- Email: [email protected]
Differential allograft gene expression in acute cellular rejection and recurrence of hepatitis C after liver transplantation
Differential Allograft Gene Expression in Acute Cellular Rejection and Recurrence of Hepatitis C After Liver Transplantation Raghavakaimal Sreekumar, Deborah L. Rasmussen, Russell H. Wiesner, and Michael R. Charlton Treatment of acute cellular rejection (ACR ) is associated with increased viral load, more severe histologic recurrence, and diminished patient and graft survival after liver transplantation for hepatitis C virus (HCV). Recurrence of HCV may be difficult to distinguish histologically from ACR . Because the immunologic mechanisms of ACR and HCV recurrence are likely to differ, we hypothesized that ACR is associated with the expression of a specific subset of immune activation genes that may serve as a diagnostic indicator of ACR and provide mechanistic insight into the pathophysiology of ACR and recurrence of HCV. The goal of the study was to study intragraft gene expression patterns in ACR and during recurrence of HCV in HCVinfected recipients. High-density microarrays were used to determine relative intragraft gene expression in two groups of HCV-infected liver transplant recipients: four with steroid responsive ACR by Banff criteria and four age- and gender-matched HCV-infected recipients with similar necroinflammatory activity but without histological criteria for rejection (no cholangitis or endotheliitis). Immunosuppression was similar in both groups. Other etiologies of graft dysfunction were excluded by ultrasound, cholangiography, and cultures. High-quality total RNA was extracted from snap frozen liver biopsies, reverse transcribed, labeled with biotin, and fragmented according to established protocol. Twenty-five genes were relatively overexpressed, and 15 were relatively underexpressed by >twofold in the ACR when compared with the HCV group. ACR was most notably associated with the relative overexpression of genes associated with major histocompatibility complex I and II, insulin-like growth factor–1 expression, apoptosis induction, and T-cell activation. In HCV-infected liver transplant recipients, ACR is associated with an intragraft gene expression profile that is distinct from that seen during recurrence of HCV. These experiments provide evidence that alloimmunity, as indicated by expression of T-cell activation and apoptosisinducing genes, is less important in recurrence of HCV
From the Transplant Center, Mayo Clinic and Foundation, Rochester, MN. Supported by a grant from the Carlson-Nelson Foundation, Minneapolis, MN. Address reprint requests to Michael R. Charlton, MD, Division of Gastroenterology and Hepatology, Mayo Clinic and Foundation, 200 First St SW, Rochester, MN 55905. Telephone: 507-266-7054; FAX: 507-266-1856; E-mail: [email protected] Copyright © 2002 by the American Association for the Study of Liver Diseases 1527-6465/02/0809-0096$35.00/0 doi:10.1053/jlts.2002.35173
814
than in ACR. Further studies are required to determine whether gene expression profiles, either intragraft or in serum, can be used for the diagnosis and differentiation of ACR from recurrence of HCV. (Liver Transpl 2002;8: 814-821.)
H
igher pretransplantation and posttransplantation levels of viremia are associated with more severe histologic recurrence of hepatitis C after liver transplantation.1-3 Higher average daily steroid dose4,5 and use of OKT36,7 (Orthoclone; Ortho Biotech, Raritan, NJ) have both been associated with more severe recurrence of hepatitis C, presumably through enhancing viral replication and/or attenuating viral clearance. Minimizing exposure to immunosuppression for hepatitis C–infected liver transplant recipients thus requires the accurate distinction of recurrence of hepatitis C from acute cellular rejection (ACR). The diagnosis of ACR is based on the detection of biochemical evidence of graft dysfunction and the presence of suggestive allograft histology, including distinct lymphocytic infiltrate patterns.8 The presence of a modest cellular infiltrate and biochemical abnormalities, however, are not specific to ACR. The clinical and histologic distinction of the effects of hepatitis C reinfection from ACR is made more difficult by the fact that these events may occur simultaneously. Because therapeutic agents that inhibit rejection are associated with more severe recurrence of hepatitis C virus (HCV), it seems likely that these events are mechanistically distinct. ACR is characterized by antigen-triggered T-cell activation and the subsequent migration of activated CD4⫹ and CD8⫹ T-cells, macrophages, and natural killer cells.9 Because T-cell activation is characterized by a preprogrammed sequence of tightly regulated gene expression, we hypothesized that ACR is associated with the expression of a subset of T-cell– dependent immune activation genes that is distinct from those associated with recurrence of HCV. To evaluate this hypothesis requires a method that can provide simultaneous analysis of the relative expression of a large number of genes in a sensitive manner. We used high-density oligonucleotide microarrays to determine the relative expression of more than 6400
Liver Transplantation, Vol 8, No 9 (September), 2002: pp 814-821
815
Differential Allograft Gene Expression
genes during acute cellular rejection.10 The direct, combinatorial synthesis of oligonucleotides based on sequence information, as used in high-density microarrays, allows the hybridization patterns and signal intensities to be determined for specific genes without additional sequencing or characterization.11 In addition to facilitating the distinction of ACR from recurrence of HCV, identification of differentially expressed genes associated with allograft rejection may generate novel insights into the pathophysiology of ACR . Furthermore, differential cytokine expression analysis might facilitate adjustments to immunosuppression before rejection, producing histologically apparent allograft injury.
Table 1. Demographics
Age (yr) Height (m) Weight (kg) HAI Fibrosis stage Creatinine (mg/dL) Albumin (g/dL) Bilirubin (g/dL)
ACR
HCV
48.3 ⫾ 6.1 1.65 ⫾ 0.16 72.0 ⫾ 8.9 5.3 ⫾ 2.7 0 ⫾ 0.0 1.4 ⫾ 0.3 3.0 ⫾ 0.5 2.9 ⫾ 3.0
46.0 ⫾ 7.9 1.72 ⫾ 0.18 75.3 ⫾ 9.1 5.7 ⫾ 3.9 0 ⫾ 0.0 1.5 ⫾ 0.4 3.2 ⫾ 0.6 3.3 ⫾ 2.6
Abbreviations: ACR, HCV-infected subjects with acute cellular rejection on day 21 liver biopsy; HCV, HCV-infected subjects without acute cellular rejection on day 21 liver biopsy or at any time subsequently; HAI, histology activity index.
Materials and Methods Subjects The primary purpose of this study was to examine differences in gene expression that occur during ACR when compared with recurrence of HCV. Two groups of subjects were studied. ACR group. This group comprised four HCV-infected liver transplant recipients with ACR on protocol-based posttransplant day 21 liver biopsies (regardless of biochemical profile). All biopsies had cholangitis, endotheliitis, and portal hepatitis in ⱖ50% of portal tracts (mild ACR by Banff criteria). HCV group. This group comprised four HCV-infected liver transplant recipients with similar degrees of necroinflammatory activity on protocol-based posttransplant day 21 liver biopsies (as measured by histology activity index12,13) as the ACR group, but without cholangitis or endotheliitis. Recipients in this group did not receive treatment for ACR. Biopsies were obtained according to clinical protocol, regardless of biochemical profile. Both the ACR and HCV groups were matched for race, age, gender, and body mass index (BMI) and histology activity index (Table 1). Liver biopsies for both groups were performed using an 18-gauge biopsy needle and were immediately snap-frozen in liquid nitrogen and stored at ⫺80°C. All subjects in the ACR group who developed classical acute rejection within 21 days posttransplantation showed complete resolution of ACR histologically and biochemically with a single course of methyl prednisolone therapy (1000 mg on alternate days for a total of 3000 mg). Subjects in the HCV group did not require treatment for ACR at any time and had subsequent allograft histology showing persistent and progressive histologic changes (hepatitis activity index ⱖ3 and fibrosis stage ⱖ2) consistent with recurrence of hepatitis C (defined by portal ⫾ lobular hepatitis without zone three predominance, balloon degeneration of hepatocytes ⫾ mild steatosis ⫾ fibrosis in the absence of cholangitis and endotheliitis) at 1 year posttransplantation. Protocol ultrasounds
(obtained in all subjects before biopsy), cholangiograms, and cultures were used to exclude vascular, biliary, and infectious etiologies of allograft dysfunction. Immunosuppression regimens were identical in both groups, consisting of tacrolimus (mean trough levels 11.5 ⫾1.8 and 12.4 ⫾ 2.5 ng/mL at the time of biopsy in the ACR and HCV groups respectively), intraoperative methylprednisolone (1000 mg), and a standard oral prednisone taper. Quantitative HCV RNA levels were not measured as part of this analysis.
