and for the collection of patient clinical data. Acute HCV: systemic PBMC response Dr. Kamal follows a population of Egyptian health care workers who become infected acutely with HCV, and she routinely isolates PBMCs from them. This clinical population is composed primarily of Genotype 1 and Genotype 4 infected individuals, and is thoroughly described for patient demographics, RNA titer at the time of blood collection, and progression to chronic infection. Given the relative rarity of Genotype 4 in our local population our initial analysis will focus on Genotype 1 acute HCV, but we intend to study genotype 4 as well, since this genotype has great relevance to HCV worldwide. Based on our previous work we anticipate needing between 10-20 patients in a given group in order to clearly define consistent gene expression changes across a spectrum of clinical variability. We expect to collect and study PBMC samples from at least 20 acute genotype 1 patients per year. Samples from 20 healthy control volunteers will also be obtained over the 3 year study period (recruited by poster). Acute HCV: Hepatic response to HCV recurrence post-transplantation Our center performs 30-40 liver transplants each year in persons infected with HCV. All clinical data in our liver transplant population is entered into an Oracle-based database. Liver biopsies are routinely taken from the HCV-naïve graft prior to transplantation and at 12, 24, and 52 weeks post-transplantation. Based on previous experience we expect an accrual rate of at least 80%; thus, we will be able to collect data and specimens from roughly 24-32 patients per year. Biopsies of normal liver are taken as part of an ongoing study in our living donor population (PI: I. McGilvray). 30-40 such biopsies are taken every year and represent a unique source of control liver tissue. Chronic HCV: systemic PBMC response Dr. Heathcote follows a large population of patients with chronic HCV. Blood samples will be taken from patients in whom treatment with PegIFN/Rib is being considered and PBMCs will be isolated. Based on our experience with the liver biopsy studies outlined in the Progress Report above, we expect to obtain PBMC preparations from at least 50 patients each year. Roughly 2/3 of these are Genotype 1 patients. IIb. Data Analysis: Acute HCV: Gene expression changes in response to acute HCV in circulating PBMCs Gene expression profiles from PBMCs from patients with acute Genotype 1 HCV will be determined and compared to those from normal healthy controls and chronic genotype 1 HCV. Genes will be identified by statistical and fold differences between groups (p value of \<0.01, fold change of at least 1.5). This comparison will define which gene expression alterations are specific to acute HCV infection. In order to associate individual genes with the progression to chronic disease, subgroup analyses will be done in which gene expression in acutely infected individuals is compared between those who went on to chronic infection and those who spontaneously eliminated the virus. After real-time PCR confirmation of these genes we will use a number of unsupervised (hierarchical clustering, principal components) and supervised (nearest neighbours, linear discriminants) analyses to develop predictive gene sets which accurately classify patients who go on to develop chronic infection. Taken together these results will identify those genes whose expression is altered in the systemic immunologic response to acute HCV and how these relate to the establishment of chronic disease. Genes which are of interest as possible therapeutic targets will be further studied using the HCV replicon model, as described earlier. By the end of the second year of the study we expect to have analyzed 40 acute genotype 1 patients, and in the third year we will analyze 20 genotype 4 patients. Acute HCV: local hepatic gene expression in post-transplantation HCV infection Gene expression profiles will be determined from liver biopsies at organ retrieval, prior to infection by HCV, and at 3, 6 and 12 months after transplantation. Gene changes can therefore be followed in the same patient and between patients; genes levels will be compared to normal liver tissue and to baseline gene expression in HCV-naïve grafts, using statistical methods developed in our lab and at the EBI. Currently 70% of the patients transplanted with HCV in our center are treated for recurrent HCV post-transplantation after evidence of recurrent HCV on liver biopsy, and in these patients quantitative HCV titers are done at 3 or 6 months post-transplantation. Roughly 80% of our patients complete therapy over a 12 month course. Thus, by the end of our 3 year study we will know the response to treatment of between 28-36 patients (we expect a 60% response rate, thus roughly 17-22 responders and 11-14 nonresponders), and eventually will have data on the 42-54 patients expected to complete therapy. With this sample size we can determine which gene subsets are predictive of response to treatment (using the methods noted above). Equally we will be able to relate gene expression to HCV RNA titer in 50-60 patients. By the end of the three year study we will have clinical and genomic data on 72 to 90 patients. Ultimately we will relate gene expression to the development of recurrent cirrhosis, and we anticipate that roughly 18-30 patients will have gone on to this state within 5 years of transplantation. In order to control for the effects of the many confounding factors that could influence hepatic gene expression post-transplantation (eg. acute rejection, immunosuppression) we will perform subgroup analyses using our extensive clinical database. We will compare HCV treatment responders to nonresponders, rejection to no rejection, and different immunosuppressive regimens, in addition to comparing gene expression profiles from acute hepatic HCV to those already determined for chronic HCV. These analyses will allow us to distinguish and describe the effects of acute HCV infection on hepatic gene expression. Chronic HCV: circulating PBMC response and prediction of treatment outcome: Over the first 2 years of our study we will use gene expression profiling to study PBMCs from roughly 30 patients with chronic genotype 1 HCV. This data will serve as a crucial control for our studies of acute HCV in PBMCs. However, all of these samples are taken from patients who will go on to treatment with PegIFN/Rib. Therefore we will use the gene expression data from these studies not only for comparison with the acutely infected PBMCs, but also to determine which gene subsets are predictive of treatment outcome. These results will form the basis for a noninvasive prognostic test. Conclusions II: These studies will provide a comprehensive description of the gene expression changes that accompany acute HCV infection at both the systemic and the local level. They will clearly identify genes which are important to the progression to chronic disease and to adverse clinical outcomes, thereby suggesting novel prognostic and therapeutic targets.