Portal de Periódicos da CAPES

Gene Expression Patterns That Correlate With Hepatitis C and Early Progression to Fibrosis in Liver Transplant Recipients

2006; Elsevier BV; Volume: 130; Issue: 1 Linguagem: Inglês

10.1053/j.gastro.2005.08.015

1528-0012

M. W. Smith, Kathie–Anne Walters, Marcus J. Korth, Matthew Fitzgibbon, Sean Proll, Jill C. Thompson, Matthew M. Yeh, Margaret C. Shuhart, Jeffrey C. Furlong, Paula P. Cox, David L. Thomas, John D. Phillips, James P. Kushner, Nelson Fausto, Robert L. Carithers, Michael G. Katze,

Hepatitis B Virus Studies

Background & Aims: Liver transplant recipients infected with hepatitis C virus (HCV) develop recurrent hepatitis soon after transplantation and, in some cases, progress to fibrosis within the first 2 years. Our goals were to identify molecular processes influencing the liver disease progression and to find potential gene markers of early fibrosis.Methods: We performed gene expression profiling on serial liver biopsy specimens obtained from 13 (11 infected and 2 uninfected) transplant recipients within the first year after transplantation at 0, 3, 6, and 12 months. The data were compared with clinical observations and with a gene expression database obtained for 55 nontransplant HCV-infected and uninfected liver samples.Results: We identified several specific gene expression patterns. The first pattern was unique for the transplant recipients regardless of their infection status. The corresponding genes encoded stress response proteins and blood proteins involved in coagulation that were differentially expressed in response to posttransplantation graft recovery. The second pattern was specific to HCV-infected samples and included up-regulation of genes encoding components of the interferon-mediated antiviral response and immune system (antigen presentation, cytotoxic response). This up-regulation pattern was absent or suppressed in the patients who developed early fibrosis, indicating that the disease progression might result from an impaired liver response to infection. Finally, we identified gene expression patterns that were specific for 12-month biopsy specimens in all 4 HCV-infected patients who developed early fibrosis.Conclusions: The identified gene expression patterns may prove useful for diagnostic and prognostic applications in HCV-infected patients, including predicting early progression to fibrosis. Background & Aims: Liver transplant recipients infected with hepatitis C virus (HCV) develop recurrent hepatitis soon after transplantation and, in some cases, progress to fibrosis within the first 2 years. Our goals were to identify molecular processes influencing the liver disease progression and to find potential gene markers of early fibrosis. Methods: We performed gene expression profiling on serial liver biopsy specimens obtained from 13 (11 infected and 2 uninfected) transplant recipients within the first year after transplantation at 0, 3, 6, and 12 months. The data were compared with clinical observations and with a gene expression database obtained for 55 nontransplant HCV-infected and uninfected liver samples. Results: We identified several specific gene expression patterns. The first pattern was unique for the transplant recipients regardless of their infection status. The corresponding genes encoded stress response proteins and blood proteins involved in coagulation that were differentially expressed in response to posttransplantation graft recovery. The second pattern was specific to HCV-infected samples and included up-regulation of genes encoding components of the interferon-mediated antiviral response and immune system (antigen presentation, cytotoxic response). This up-regulation pattern was absent or suppressed in the patients who developed early fibrosis, indicating that the disease progression might result from an impaired liver response to infection. Finally, we identified gene expression patterns that were specific for 12-month biopsy specimens in all 4 HCV-infected patients who developed early fibrosis. Conclusions: The identified gene expression patterns may prove useful for diagnostic and prognostic applications in HCV-infected patients, including predicting early progression