Microarrays To determine the relative intrahepatic gene expression profiles in the two study groups, we analyzed the relative abundance of mRNA using high-density oligonucleotide microarrays containing probes for 6412 known genes (Hu6800 GeneChip; Affymetrix, Santa Clara, CA). Hu6800 arrays contain collections of pairs of probes consisting of a 25-mer that is perfectly complimentary (referred to as a perfect match, or PM) to a subsequence of a particular message and a companion 25-mer that is identical except for a single base difference in a central position. The mismatch (MM) probe of each pair serves as an internal control for hybridization specificity. The analysis of PM/MM pairs allows low-intensity hybridization patterns from mRNAs to be sensitively and accurately recognized in the presence of cross-hybridization signals.10 RNA Isolation Total RNA was isolated from frozen liver tissue by using TRIZOL reagent (Life Technologies, Gaithersberg, MD), which was further purified using an affinity resin column (RNeasy; Qiagen, Chatsworth, CA). Total isolated RNA was converted to cDNA using the
816
Sreekumar et al
Superscript cDNA synthesis kit (Gibco-BRL, Gaithersberg, MD). Double-stranded cDNA was then purified by phase lock gel (Eppendorf, Westbury, NY) with phenol/chloroform extraction.14 Liver tissue from each of the subjects was processed separately. Sample Preparation, Fragmentation, Array Hybridization, and Scanning The purified cDNA was used as a template for the in vitro transcription reaction for the synthesis of biotinylated cRNA using RNA transcript labeling reagent (Affymetrix). This labeled cRNA was fragmented and hybridized onto the Hu6800 array as previously described.14 Briefly, fragmented cRNA and control oligonucleotide B2 were added along with control cRNA (BioB, BioC, BioD), herring sperm DNA and bovine serum albumin to the hybridization buffer. The hybridization mixture was heated at 99°C for 5 minutes, followed by incubation at 45°C for 5 minutes, before injecting the sample into the microarray. Hybridization was then carried out at 45°C for 16 hours with mixing on a rotisserie at 60 rpm. After hybridization, the solutions were removed and the arrays were washed and stained with streptavidin-phycoerythrin (Molecular Probes, Portland, OR). After washes, probe arrays were scanned using the Hewlett-Packard gene chip system confocal scanner.14 Five to 10 g of total RNA was required from each biopsy sample for these analyses. Data Analysis After scanning, GeneChip 4.0 (Affymetrix) software calculated intensity values for each probe cell and made a presence or absence call for each mRNA. Algorithms in the software use probe cell intensities to calculate an average intensity for each set of probe pairs representing a gene, which directly correlates with the abundance of mRNA. Spotfire (Spotfire, Cambridge, MA) and Microsoft Excel (Microsoft, Redmond, WA) were also used for data analysis. For assessing the relative abundance of mRNA expression in the ACR group, gene expression profiles of each patient were compared with those of age-, gender-, and BMI-matched subjects from the HCV group in side-by-side experiments. The fold changes represent the average of all four possible pairwise comparisons among individual subjects on the basis of their age, gender, and BMI. The selection criteria for genes reported as differentially expressed were as follows: (1) a call of present in absolute analysis; (2) a difference call of either increased or decreased; (3) average difference in hybridization intensity of ⱖ1000 on both HCV subjects and ACR patients; (4) ⱖtwofold changes in all four pairwise comparisons.
Only genes with transcript levels that satisfied all four criteria were considered as significantly differentially expressed (either relatively overexpressed or underexpressed in one group when compared with the other). Differentially expressed genes were classified into different groups according to function class. Log average differences of patients in the HCV group were compared with those of well-matched patients in the ACR group using an unpaired, two-tailed t-test.
Results Study Population The clinical characteristics of both groups are summarized in Table 1. Subjects in the ACR and HCV groups had similar mean histology activity indices and biochemical profiles. BMI (weight in kg/height in m2) was also similar between the two groups. Relative Intragraft Gene Expression Of the 6412 genes surveyed, 25 were relatively overexpressed by ⱖtwofold and 15 were relatively underexpressed in the ACR group when compared with the HCV group. Differentially expressed genes included genes involved in protein synthesis/degradation and folding (32%), transcription factors (5%), immune function (5%), growth factors (8%), glucose metabolism (8%), fatty acid metabolism (12%), DNA metabolism (5%), and apoptosis (10%). The largest differential expression between ACR and HCV group (decreased 5.3-fold) was observed for the gene encoding the free radical scavenging enzyme glutathione peroxidase. The Hu6800 GeneChip contains oligonucleotide probe sets specific for multiple housekeeping genes, including GAPDH and 28S, that were expressed similarly in both study groups in these experiments. The 25 genes that were relatively overexpressed in the ACR group included: transcription factor ISGF-3 and interferon-responsive transcription factor (transcription factors), heat shock protein 70 (stress response/ chaperone), ubiquitin-conjugating enzyme E2, ubiquitin, ubiquitin-activating enzyme E1 and granzyme B (protein degradation), nicotinamide N-methyltransferase (nicotinamide metabolism), major histocompatibility complex (MHC) class I and II (immune function), transforming growth factor (TGF)– beta and insulin-like growth factor I (growth factors), glycogen synthase and phosphoenolpyruvate carboxykinase (glucose metabolism), cytidine triphosphate (CTP) synthetase, medium-chain acyl-CoA dehydrogenase and triglyceride lipase (fatty acid metabolism), complement components C1q and C3 (complement activation),
Differential Allograft Gene Expression
817
Table 2. Genes Differentially Expressed in Acute Cellular Rejection Probe Set T-cell and B-cell activation K03430 HG2915 K01160 M28879 X00695 K02765 U25956 Apoptosis U37518 U69611 M59465 X02910 Growth factors X05839 X57025 M74587 Transcription factors/stress response M97935 M87503 M11717
Fold ⌬
Gene Name
13.8 13.1 13.0 12.9 12.8 12.4 12.1
Complement C1q MHC class I MHC class II Granzyme B Interleukin-2 Complement component C3 P-selectin
⬍.05 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01
14.9 13.0 12.4 12.1
TNF-related apoptosis inducing ligand TNF-alpha converting enzyme TNF-alpha inducible protein A20 TNF-alpha
⬍.01 ⬍.01 ⬍.01 ⬍.01
13.2 12.5 22.8
Transforming growth factor beta 1 Insulin-like growth factor I IGF-binding protein
⬍.05 ⬍.05 ⬍.01
13.1 12.3 12.4
Transcription factor ISGF-3 IFN-responsive transcription factor Heat shock protein 70
⬍.01 ⬍.01 ⬍.01
P Value
Note. Fold change indicates relative abundance of mRNA in allografts from HCV-infected patients with acute cellular rejection (ACR group) when compared with those with HCV and no rejection (HCV group). Abbreviations: TNF, tumor necrosis factor, IGF, insulin-like growth factor.
p-selectin (cell adhesion), tumor necrosis factor (TNF)–related apoptosis inducing ligand (TRAIL), TNF-alpha converting enzyme, TNF-alpha inducible protein A20 and TNF-alpha (apoptosis). A summary of these results is shown in Tables 2 and 3. Genes that were relatively underexpressed in the ACR group included: alanyl-tRNA synthetase, ribosomal protein-L8, elongation TU, protein synthesis factor eIF-4C, elongation factor-2, eukaryotic initiation factor-4AI and elongation factor-1 alpha (protein synthesis), chaperonin 10 and protein disulfide isomerase (protein folding), insulin-like growth factor (IGF)– binding protein (growth factor), GLUT-2 (glucose metabolism), very-long-chain acyl CoA dehydrogenase and fatty acid omega hydroxylase (fatty acid metabolism), MT-1 and glutathione peroxidase (DNA metabolism). These results are summarized in Tables 2 and 3.
Discussion In this study of HCV-infected liver transplant recipients with comparable histologic activity indices, we found that of 6412 genes surveyed, 25 were relatively overexpressed and 15 were relatively underexpressed during ACR when compared with recipients with
uncomplicated early recurrence of HCV infection. Our principal finding is thus that, although ACR and early recurrence of HCV may have similar clinical manifestations, they are quite distinct at the level of mRNA abundance. The results of these experiments suggest that recurrence of HCV is not, primarily, an immunemediated process. When interpreting the results of this type of study, it is important to bear in mind that differential expression will be secondary in nature in the great majority of instances. Furthermore, the results of high-density microarray studies provide insight into gene expression, not transcription nor posttranscriptional regulation. It is also likely that there are genes that were differentially expressed at levels that were beyond the sensitivity of the method used in this study. The information that this type of microarray study generates provides a relatively global view of the relative abundance of mRNA for many of the human genes of known function and thus clues, rather than answers, concerning the relative pathophysiology of ACR and recurrence of HCV. These caveats notwithstanding, ACR was most notably associated with the overexpression of genes associated with MHC I and II, apoptosis, and a distinct
818
Sreekumar et al
Table 3. Genes Differentially Expressed in Acute Cellular Rejection Probe Set Protein metabolism D83004 U49869 M58028 D32050 Z28407 X03689 L18960 X51466 D13748 J04617 U07550 Z49835 Glucose metabolism S70004 L05144 J03810 X52142 Fatty acid metabolism M91432 M29194 D43682 D13705 DNA/nicotinamide metabolism X76717 D00632
Fold ⌬
Gene Name
13.0 12.1 12.0 22.1 22.2 22.6 22.9 22.9 23.2 23.6 22.3 22.7
Ubiquitin-conjugating enzyme E2 Ubiquitin Ubiquitin-activating enzyme E1 Alanyl-tRNA synthetase Ribosomal protein L8 Elongation factor TU Protein synthesis factor (elF-4C) Elongation factor 2 Eukaryotic initiation factor 4AI Elongation factor EF-1-alpha Chaperonin 10 Protein disulfide isomerase
12.8 12.2 22.3 14.3
Glycogen synthase Phosphoenolpyruvate carboxykinase GLUT2 CTP synthetase
⬍0.01 ⬍0.01 ⬍0.01 ⬍0.01
13.0 12.4 22.3 22.3
Medium-chain acyl-CoA dehydrogenase Triglyceride lipase Very-long-chain acyl-CoA dehydrogense Fatty acids omega-hydroxylase
⬍0.05 ⬍0.01 ⬍0.01 ⬍0.01
24.1 25.3
MT-1 Glutathione peroxidase
⬍0.01 ⬍0.01
P Value ⬍.05 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.05 ⬍.01 ⬍.05 ⬍.01 ⬍.01
Note. Fold change indicates relative abundance of mRNA in allografts from HCV-infected patients with acute cellular rejection (ACR group) when compared with those with HCV and no rejection (HCV group).
subset of T-cell activation genes. Interleukin 2 (IL-2) is an immunoregulatory lymphokine produced by lectinor antigen-activated T cells. IL-2 can act as a mitogen for both B- and T-lymphocytes and is known to be upregulated in ACR.15 Interferon-stimulated gene factor-3 (ISGF3), an interferon-dependent (␣ and ), positive-acting transcription factor that is cytoplasmically activated,16 was also relatively overexpressed in the ACR group.17 Expression of ISGF3 is known to be inhibited by the HCV NS5 protein.18 Because it is not known whether HCV RNA levels were different between the ACR and HCV groups, a mechanistic role of HCV in the differential expression of ISGF3 in the ACR group can only be speculated on. Granzyme B, a serine protease produced by cytotoxic T-lymphocytes and natural killer/lymphokine-activated cells, was also relatively overexpressed in the ACR group. Perforin and granzyme B act synergistically to trigger an endogenous apoptosis pathway resulting in the destruction of target cell nuclear membrane and DNA fragmentation.19 Gene transcript levels of granzyme B are known to be enhanced during ACR after liver and kidney transplan-
tation.20-23 Members of the tumor necrosis factor (TNF) family of cytokines and receptors are critically involved in cellular activation, proliferation, and cell death.23 Within this family, Fas ligand and TNF-related apoptosis-inducing ligand (TRAIL)24-26 have been shown to be mediators of apoptosis. Two further genes from the TNF superfamily, TNF-alpha converting enzyme and TNF-alpha inducible protein A20, were also relatively overexpressed in the ACR group. Transforming growth factor-beta 1, in addition to activating stellate cells, is proapoptotic.27,28 Expression of P-selectin, a 140-kD adhesion molecule that mediates the interaction of activated endothelial cells or platelets with leukocytes,29,30 was also increased in patients with ACR. Overexpression of this group of genes in the ACR group suggests that T-cell activation and apoptosis are mechanistically relatively more important in ACR than in recurrence of HCV. Because apoptosis is not thought to be pretranscriptionally regulated, it is likely that the relatively greater abundance of mRNA for Fas ligand and TRAIL in the ACR group is on the basis of differences in the relative abundance of lymphocyte cell types
Differential Allograft Gene Expression
within allograft cellular infiltrates (e.g., greater proportion of T than B lymphocytes in ACR allografts). The complement system plays a paradoxical role in the development and expression of autoimmunity and alloimmunity in humans.31 Transcript levels of genes for both C1q and C3 were relatively greater in the ACR group. C1q and C3 are known to be increased in acute cellular rejection.32 Subcomponent C1q binds to immunoglobulin complexes with resulting serial activation of C1r (enzyme), C1s (proenzyme), and the other eight components of complement. Complement component C3 is an acute phase reactant. Increased synthesis of C3 is also induced during acute inflammation.33 The liver is the main site of synthesis of C3, although small amounts are also produced by activated monocytes and macrophages. Overexpression of C1q and C3 are in keeping with enhanced complement activation during ACR. The relatively greater abundance of mRNA for immune activation genes in the ACR than HCV groups supports the concept that posttransplant recurrence of HCV infection is less immune mediated and may primarily involve a direct cytopathic effect of HCV infection, perhaps through diminished clearance of HCV from plasma and infected hepatocytes. This is in keeping with the clinical association of very high levels of HCV RNA with severe recurrence of HCV. It would be of substantial interest to compare hepatic gene expression in patients with posttransplant recurrence of HCV with that of patients with progressive HCV infection in the nontransplant setting to determine whether there are fundamental differences in the mechanisms of hepatic injury in these two settings. Whether the differential expression of immune activation genes in ACR and recurrence of HCV, as observed in these experiments, can be used to determine adequacy of immunosuppression or to differentiate ACR from recurrence of HCV in individual patients would require prospective studies. The possibility is, however, suggested by our results and those of previous reports of the utility of cytokine profiles in the diagnosis of acute cellular rejection.15 Increases in levels of IL-2 and expression of IL-2 and TNF-alpha receptors have been shown to precede biochemical and histologic abnormalities by several days in liver transplant recipients with acute cellular rejection.15 A similar finding has been reported in acute renal allograft rejection, in which expression of IL-10 and IL-15 also accompanies clinically apparent rejection.34 Others have reported the use of serum human lymphocyte antigen (HLA) class I antigens as a marker of acute rejection in human liver transplantations.35 Nonimmunomodulatory genes were also differen-