to fibrosis. The prolonged time frame of several decades associated with progression to fibrosis in hepatitis C virus (HCV)-infected patients greatly complicates efforts to investigate the molecular basis of disease progression. However, 8%–30% of patients who have undergone liver transplantation due to HCV-associated liver disease develop rapidly progressive fibrosis and subsequent cirrhosis within 5–7 years after transplantation.1Prieto M. Berenguer M. Rayon J.M. Cordoba J. Arguello L. Carrasco D. Garcia-Herola A. Olaso V. De Juan M. Gobernado M. Mir J. Berenguer J. High incidence of allograft cirrhosis in hepatitis C virus genotype 1b infection following transplantation relationship with rejection episodes.Hepatology. 1999; 29: 250-256Crossref PubMed Scopus (487) Google Scholar, 2Berenguer M. Lopez-Labrador F.X. Wright T.L. Hepatitis C and liver transplantation.J Hepatol. 2001; 35: 666-678Abstract Full Text Full Text PDF PubMed Scopus (262) Google Scholar, 3Larson A.M. Carithers R.L. Hepatitis C in clinical practice.J Intern Med. 2001; 249: 111-120Crossref PubMed Scopus (14) Google Scholar, 4McCaughan G.W. Zekry A. Pathogenesis of hepatitis C virus recurrence in the liver allograft.Liver Transpl. 2002; 8: S7-S13Crossref PubMed Scopus (58) Google Scholar, 5Rosen H.R. Hepatitis C virus in the human liver transplantation model.Clin Liver Dis. 2003; 7: 107-125Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar Although a number of studies have attempted to identify factors influencing the incidence and severity of recurrent liver disease following transplantation, no single clinical variable appears to be an accurate predictor.1Prieto M. Berenguer M. Rayon J.M. Cordoba J. Arguello L. Carrasco D. Garcia-Herola A. Olaso V. De Juan M. Gobernado M. Mir J. Berenguer J. High incidence of allograft cirrhosis in hepatitis C virus genotype 1b infection following transplantation relationship with rejection episodes.Hepatology. 1999; 29: 250-256Crossref PubMed Scopus (487) Google Scholar, 2Berenguer M. Lopez-Labrador F.X. Wright T.L. Hepatitis C and liver transplantation.J Hepatol. 2001; 35: 666-678Abstract Full Text Full Text PDF PubMed Scopus (262) Google Scholar, 6Shuhart M.C. Bronner M.P. Gretch D.R. Thomassen L.V. Wartelle C.F. Tateyama H. Emerson S.S. Perkins J.D. Carithers R.L. Histological and clinical outcome after liver transplantation for hepatitis C.Hepatology. 1997; 26: 1646-1652Crossref PubMed Scopus (103) Google Scholar, 7Firpi R.J. Abdelmalek M.F. Soldevila-Pico C. Cabrera R. Shuster J.J. Theriaque D. Reed A.I. Hemming A.W. Liu C. Crawford J.M. Nelson D.R. One-year protocol liver biopsy can stratify fibrosis progression in liver transplant recipients with recurrent hepatitis C infection.Liver Transpl. 2004; 10: 1240-1247Crossref PubMed Scopus (149) Google Scholar Thus, at the University of Washington, the individual transplant recipients are monitored by performing consecutive core needle liver biopsies.3Larson A.M. Carithers R.L. Hepatitis C in clinical practice.J Intern Med. 2001; 249: 111-120Crossref PubMed Scopus (14) Google Scholar In addition to valuable clinical information, these biopsy specimens provide a unique resource of sequentially obtained graft material for studies of molecular mechanisms of HCV pathogenesis. Global gene expression profiling provides a way to obtain detailed pictures of molecular events occurring in clinical liver samples.8Honda M. Kaneko S. Kawai H. Shirota Y. Kobayashi K. Differential gene expression between chronic hepatitis B and C hepatic lesion.Gastroenterology. 2001; 120: 955-966Abstract Full Text Full Text PDF PubMed Google Scholar, 9Shackel N.A. McGuinness P.H. Abbott C.A. Gorrell M.D. McCaughan G.W. Identification of novel molecules and pathogenic pathways in primary biliary cirrhosis cDNA array analysis of intrahepatic differential gene expression.Gut. 2001; 49: 565-576Crossref PubMed Scopus (136) Google Scholar, 10Shackel N.A. Gorrell M.D. McCaughan G.W. Gene array analysis and the liver.Hepatology. 