819
tially expressed during ACR. Gene transcript levels of ribosomal protein L8, protein synthesis factor eIF-4C, alanyl-tRNA synthetase, elongation factors TU, 2, and EF-1 alfa were all relatively underexpressed in the ACR group. Relative underexpression of these genes in the ACR group suggests a global impairment in liver protein synthesis at the initiation, elongation, and folding phases as well as stimulation of protein breakdown during ACR. Because immunosuppression type and levels were similar between the ACR and HCV groups, the differential expression of these genes is likely to have been secondary to events associated with rejection (e.g., diminished substrate availability because of endotheliitis-induced ischemia). Specific effects of lower transcript levels would be inhibition of 43S preinitiation complex formation (ribosomal protein L8 and eIF-4C), ribosomal translocation of tRNAs and mRNA (EF-TU and alanyl-tRNA synthetase), and protein translocation (EF-2). Chaperonin 10, which was also relatively underexpressed in the ACR group, is involved in ATPdependent protein folding and may have been downregulated secondary to diminished protein synthesis.36 Ubiquitin and ubiquitin activating enzyme E1 were relatively overexpressed in the ACR group. Enhanced protein ubiquitination, secondary to overexpression of ubiquitin and ubiquitin activating enzyme E1, would be expected to stimulate proteolysis.37,38 Differential expression of these genes may have been secondary to relatively greater local cytokine levels. Stimulation of proteolysis and inhibition of protein synthesis by cytokines (e.g., TGF-beta and TNF) has been well described.39-41 Heat shock protein 70, which is thought to be involved in protection of the cytoskeleton during stress,42 was relatively overexpressed in the ACR group, possibly reflecting higher oxidative stress in patients with ACR. The transcriptional activation of stress response gene (HSP 70) and decreased expression of chaperonin 10 and protein disulfide-isomerase that process damaged or misfolded proteins during the ACR suggest a central role for protein modifications associated with ACR. Transcript levels of genes coding for reactive oxygen species scavenging enzymes, metallothionine 1 and glutathione peroxidase, were relatively underexpressed in the ACR group. Both these genes play a protective role against oxidative damage by neutralizing reactive oxygen species.43-45 Removal of the superoxide radical by superoxide dismutate and hydrogen peroxide by glutathione peroxidase prevents formation of reactive hydroxyl radicals, which are postulated to be responsible for oxidative cellular injury. Although the mecha-
820
Sreekumar et al
nism is unclear, HCV infection is known to induce oxidative stress and DNA damage.46-48 Finally, three genes involved in glucose metabolism, glycogen synthase, phosphoenolpyruvate carboxykinase, and GLUT2, were differentially underexpressed in the ACR group. Glycogen synthase catalyzes the rate-limiting step in glycogen synthesis, whereas phosphoenolpyruvate carboxykinase is a key enzyme in gluconeogenesis. The gene transcript levels of both these genes were relatively higher in the ACR group. In contrast, GLUT2 (a low-affinity transporter of glucose present in the plasma membrane of hepatocytes) and insulin-like growth factor-1 (IGF-I) in hepatocytes, a stimulator of glycolytic metabolism through activation of the calcium calmodulin-dependent phosphatase calcineurin,49-52 were relatively overexpressed in the ACR group. The finding of relatively increased expression of IGF-1 among patients with ACR is of particular interest because IGF-1 stimulates calcineurin expression, opposing the effects of calcineurin inhibitors. In light of this observation, we are currently studying whether higher posttransplant IGF-1 levels are associated with increased risk of ACR in general. IGF-1 induces expression of the transcription factor GATA-2, which associates with calcineurin and a specific dephosphorylated isoform of the transcription factor NF-ATc1.50 Thus, IGF-1 induces calcineurin-mediated signaling. In summary, ACR and recurrence of HCV are associated with distinct mRNA expression patterns. In all likelihood, this reflects mechanistic differences in these two events. The differential expression of intrahepatic mRNA raises the possibility that gene expression, e.g., of immunomodulatory peptides, may be useful in the distinction of ACR from recurrence of HCV. In addition, these experiments provide evidence that alloimmunity, as indicated by expression of T-cell activation and apoptosis-inducing genes, is less important in recurrence of HCV than in ACR.
References 1. Charlton M, Seaberg E, Wiesner R, Everhart J, Zetterman R, Lake J, et al. Predictors of patient and graft survival after liver transplantation for hepatitis C. Hepatology 1998;28:823-830. 2. Sreekumar R, Gonzalez-Koch A, Maor-Kendler Y, Batts K, Moreno-Luna L, Poterucha J, et al. Early identification of recipients with progressive histologic recurrence of hepatitis C after liver transplantation. Hepatology 2000;32:1125-1130. 3. Berenguer M, Ferrell L, Watson J, Prieto M, Kim M, Rayon M, et al. HCV-related fibrosis progression after liver transplantation: increase in recent years. J Hepatol 2000;32:673-684. 4. Gane EJ, Naoumov NV, Qian KP, Mondelli MU, Maertens G, Portmann BC, et al. A longitudinal analysis of hepatitis C virus
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
replication after liver transplantation. Gastroenterology 1996; 110:167-177. Charlton M, Seaberg E. Impact of immunosuppression and acute rejection on recurrence of hepatitis C: Results of the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database. Liver Transpl 1999; 5(suppl 1):S107-S114. Rosen HR, Shackleton CR, Higa L, Gralnek IM, Farmer DA, McDiarmid SV, et al. Use of OKT3 is associated with early and severe recurrence of hepatitis C after liver transplantation [see comments]. Am J Gastroenterol 1997;92:1453-1457. Sheiner PA, Schwartz ME, Mor E, Schluger LK, Theise N, Kishikawa K, et al. Severe or multiple rejection episodes are associated with early recurrence of hepatitis C after orthotopic liver transplantation. Hepatology 1995;21:30-34. Demetris AJ, Seaberg EC, Batts KP, Ferrell LD, Ludwig J, Markin RS, et al. Reliability and predictive value of the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database nomenclature and grading system for cellular rejection of liver allografts. Hepatology 1995;21:408416. Si L, Whiteside TL, Van Thiel DH, Rabin BS. Lymphocyte subpopulations at the site of “piecemeal” necrosis in end stage chronic liver diseases and rejecting liver allografts in cyclosporine-treated patients. Lab Invest 1984;50:341-347. Lockhart DJ, Dong H, Byrne MC, Follettie MT, Gallo MV, Chee MS, et al. Expression monitoring by hybridization to highdensity oligonucleotide arrays. Nature Biotechnology 1996;14: 1675-1680. Fodor SP, Read JL, Pirrung MC, Stryer L, Lu AT, Solas D. Light-directed, spatially addressable parallel chemical synthesis. Science 1991;251:767-773. Knodell RG, Ishak KG, Black WC, Chen TS, Craig R, Kaplowitz N, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981;1:431-435. Batts KP, Ludwig J. Chronic hepatitis: An update on terminology and reporting. Surg Pathol 1995;19:1409-1417. Mahadevappa M, Warrington JA. A high-density probe array sample preparation method using 10- to 100-fold fewer cells. Nature Biotechnology 1999;17:1134-1136. Lowenthal JW, Zubler RH, Nabholz M, MacDonald HR. Similarities between interleukin-2 receptor number and affinity on activated B and T lymphocytes. Nature 1985;315:669-672. Levy DE, Kessler DS, Pine R, Darnell JE Jr. Cytoplasmic activation of ISGF3, the positive regulator of interferon-alpha-stimulated transcription, reconstituted in vitro. Genes Devel 1989; 3:1362-1371. Akalin E, Hendrix RC, Polavarapu RG, Pearson TC, Neylan JF, Larsen CP, et al. Gene expression analysis in human renal allograft biopsy samples using high-density oligoarray technology. Transplantation 2001;72:948-953. Polyak SJ, Khabar KS, Paschal DM, Ezelle HJ, Duverlie G, Barber GN, et al. Hepatitis C virus nonstructural 5A protein induces interleukin-8, leading to partial inhibition of the interferon-induced antiviral response. J Virol 2001;75:6095-6106. Trapani JA, Browne KA, Smyth MJ, Jans DA. Localization of granzyme B in the nucleus. A putative role in the mechanism of cytotoxic lymphocyte-mediated apoptosis. J Biol Chem 1996; 271:4127-4133. Krams SM, Villanueva JC, Quinn MB, Martinez OM. Expres-
Differential Allograft Gene Expression
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
sion of the cytotoxic T cell mediator granzyme B during liver allograft rejection. Transpl Immunol 1995;3:162-166. Tannapfel A, Kohlhaw K, Ebelt J, Hauss J, Liebert U, Berr F, et al. Apoptosis and the expression of Fas and Fas ligand (FasL) antigen in rejection and reinfection in liver allograft specimens. Transplantation 1999;67:1079-1083. Lipman ML, Stevens AC, Strom TB. Heightened intragraft CTL gene expression in acutely rejecting renal allografts. J Immunol 1994;152:5120-5127. Sharma VK, Bologa RM, Li B, Xu GP, Lagman M, Hiscock W, et al. Molecular executors of cell death— differential intrarenal expression of Fas ligand, Fas, granzyme B, and perforin during acute and/or chronic rejection of human renal allografts. Transplantation 1996;62:1860-1866. Smith CA, Farrah T, Goodwin RG. The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death. Cell 1994;76:959-962. Musgrave BL, Phu T, Butler JJ, Makrigiannis AP, Hoskin DW. Murine TRAIL (TNF-related apoptosis inducing ligand) expression induced by T cell activation is blocked by rapamycin, cyclosporin A, and inhibitors of phosphatidylinositol 3-kinase, protein kinase C, and protein tyrosine kinases: Evidence for TRAIL induction via the T cell receptor signaling pathway. Exp Cell Res 1999;252:96-103. Wu GS, Burns TF, McDonald ER III, Meng RD, Kao G, Muschel R, et al. Induction of the TRAIL receptor KILLER/DR5 in p53-dependent apoptosis but not growth arrest. Oncogene 1999;18:6411-6418. Ruuls SR, Sedgwick JD. Unlinking tumor necrosis factor biology from the major histocompatibility complex: lessons from human genetics and animal models. [Review] [38 refs]. Am J Hum Genet 1999;65:294-301. Fukuda K, Kojiro M, Chiu JF. Induction of apoptosis by transforming growth factor-beta 1 in the rat hepatoma cell line McARH7777: A possible association with tissue transglutaminase expression. Hepatology 1993;18:945-953. Singbartl K, Forlow SB, Ley K. Platelet, but not endothelial, P-selectin is critical for neutrophil-mediated acute postischemic renal failure. FASEB J 2001;15:2337-2344. Coughlan AF, Berndt MC, Dunlop LC, Hancock WW. In vivo studies of P-selectin and platelet activating factor during endotoxemia, accelerated allograft rejection, and discordant xenograft rejection. Transplant Proc 1993;25:2930-2931. Kawano M. Complement regulatory proteins and autoimmunity. [Review] [52 refs]. Archivum Immunologiae et Therapiae Experimentalis 2000;48:367-372. Pratt JR, Abe K, Miyazaki M, Zhou W, Sacks SH. In situ localization of C3 synthesis in experimental acute renal allograft rejection. Am J Pathol 2000;157:825-831. Barrington R, Zhang M, Fischer M, Carroll MC. The role of complement in inflammation and adaptive immunity [Review] [42 refs]. Immunol Rev 2001;180:5-15. McLean AG, Hughes D, Welsh KI, Gray DW, Roake J, Fuggle SV, et al. Patterns of graft infiltration and cytokine gene expression during the first 10 days of kidney transplantation. Transplantation 1997;63:374-380. Renna ME, Puppo F, Brenci S, Scudeletti M, Cinti P, Alfani D, et al. Serum HLA class I soluble antigens: A marker of acute rejection after liver transplantation. Transplant Proc 1995;27: 1155-1156. Todd MJ, Viitanen PV, Lorimer GH. Dynamics of the chapero-
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
821
nin ATPase cycle: Implications for facilitated protein folding [Review] [57 refs]. Science 1994;265:659-666. Zwickl P, Voges D, Baumeister W. The proteasome: A macromolecular assembly designed for controlled proteolysis [Review] [139 refs]. Philos Trans R Soc Lond Series B: Biological Sciences 1999;354:1501-1511. Ciechanover A. Ubiquitin-mediated degradation of cellular proteins: why destruction is essential for construction, and how it got from the test tube to the patient’s bed [Review] [40 refs]. Isr Med Assoc J: Imaj 2001;3:319-327. Charters Y, Grimble RF. Effect of recombinant human tumour necrosis factor alpha on protein synthesis in liver, skeletal muscle and skin of rats. Biochem J 1989;258:493-497. Llovera M, Garcia-Martinez C, Lopez-Soriano FJ, Argiles JM. The effect of chronic tumour necrosis factor-alpha treatment on urinary nitrogen excretion in the rat. Biochem Mol Biol Int 1994;33:681-689. Tayek JA. Effects of tumor necrosis factor alpha on skeletal muscle amino acid metabolism studied in-vivo. J Am Coll Nutr 1996;15:164-168. Wang TT, Chiang AS, Chu JJ, Cheng TJ, Chen TM, Lai YK. Concomitant alterations in distribution of 70 kDa heat shock proteins, cytoskeleton and organelles in heat shocked 9L cells. Int J Biochem Cell Biol 1998;30:745-759. Esposito LA, Kokoszka JE, Waymire KG, Cottrell B, MacGregor GR, Wallace DC. Mitochondrial oxidative stress in mice lacking the glutathione peroxidase-1 gene. Free Radic Biol Med 2000; 28:754-766. Gobel H, van der Wal AC, Teeling P, van der Loos CM, Becker AE. Metallothionein in human atherosclerotic lesions: a scavenger mechanism for reactive oxygen species in the plaque? Virchows Arch 2000;437:528-533. Ren Y, Smith A. Mechanism of metallothionein gene regulation by heme-hemopexin. Roles of protein kinase C, reactive oxygen species, and cis-acting elements. J Biol Chem 1995;270:2398823995. Cardin R, Saccoccio G, Masutti F, Bellentani S, Farinati F, Tiribelli C. DNA oxidative damage in leukocytes correlates with the severity of HCV-related liver disease: Validation in an open population study. J Hepatol 2001;34:587-592. Vendemiale G, Grattagliano I, Portincasa P, Serviddio G, Palasciamo G, Altomare E. Oxidative stress in symptom-free HCV carriers: relation with ALT flare-up. Eur J Clin Invest 2001;31: 54-63. Sumida Y, Nakashima T, Yoh T, Nakajima Y, Ishikawa H, Mitsuyoshi H, et al. Serum thioredoxin levels as an indicator of oxidative stress in patients with hepatitis C virus infection. J Hepatol 2000;33:616-622. Gooch JL, Tang Y, Ricono JM, Abboud HE. Insulin-like growth factor-I induces renal cell hypertrophy via a calcineurin-dependent mechanism. J Biol Chem 2001;276:42492-42500. Musaro A, McCullagh KJ, Naya FJ, Olson EN, Rosenthal N. IGF-1 induces skeletal myocyte hypertrophy through calcineurin in association with GATA-2 and NF-ATc1. Nature 1999;400:581-585. Pons S, Torres-Aleman I. Insulin-like growth factor-I stimulates dephosphorylation of ikappa B through the serine phosphatase calcineurin (protein phosphatase 2B). J Biol Chem 2000;275: 38620-38625. Semsarian C, Wu MJ, Ju YK, Marciniec T, Yeoh T, Allen DG, et al. Skeletal muscle hypertrophy is mediated by a Ca2⫹-dependent calcineurin signaling pathway. Nature 1999;400:576-581.