2002; 36: 1313-1325Crossref PubMed Google Scholar, 11Smith M.W. Yue Z.N. Geiss G.K. Sadovnikova N.Y. Carter V.S. Boix L. Lazaro C.A. Rosenberg G.B. Bumgarner R.E. Fausto N. Bruix J. Katze M.G. Identification of novel tumor markers in hepatitis C virus-associated hepatocellular carcinoma.Cancer Res. 2003; 63: 859-864PubMed Google Scholar, 12Smith M.W. Yue Z.N. Korth M.J. Do H.A. Boix L. Fausto N. Bruix J. Carithers Jr, R.L. Katze M.G. Hepatitis C virus and liver disease global transcriptional profiling and identification of potential markers.Hepatology. 2003; 38: 1458-1467Crossref PubMed Google Scholar We used this approach to analyze serial liver biopsy specimens obtained from 13 transplant recipients. This allowed us to correlate the dynamics of gene expression changes with clinical findings. We also analyzed our data in the context of a gene expression database obtained using 55 nontransplant and uninfected liver specimens. Thus, we evaluated whether specific gene expression patterns are associated with the response to HCV infection or the onset and severity of liver disease. We identified several sets of differentially expressed genes that may serve as markers for early progression to fibrosis and that may provide new insights into the molecular mechanisms underlying liver disease progression. Core needle liver biopsy specimens were obtained with informed consent as approved by the Human Subjects Review Committee from patients who underwent liver transplantation at the University of Washington between July 2000 and October 2003. In total, 13 patients were included in this study based on availability of the baseline biopsy specimen obtained at the time of transplantation and biopsy specimens obtained at 3, 6, or 12 months after transplantation (Table 1). Two uninfected patients (521 and 522) underwent transplantation due to cryptogenic cirrhosis and alcoholic cirrhosis with hepatocellular carcinoma, respectively. Eleven patients underwent transplantation due to HCV-associated cirrhosis, and all developed recurrent HCV infection (none underwent therapy for HCV after transplantation). All patients received different combinations of immune-suppressing drugs (Table 1); however, their clinical outcome did not correlate with the drug combinations. Inflammation (grade) and fibrosis (stage) were scored using the Batts–Ludwig scoring system (grade 0–4, stage 0–4).13Batts K.P. Ludwig J. Chronic hepatitis. An update on terminology and reporting.Am J Surg Pathol. 1995; 19: 1409-1417Crossref PubMed Scopus (940) Google Scholar Three of 11 patients showed progression to fibrosis stage 1 (504) or 2 (510 and 512) within the first year and progressed to stage 3 during the second and third years after transplantation. Patient 509 developed fibrosis stage 1 at 15 months and progressed to cirrhosis during the second year. All other patients remained at stage 0 for the duration of the follow-up.Table 1Clinical Information for the Transplant Recipients Included in This StudyPatient-monthsAge (y)EthnicitySexHCV genotypePast alcoholicPast drug useGradeFibrosis stageAlanine aminotransferase (U/L)Immune suppression501-0649WhiteMale1aYesYes0064T, A501-120032T, P504-0148WhiteMale3aYesYes0041T, P504-121245T, A, P507-0349Native AmericanMale1aYesYes00304T, P509-0353WhiteFemale2bYesNo0030T, A, P509-0600257T, A, P509-120027T, A, P510-0350WhiteMale2bYesNo20185C, P510-0621341C, P510-1222274C, P511-0353WhiteFemale2aYesNo30740C, A, P511-0630419C, A511-1220227C, A, P512-0353AsianFemale2aYesNo00263T, P512-0630270T, P512-1222128T, P513-0352WhiteMale3aYesYes0019T, S518-0347AsianMale2aYesNo00260C, P518-0800234C, P518-1200165C, P519-0348WhiteMale1aYesYes0049R519-080064R519-1200142T, R520-0354WhiteFemale1aYesYes0040C, A, P520-060072T, A, P520-1200108T, A, P521-03aDiagnosis of cryptogenic cirrhosis.19BlackFemale—NoNo0021T, A522-03bDiagnosis of alcoholic cirrhosis and hepatocellular carcinoma.49WhiteFemale—YesYes0070T522-060025TNOTE. Inflammation (grade 0–4) and fibrosis (stage 0–4) were scored using the Batts–Ludwig system.Four patients who developed fibrosis within the first 1.5 years after transplantation are shown in bold.A, azathioprine; C, cyclosporine; P, prednisone; R, rapamune; S, sirolimus; T, tacrolimus.a Diagnosis of cryptogenic cirrhosis.b Diagnosis of alcoholic cirrhosis and hepatocellular carcinoma. Open table in a new tab NOTE. Inflammation (grade 0–4) and fibrosis (stage 0–4) were scored using the Batts–Ludwig system. Four patients who developed fibrosis within the first 1.5 years after transplantation are shown in bold. A, azathioprine; C, cyclosporine; P, prednisone; R, rapamune; S, sirolimus; T, tacrolimus. Core needle liver biopsy specimens of nontransplant livers were collected from infected and uninfected patients before medical treatment at the University of Washington, University of Utah, and John Hopkins University according to protocols approved by human subjects review committees.14Rai R. Wilson L.E. Astemborski J. Anania F. Torbenson M. Spoler C. Vlahov D. Strathdee S.A. Boitnott J. Nelson K.E. Thomas D.L. Severity and correlates of liver disease in hepatitis C virus-infected injection drug users.Hepatology. 2002; 35: 1247-1255Crossref PubMed Scopus (69) Google Scholar Surgical samples were collected from 8 livers with HCV-associated cirrhosis and 8 livers with alcoholic liver cirrhosis.11Smith M.W. Yue Z.N. Geiss G.K. Sadovnikova N.Y. Carter V.S. Boix L. Lazaro C.A. Rosenberg G.B. Bumgarner R.E. Fausto N. Bruix J. Katze M.G. Identification of novel tumor markers in hepatitis C virus-associated hepatocellular carcinoma.Cancer Res. 2003; 63: 859-864PubMed Google Scholar, 12Smith M.W. Yue Z.N. Korth M.J. Do H.A. Boix L. Fausto N. Bruix J. Carithers Jr, R.L. Katze M.G. Hepatitis C virus and liver disease global transcriptional profiling and identification of potential markers.Hepatology. 2003; 38: 1458-1467Crossref PubMed Google Scholar A standard normal liver reference was pooled from 8 surgical samples of normal, uninfected human liver tissue and used for all experiments with nontransplant samples. Frozen liver tissues were disrupted in TRIzol reagent (Invitrogen, Carlsbad, CA) using a Polytron homogenizer (PowerGene 700; Fisher Scientific, Hampton, NH). Total RNA was isolated according to the TRIzol protocol with the yield of 1–5 μg from a liver biopsy specimen of 2 mg. All total RNA samples were double amplified with the RiboAmp RNA Amplification kit (Arcturus, Mountain View, CA). The quality of amplified RNA was evaluated by capillary electrophoresis using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). Microarray format, protocols for probe labeling, and array hybridization are described at http://expression.microslu.washington.edu. This Web site is also used to publish all primary data.15Brazma A. Hingamp P. Quackenbush J. Sherlock G. Spellman P. Stoeckert C. Aach J. Ansorge W. Ball C.A. Causton H.C. Gaasterland T. Glenisson P. Holstege F.C.P. Kim I.F. Markowitz V. Matese J.C. Parkinson H. Robinson A. Sarkans U. Schulze-Kremer S. Stewart J. Taylor R. Vilo J. Vingron M. Minimum information about a microarray experiment (MIAME) toward standards for microarray data.Nat Genet. 2001; 29: 365-371Crossref PubMed Scopus (3324) Google Scholar A typical experiment was a comparison between 2 messenger RNA samples, one of which was a baseline biopsy specimen obtained at the time of transplantation and another of which was a time point biopsy specimen from the same patient at 3, 6, or 12 months after transplantation. Each experiment was performed with 4 replicate human complementary DNA expression arrays carrying 13,026 unique complementary DNA clones (Agilent Technologies) using the dye label reverse technique.11Smith M.W. Yue Z.N. Geiss G.K. Sadovnikova N.Y. Carter V.S. Boix L. Lazaro C.A. Rosenberg G.B. Bumgarner R.E. Fausto N. Bruix J. Katze M.G. Identification of novel tumor markers in hepatitis C virus-associated hepatocellular carcinoma.Cancer Res. 2003; 63: 859-864PubMed Google Scholar, 12Smith M.W. Yue Z.N. Korth M.J. Do H.A. Boix L. Fausto N. Bruix J. Carithers Jr, R.L. Katze M.G. Hepatitis C virus and liver disease global transcriptional profiling and identification of potential markers.Hepatology. 2003; 38: 1458-1467Crossref PubMed Google Scholar, 16Geiss G.K. An M.C. Bumgarner R.E. Hammersmark E. Cunningham D. Katze M.G. Global impact of influenza virus on cellular pathways is mediated by both replication-dependent and -independent events.J Virol. 