From the Transplant Center, Mayo Clinic and Foundation, Rochester, MN. Supported by a grant from the Carlson-Nelson Foundation, Minneapolis, MN. Address reprint requests to Michael R. Charlton, MD, Division of Gastroenterology and Hepatology, Mayo Clinic and Foundation, 200 First St SW, Rochester, MN 55905. Telephone: 507-266-7054; FAX: 507-266-1856; E-mail: [email protected] Copyright © 2002 by the American Association for the Study of Liver Diseases 1527-6465/02/0809-0096$35.00/0 doi:10.1053/jlts.2002.35173
814
than in ACR. Further studies are required to determine whether gene expression profiles, either intragraft or in serum, can be used for the diagnosis and differentiation of ACR from recurrence of HCV. (Liver Transpl 2002;8: 814-821.)
H
igher pretransplantation and posttransplantation levels of viremia are associated with more severe histologic recurrence of hepatitis C after liver transplantation.1-3 Higher average daily steroid dose4,5 and use of OKT36,7 (Orthoclone; Ortho Biotech, Raritan, NJ) have both been associated with more severe recurrence of hepatitis C, presumably through enhancing viral replication and/or attenuating viral clearance. Minimizing exposure to immunosuppression for hepatitis C–infected liver transplant recipients thus requires the accurate distinction of recurrence of hepatitis C from acute cellular rejection (ACR). The diagnosis of ACR is based on the detection of biochemical evidence of graft dysfunction and the presence of suggestive allograft histology, including distinct lymphocytic infiltrate patterns.8 The presence of a modest cellular infiltrate and biochemical abnormalities, however, are not specific to ACR. The clinical and histologic distinction of the effects of hepatitis C reinfection from ACR is made more difficult by the fact that these events may occur simultaneously. Because therapeutic agents that inhibit rejection are associated with more severe recurrence of hepatitis C virus (HCV), it seems likely that these events are mechanistically distinct. ACR is characterized by antigen-triggered T-cell activation and the subsequent migration of activated CD4⫹ and CD8⫹ T-cells, macrophages, and natural killer cells.9 Because T-cell activation is characterized by a preprogrammed sequence of tightly regulated gene expression, we hypothesized that ACR is associated with the expression of a subset of T-cell– dependent immune activation genes that is distinct from those associated with recurrence of HCV. To evaluate this hypothesis requires a method that can provide simultaneous analysis of the relative expression of a large number of genes in a sensitive manner. We used high-density oligonucleotide microarrays to determine the relative expression of more than 6400
Liver Transplantation, Vol 8, No 9 (September), 2002: pp 814-821
815
Differential Allograft Gene Expression
genes during acute cellular rejection.10 The direct, combinatorial synthesis of oligonucleotides based on sequence information, as used in high-density microarrays, allows the hybridization patterns and signal intensities to be determined for specific genes without additional sequencing or characterization.11 In addition to facilitating the distinction of ACR from recurrence of HCV, identification of differentially expressed genes associated with allograft rejection may generate novel insights into the pathophysiology of ACR . Furthermore, differential cytokine expression analysis might facilitate adjustments to immunosuppression before rejection, producing histologically apparent allograft injury.
Table 1. Demographics
Age (yr) Height (m) Weight (kg) HAI Fibrosis stage Creatinine (mg/dL) Albumin (g/dL) Bilirubin (g/dL)
ACR
HCV
48.3 ⫾ 6.1 1.65 ⫾ 0.16 72.0 ⫾ 8.9 5.3 ⫾ 2.7 0 ⫾ 0.0 1.4 ⫾ 0.3 3.0 ⫾ 0.5 2.9 ⫾ 3.0
46.0 ⫾ 7.9 1.72 ⫾ 0.18 75.3 ⫾ 9.1 5.7 ⫾ 3.9 0 ⫾ 0.0 1.5 ⫾ 0.4 3.2 ⫾ 0.6 3.3 ⫾ 2.6
Abbreviations: ACR, HCV-infected subjects with acute cellular rejection on day 21 liver biopsy; HCV, HCV-infected subjects without acute cellular rejection on day 21 liver biopsy or at any time subsequently; HAI, histology activity index.
Materials and Methods Subjects The primary purpose of this study was to examine differences in gene expression that occur during ACR when compared with recurrence of HCV. Two groups of subjects were studied. ACR group. This group comprised four HCV-infected liver transplant recipients with ACR on protocol-based posttransplant day 21 liver biopsies (regardless of biochemical profile). All biopsies had cholangitis, endotheliitis, and portal hepatitis in ⱖ50% of portal tracts (mild ACR by Banff criteria). HCV group. This group comprised four HCV-infected liver transplant recipients with similar degrees of necroinflammatory activity on protocol-based posttransplant day 21 liver biopsies (as measured by histology activity index12,13) as the ACR group, but without cholangitis or endotheliitis. Recipients in this group did not receive treatment for ACR. Biopsies were obtained according to clinical protocol, regardless of biochemical profile. Both the ACR and HCV groups were matched for race, age, gender, and body mass index (BMI) and histology activity index (Table 1). Liver biopsies for both groups were performed using an 18-gauge biopsy needle and were immediately snap-frozen in liquid nitrogen and stored at ⫺80°C. All subjects in the ACR group who developed classical acute rejection within 21 days posttransplantation showed complete resolution of ACR histologically and biochemically with a single course of methyl prednisolone therapy (1000 mg on alternate days for a total of 3000 mg). Subjects in the HCV group did not require treatment for ACR at any time and had subsequent allograft histology showing persistent and progressive histologic changes (hepatitis activity index ⱖ3 and fibrosis stage ⱖ2) consistent with recurrence of hepatitis C (defined by portal ⫾ lobular hepatitis without zone three predominance, balloon degeneration of hepatocytes ⫾ mild steatosis ⫾ fibrosis in the absence of cholangitis and endotheliitis) at 1 year posttransplantation. Protocol ultrasounds
(obtained in all subjects before biopsy), cholangiograms, and cultures were used to exclude vascular, biliary, and infectious etiologies of allograft dysfunction. Immunosuppression regimens were identical in both groups, consisting of tacrolimus (mean trough levels 11.5 ⫾1.8 and 12.4 ⫾ 2.5 ng/mL at the time of biopsy in the ACR and HCV groups respectively), intraoperative methylprednisolone (1000 mg), and a standard oral prednisone taper. Quantitative HCV RNA levels were not measured as part of this analysis.
Microarrays To determine the relative intrahepatic gene expression profiles in the two study groups, we analyzed the relative abundance of mRNA using high-density oligonucleotide microarrays containing probes for 6412 known genes (Hu6800 GeneChip; Affymetrix, Santa Clara, CA). Hu6800 arrays contain collections of pairs of probes consisting of a 25-mer that is perfectly complimentary (referred to as a perfect match, or PM) to a subsequence of a particular message and a companion 25-mer that is identical except for a single base difference in a central position. The mismatch (MM) probe of each pair serves as an internal control for hybridization specificity. The analysis of PM/MM pairs allows low-intensity hybridization patterns from mRNAs to be sensitively and accurately recognized in the presence of cross-hybridization signals.10 RNA Isolation Total RNA was isolated from frozen liver tissue by using TRIZOL reagent (Life Technologies, Gaithersberg, MD), which was further purified using an affinity resin column (RNeasy; Qiagen, Chatsworth, CA). Total isolated RNA was converted to cDNA using the
816
Sreekumar et al
Superscript cDNA synthesis kit (Gibco-BRL, Gaithersberg, MD). Double-stranded cDNA was then purified by phase lock gel (Eppendorf, Westbury, NY) with phenol/chloroform extraction.14 Liver tissue from each of the subjects was processed separately. Sample Preparation, Fragmentation, Array Hybridization, and Scanning The purified cDNA was used as a template for the in vitro transcription reaction for the synthesis of biotinylated cRNA using RNA transcript labeling reagent (Affymetrix). This labeled cRNA was fragmented and hybridized onto the Hu6800 array as previously described.14 Briefly, fragmented cRNA and control oligonucleotide B2 were added along with control cRNA (BioB, BioC, BioD), herring sperm DNA and bovine serum albumin to the hybridization buffer. The hybridization mixture was heated at 99°C for 5 minutes, followed by incubation at 45°C for 5 minutes, before injecting the sample into the microarray. Hybridization was then carried out at 45°C for 16 hours with mixing on a rotisserie at 60 rpm. After hybridization, the solutions were removed and the arrays were washed and stained with streptavidin-phycoerythrin (Molecular Probes, Portland, OR). After washes, probe arrays were scanned using the Hewlett-Packard gene chip system confocal scanner.14 Five to 10 g of total RNA was required from each biopsy sample for these analyses. Data Analysis After scanning, GeneChip 4.0 (Affymetrix) software calculated intensity values for each probe cell and made a presence or absence call for each mRNA. Algorithms in the software use probe cell intensities to calculate an average intensity for each set of probe pairs representing a gene, which directly correlates with the abundance of mRNA. Spotfire (Spotfire, Cambridge, MA) and Microsoft Excel (Microsoft, Redmond, WA) were also used for data analysis. For assessing the relative abundance of mRNA expression in the ACR group, gene expression profiles of each patient were compared with those of age-, gender-, and BMI-matched subjects from the HCV group in side-by-side experiments. The fold changes represent the average of all four possible pairwise comparisons among individual subjects on the basis of their age, gender, and BMI. The selection criteria for genes reported as differentially expressed were as follows: (1) a call of present in absolute analysis; (2) a difference call of either increased or decreased; (3) average difference in hybridization intensity of ⱖ1000 on both HCV subjects and ACR patients; (4) ⱖtwofold changes in all four pairwise comparisons.
Only genes with transcript levels that satisfied all four criteria were considered as significantly differentially expressed (either relatively overexpressed or underexpressed in one group when compared with the other). Differentially expressed genes were classified into different groups according to function class. Log average differences of patients in the HCV group were compared with those of well-matched patients in the ACR group using an unpaired, two-tailed t-test.
Results Study Population The clinical characteristics of both groups are summarized in Table 1. Subjects in the ACR and HCV groups had similar mean histology activity indices and biochemical profiles. BMI (weight in kg/height in m2) was also similar between the two groups. Relative Intragraft Gene Expression Of the 6412 genes surveyed, 25 were relatively overexpressed by ⱖtwofold and 15 were relatively underexpressed in the ACR group when compared with the HCV group. Differentially expressed genes included genes involved in protein synthesis/degradation and folding (32%), transcription factors (5%), immune function (5%), growth factors (8%), glucose metabolism (8%), fatty acid metabolism (12%), DNA metabolism (5%), and apoptosis (10%). The largest differential expression between ACR and HCV group (decreased 5.3-fold) was observed for the gene encoding the free radical scavenging enzyme glutathione peroxidase. The Hu6800 GeneChip contains oligonucleotide probe sets specific for multiple housekeeping genes, including GAPDH and 28S, that were expressed similarly in both study groups in these experiments. The 25 genes that were relatively overexpressed in the ACR group included: transcription factor ISGF-3 and interferon-responsive transcription factor (transcription factors), heat shock protein 70 (stress response/ chaperone), ubiquitin-conjugating enzyme E2, ubiquitin, ubiquitin-activating enzyme E1 and granzyme B (protein degradation), nicotinamide N-methyltransferase (nicotinamide metabolism), major histocompatibility complex (MHC) class I and II (immune function), transforming growth factor (TGF)– beta and insulin-like growth factor I (growth factors), glycogen synthase and phosphoenolpyruvate carboxykinase (glucose metabolism), cytidine triphosphate (CTP) synthetase, medium-chain acyl-CoA dehydrogenase and triglyceride lipase (fatty acid metabolism), complement components C1q and C3 (complement activation),
Differential Allograft Gene Expression
817
Table 2. Genes Differentially Expressed in Acute Cellular Rejection Probe Set T-cell and B-cell activation K03430 HG2915 K01160 M28879 X00695 K02765 U25956 Apoptosis U37518 U69611 M59465 X02910 Growth factors X05839 X57025 M74587 Transcription factors/stress response M97935 M87503 M11717
Fold ⌬
Gene Name
13.8 13.1 13.0 12.9 12.8 12.4 12.1
Complement C1q MHC class I MHC class II Granzyme B Interleukin-2 Complement component C3 P-selectin
⬍.05 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01
14.9 13.0 12.4 12.1
TNF-related apoptosis inducing ligand TNF-alpha converting enzyme TNF-alpha inducible protein A20 TNF-alpha
⬍.01 ⬍.01 ⬍.01 ⬍.01
13.2 12.5 22.8
Transforming growth factor beta 1 Insulin-like growth factor I IGF-binding protein
⬍.05 ⬍.05 ⬍.01
13.1 12.3 12.4
Transcription factor ISGF-3 IFN-responsive transcription factor Heat shock protein 70
⬍.01 ⬍.01 ⬍.01
P Value
Note. Fold change indicates relative abundance of mRNA in allografts from HCV-infected patients with acute cellular rejection (ACR group) when compared with those with HCV and no rejection (HCV group). Abbreviations: TNF, tumor necrosis factor, IGF, insulin-like growth factor.
p-selectin (cell adhesion), tumor necrosis factor (TNF)–related apoptosis inducing ligand (TRAIL), TNF-alpha converting enzyme, TNF-alpha inducible protein A20 and TNF-alpha (apoptosis). A summary of these results is shown in Tables 2 and 3. Genes that were relatively underexpressed in the ACR group included: alanyl-tRNA synthetase, ribosomal protein-L8, elongation TU, protein synthesis factor eIF-4C, elongation factor-2, eukaryotic initiation factor-4AI and elongation factor-1 alpha (protein synthesis), chaperonin 10 and protein disulfide isomerase (protein folding), insulin-like growth factor (IGF)– binding protein (growth factor), GLUT-2 (glucose metabolism), very-long-chain acyl CoA dehydrogenase and fatty acid omega hydroxylase (fatty acid metabolism), MT-1 and glutathione peroxidase (DNA metabolism). These results are summarized in Tables 2 and 3.