2001; 75: 4321-4331Crossref PubMed Scopus (107) Google Scholar The data were analyzed in our custom database, Expression Array Manager, Resolver 4.0 (Rosetta Biosoftware, Seattle, WA), and Decision Suite for Functional Genomics 7.1.1 (Spotfire, Somerville, MA). Data normalization was performed using a custom error model that was developed by Rosetta Biosoftware (as described in our Web site). Selection of genes for data analysis was based on a >95% probability of being differentially expressed (P ≤ .05) and a fold change of ≥2. The resultant false-positive discovery rate was estimated in previous work as <1% by using semiquantitative reverse-transcriptase polymerase chain reaction (PCR).11Smith M.W. Yue Z.N. Geiss G.K. Sadovnikova N.Y. Carter V.S. Boix L. Lazaro C.A. Rosenberg G.B. Bumgarner R.E. Fausto N. Bruix J. Katze M.G. Identification of novel tumor markers in hepatitis C virus-associated hepatocellular carcinoma.Cancer Res. 2003; 63: 859-864PubMed Google Scholar, 12Smith M.W. Yue Z.N. Korth M.J. Do H.A. Boix L. Fausto N. Bruix J. Carithers Jr, R.L. Katze M.G. Hepatitis C virus and liver disease global transcriptional profiling and identification of potential markers.Hepatology. 2003; 38: 1458-1467Crossref PubMed Google Scholar First, the array data were validated based on genes that were represented on the arrays with at least 2 independent complementary DNA spots corresponding to different PCR products. Among these genes, 236 were differentially expressed in at least 6 of 30 experiments, and in 83% of the cases the duplicate spots showed relatively high correlation coefficients exceeding 0.85. Second, the array data were directly validated for 6 genes using real-time PCR. Total RNA samples from a subset of 4 patients were treated with deoxyribonuclease reagents (Ambion, Austin, TX). Reactions were performed in quadruplicate on the ABI 7500 Real Time PCR System (Applied Biosystems, Foster City, CA) using TaqMan chemistry with primer and probe sets from the Assays-on-Demand list (Applied Biosystems) and 2 controls: glyceraldehyde-3-phosphate dehydrogenase and 18S ribosomal RNA. On average, the ratios calculated from real-time PCR data exceeded the ones obtained using expression arrays. However, the direction of changes was the same for both types of data, providing that the array data met our selection criteria of P ≤ .05 and a fold change of ≥2. In total, 30 experiments were performed to compare baseline biopsy specimens obtained at the time of transplantation with the consecutive biopsy specimens obtained at 3, 6, or 12 months posttransplantation from the same patients (Table 1). A hierarchical 2-dimensional clustering of these 7586 genes differentially expressed (more than 2-fold with P value ≤ .05) in at least 2 experiments is shown in Figure 1A. Experiments performed using material from the same patient tended to group together. A notable exception was observed for the 12-month biopsy specimens from patients with early progression to fibrosis (504, 509, 510, and 512), which had much less similarity with the earlier time points from the same patients and showed highly elevated numbers of gene expression changes (Figure 1B). To associate gene expression patterns with common pathologic processes occurring in the patients, we selected a set of 3536 genes that were differentially expressed in at least 6 of 30 experiments with liver transplant samples. The obtained gene set was analyzed in the context of array experiments performed on nontransplant liver specimens, including (1) HCV-infected with fibrosis (n =34); (2) uninfected with porphyria cutanea tarda (n = 5); and (3) cirrhosis, both alcoholic (n = 8) and HCV-associated (n = 8). Using this approach, we selected 2 gene sets: one with expression patterns unique for the transplant recipients, and one of 169 genes that were specifically up-regulated in the majority of HCV-infected samples in both transplant and nontransplant settings. Our expression array data showed dramatic posttransplant stress responses in the baseline biopsy specimens obtained from the grafts at the time of transplantation (Table 2). These responses disappeared at 3 months (Table 2) in the process of posttransplant graft recovery. Restoration of normal liver function was also evident from up-regulation of genes encoding liver enzymes and blood proteins involved in coagulation and fibrinolysis (Figure 2A). Our array data were consistent with the clinical observations on normalization of serum albumin levels and the prothrombin time at 3 months posttransplantation (Figure 2B).Table 2Expression Ratios of 15 Stress-Related Genes in the Pool of Baseline Transplant Biopsy Specimens Compared With the Normal Liver Reference Pool and Their Average Ratios Calculated Using 11 Experiments With the Individual 3-Month Time Point Transplant Biopsy Specimens Compared With the Corresponding Baseline Biopsy SpecimensGene nameGene descriptionPool of baseline transplant biopsy specimens vs normal liver reference poolAverage of 11 3-month biopsy specimens vs baseline biopsy specimensSAA1Serum amyloid A150.01−9.72SAA1Serum amyloid A149.20−10.47SAA2.2Serum amyloid A246.59−8.97HSPA7Heat shock 70-kilodalton protein 725.01−2.27CRPC-reactive protein, pentraxin-related12.27−15.69CRPC-reactive protein, pentraxin-related12.23−18.93HSPA1AHeat shock 70-kilodalton protein 1A11.05−4.99HSPF1Heat shock protein 407.92−2.78SERPINH1Heat shock 47-kilodalton protein, member 16.15−1.84HSPH1Heat shock 105-kilodalton/110-kilodalton protein 14.71−2.78HSPCBHeat shock 90-kilodalton protein 1, β4.32−1.80HSPCAHeat shock 90-kilodalton protein 1, α3.61−2.16HSP89AHeat shock 89-α protein2.85−2.33HIF1AHypoxia-inducible factor 1, α2.58−3.07HIF1AHypoxia-inducible factor 1, α2.53−2.87HSPA5Heat shock 70-kilodalton protein 5, 78 kilodalton2.22−1.96HSPD1Heat shock 60-kilodalton protein 12.17−2.15CSDACold shock domain protein A2.02−2.58 Open table in a new tab The set of 169 genes up-regulated in correlation with HCV infection in both transplant and nontransplant settings contained 20 genes related to the interferon (IFN) system. As shown in Figure 3A, these genes were up-regulated only in HCV-infected patients. Several of them encode well-known markers of a type I IFN response. Expression array data for 2 of these genes, MX1 and STAT1, were successfully validated by using real-time PCR (Figure 3B). Up-regulation of the IFN-related genes was observed in 7 of 11 patients at all posttransplant time points (Figure 3A). However, this up-regulation was absent in patients 509 and 512 and suppressed in the 12-month biopsy specimens from patients 504 and 510 (Figure 3A), despite occurrence of high viremia. Three of these patients (504, 510, and 512) developed significant fibrosis at the end of the first year after transplantation (Table 1). Patient 509 did not have histologic signs of fibrosis at 12 months but developed fibrosis by 15 months and rapidly progressed to cirrhosis at 24 months. Thus, the lack of up-regulation of IFN-related gene expression may be related to an impaired liver response to infection and to early liver disease progression. Furthermore, as illustrated by patient 509, this expression pattern may appear before histologic evidence of fibrosis. The set of genes specifically up-regulated in association with HCV infection contained more than 40 genes associated with different immune system cells (Figure 4A). However, despite the universal establishment of HCV infection after transplantation, the expression patterns of these genes were patient specific. Six of the infected transplant recipients showed a clear and coordinated up-regulation of genes involved in the functioning of professional antigen-presenting cells (brown in Figure 4A) and 2 cytokine genes, interleukin-6 and CCL5/RANTES (red in Figure 4A), that may be associated with dendritic cell activation.17Speirs K. Lieberman L. Caamano J. Hunter C.A. Scott P. Cutting edge NF-kappa B2 is a negative regulator of dendritic cell function.J Immunol. 2004; 172: 752-756Crossref PubMed Scopus (45) Google Scholar, 18Leroy V. Vigan I. Mosnier J.F. Dufeu-Duchesne T. Pernollet M. Zarski J.P. Marche P.N. Jouvin-Marche E. Phenotypic and functional characterization of intrahepatic T lymphocytes during chronic hepatitis C.Hepatology. 2003; 38: 829-841Crossref PubMed Google Scholar The same group of 6 patients also showed an increase in the expression of 5 genes encoding components of the ubiquitin-mediated proteolysis that may be involved in antigen presentation (Figure 4B). Up-regulation of genes related to antigen presentation correlated with up-regulation of genes associated with T-cell recruitment and activation (green in Figure 4A) and with cytotoxic T-lymphocyte response and natural killer cells (blue in Figure 4A). The described gene markers of immune response were either unchanged or down-regulated in the remaining 5 HCV-infected transplant recipients (509, 510, 511, and 512; and 12 months in 504). This may indicate a reduced antigen presentation and low efficiency of cytotoxic immune responses. Four of these patients also showed impaired IFN-mediated gene expression (Figure 3A) and progressed to fibrosis within the first 12–15 months after transplantation (Table 1). Development of fibrosis in these patients was also inversely correlated with the up-regulation of a specific transcription factor, c-MAF, indicating an impaired development of the humoral immune response19Hwang E.S. White I.A. Ho I.C. An IL-4-independent and CD25-mediated function of c-maf in promoting the production of Th2 cytokines.Proc Natl Acad Sci U S A. 2002; 99: 13026-13030Crossref PubMed Scopus (52) Google Scholar (Figure 4A). To identify gene expression patterns characteristic of early progression to fibrosis, we identified 21 genes specifically up-regulated in the 12-month biopsy specimens from patients 504, 509, 510, and 512. Among these genes, 15 encoded potential markers of myofibroblasts (MFs) and MF-like cells (Figure 5A). Liver stress, injury, and development of fibrosis can cause an increase in the MF numbers due to the activation of hepatic stellate cells (HSCs) and their conversion to the contractile phenotype.20Friedman S.L. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury.J Biol Chem. 2000; 275: 2247-2250Crossref PubMed Scopus (1892) Google Scholar However, the immunochemical staining for α–smooth muscle actin did not show a significant increase in activated HSCs in our samples (data not shown). Thus, we further used expression array data to analyze genes encoding retinoid-related proteins, because activated HSCs lose retinoids that are stored in the quiescent state.20Friedman S.L. Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury.J Biol Chem. 2000; 275: 2247-2250Crossref PubMed Scopus (1892) Google Scholar Nine of these genes, and the apolipoprotein E precursor gene (which is also expressed by HSCs), were up-regulated in the majority of specimens except for the 12-month biopsy specimens from the 4 patients who developed fibrosis within the first 12–15 months after transplantation (Figure 5B). Thus, early progression to fibrosis may be associated with a decrease in the pools of quiescent HSCs, due to their activation and transdifferentiation, and with increased numbers of resulting MF-like cells. The series of liver transplant biopsy specimens allowed us to analyze gene expression changes over time, thus enabling us to distinguish progression to fibrosis from other processes caused by recurrent hepatitis C. Our data showed that each of 4 individuals who had clinical features of fibrosis at 12–15 months lacked at least a part of the up-regulation patterns that were related to IFN-mediated innate response (Figure 3), antigen presentation, and immune response (Figure 4). This lack of up-regulation was not caused by sampling errors or focal development of liver disease, because all 12-month biopsy specimens had very high numbers of up-regulated genes (Figure 1) that accompanied multiple pathologic changes observed in the collected tissue. All 4 individuals who showed the described impairment of hepatic responses to HCV infection progressed to fibrosis stage 3 and cirrhosis within 2–3 years posttransplantation. Among the remaining 7 patients, only 1 patient developed fibrosis stage 2 at 24 months, while others showed no histologic evidence of fibrosis for the duration of follow-up (4 years). Thus, we suggest that addition of the expression array profiling of 12-month biopsy specimens and use of the described patterns as markers of early fibrosis development may help in identification of transplant recipients at risk for progressive liver disease. The authors thank Zhaoxia Yue for technical assistance, Hao A. Do for clinical sample collection, and Drs Loreto Boix and Jordi Bruix for surgical liver samples.