Discussion In this study of HCV-infected liver transplant recipients with comparable histologic activity indices, we found that of 6412 genes surveyed, 25 were relatively overexpressed and 15 were relatively underexpressed during ACR when compared with recipients with
uncomplicated early recurrence of HCV infection. Our principal finding is thus that, although ACR and early recurrence of HCV may have similar clinical manifestations, they are quite distinct at the level of mRNA abundance. The results of these experiments suggest that recurrence of HCV is not, primarily, an immunemediated process. When interpreting the results of this type of study, it is important to bear in mind that differential expression will be secondary in nature in the great majority of instances. Furthermore, the results of high-density microarray studies provide insight into gene expression, not transcription nor posttranscriptional regulation. It is also likely that there are genes that were differentially expressed at levels that were beyond the sensitivity of the method used in this study. The information that this type of microarray study generates provides a relatively global view of the relative abundance of mRNA for many of the human genes of known function and thus clues, rather than answers, concerning the relative pathophysiology of ACR and recurrence of HCV. These caveats notwithstanding, ACR was most notably associated with the overexpression of genes associated with MHC I and II, apoptosis, and a distinct
818
Sreekumar et al
Table 3. Genes Differentially Expressed in Acute Cellular Rejection Probe Set Protein metabolism D83004 U49869 M58028 D32050 Z28407 X03689 L18960 X51466 D13748 J04617 U07550 Z49835 Glucose metabolism S70004 L05144 J03810 X52142 Fatty acid metabolism M91432 M29194 D43682 D13705 DNA/nicotinamide metabolism X76717 D00632
Fold ⌬
Gene Name
13.0 12.1 12.0 22.1 22.2 22.6 22.9 22.9 23.2 23.6 22.3 22.7
Ubiquitin-conjugating enzyme E2 Ubiquitin Ubiquitin-activating enzyme E1 Alanyl-tRNA synthetase Ribosomal protein L8 Elongation factor TU Protein synthesis factor (elF-4C) Elongation factor 2 Eukaryotic initiation factor 4AI Elongation factor EF-1-alpha Chaperonin 10 Protein disulfide isomerase
12.8 12.2 22.3 14.3
Glycogen synthase Phosphoenolpyruvate carboxykinase GLUT2 CTP synthetase
⬍0.01 ⬍0.01 ⬍0.01 ⬍0.01
13.0 12.4 22.3 22.3
Medium-chain acyl-CoA dehydrogenase Triglyceride lipase Very-long-chain acyl-CoA dehydrogense Fatty acids omega-hydroxylase
⬍0.05 ⬍0.01 ⬍0.01 ⬍0.01
24.1 25.3
MT-1 Glutathione peroxidase
⬍0.01 ⬍0.01
P Value ⬍.05 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.01 ⬍.05 ⬍.01 ⬍.05 ⬍.01 ⬍.01
Note. Fold change indicates relative abundance of mRNA in allografts from HCV-infected patients with acute cellular rejection (ACR group) when compared with those with HCV and no rejection (HCV group).
subset of T-cell activation genes. Interleukin 2 (IL-2) is an immunoregulatory lymphokine produced by lectinor antigen-activated T cells. IL-2 can act as a mitogen for both B- and T-lymphocytes and is known to be upregulated in ACR.15 Interferon-stimulated gene factor-3 (ISGF3), an interferon-dependent (␣ and ), positive-acting transcription factor that is cytoplasmically activated,16 was also relatively overexpressed in the ACR group.17 Expression of ISGF3 is known to be inhibited by the HCV NS5 protein.18 Because it is not known whether HCV RNA levels were different between the ACR and HCV groups, a mechanistic role of HCV in the differential expression of ISGF3 in the ACR group can only be speculated on. Granzyme B, a serine protease produced by cytotoxic T-lymphocytes and natural killer/lymphokine-activated cells, was also relatively overexpressed in the ACR group. Perforin and granzyme B act synergistically to trigger an endogenous apoptosis pathway resulting in the destruction of target cell nuclear membrane and DNA fragmentation.19 Gene transcript levels of granzyme B are known to be enhanced during ACR after liver and kidney transplan-
tation.20-23 Members of the tumor necrosis factor (TNF) family of cytokines and receptors are critically involved in cellular activation, proliferation, and cell death.23 Within this family, Fas ligand and TNF-related apoptosis-inducing ligand (TRAIL)24-26 have been shown to be mediators of apoptosis. Two further genes from the TNF superfamily, TNF-alpha converting enzyme and TNF-alpha inducible protein A20, were also relatively overexpressed in the ACR group. Transforming growth factor-beta 1, in addition to activating stellate cells, is proapoptotic.27,28 Expression of P-selectin, a 140-kD adhesion molecule that mediates the interaction of activated endothelial cells or platelets with leukocytes,29,30 was also increased in patients with ACR. Overexpression of this group of genes in the ACR group suggests that T-cell activation and apoptosis are mechanistically relatively more important in ACR than in recurrence of HCV. Because apoptosis is not thought to be pretranscriptionally regulated, it is likely that the relatively greater abundance of mRNA for Fas ligand and TRAIL in the ACR group is on the basis of differences in the relative abundance of lymphocyte cell types
Differential Allograft Gene Expression
within allograft cellular infiltrates (e.g., greater proportion of T than B lymphocytes in ACR allografts). The complement system plays a paradoxical role in the development and expression of autoimmunity and alloimmunity in humans.31 Transcript levels of genes for both C1q and C3 were relatively greater in the ACR group. C1q and C3 are known to be increased in acute cellular rejection.32 Subcomponent C1q binds to immunoglobulin complexes with resulting serial activation of C1r (enzyme), C1s (proenzyme), and the other eight components of complement. Complement component C3 is an acute phase reactant. Increased synthesis of C3 is also induced during acute inflammation.33 The liver is the main site of synthesis of C3, although small amounts are also produced by activated monocytes and macrophages. Overexpression of C1q and C3 are in keeping with enhanced complement activation during ACR. The relatively greater abundance of mRNA for immune activation genes in the ACR than HCV groups supports the concept that posttransplant recurrence of HCV infection is less immune mediated and may primarily involve a direct cytopathic effect of HCV infection, perhaps through diminished clearance of HCV from plasma and infected hepatocytes. This is in keeping with the clinical association of very high levels of HCV RNA with severe recurrence of HCV. It would be of substantial interest to compare hepatic gene expression in patients with posttransplant recurrence of HCV with that of patients with progressive HCV infection in the nontransplant setting to determine whether there are fundamental differences in the mechanisms of hepatic injury in these two settings. Whether the differential expression of immune activation genes in ACR and recurrence of HCV, as observed in these experiments, can be used to determine adequacy of immunosuppression or to differentiate ACR from recurrence of HCV in individual patients would require prospective studies. The possibility is, however, suggested by our results and those of previous reports of the utility of cytokine profiles in the diagnosis of acute cellular rejection.15 Increases in levels of IL-2 and expression of IL-2 and TNF-alpha receptors have been shown to precede biochemical and histologic abnormalities by several days in liver transplant recipients with acute cellular rejection.15 A similar finding has been reported in acute renal allograft rejection, in which expression of IL-10 and IL-15 also accompanies clinically apparent rejection.34 Others have reported the use of serum human lymphocyte antigen (HLA) class I antigens as a marker of acute rejection in human liver transplantations.35 Nonimmunomodulatory genes were also differen-
819
tially expressed during ACR. Gene transcript levels of ribosomal protein L8, protein synthesis factor eIF-4C, alanyl-tRNA synthetase, elongation factors TU, 2, and EF-1 alfa were all relatively underexpressed in the ACR group. Relative underexpression of these genes in the ACR group suggests a global impairment in liver protein synthesis at the initiation, elongation, and folding phases as well as stimulation of protein breakdown during ACR. Because immunosuppression type and levels were similar between the ACR and HCV groups, the differential expression of these genes is likely to have been secondary to events associated with rejection (e.g., diminished substrate availability because of endotheliitis-induced ischemia). Specific effects of lower transcript levels would be inhibition of 43S preinitiation complex formation (ribosomal protein L8 and eIF-4C), ribosomal translocation of tRNAs and mRNA (EF-TU and alanyl-tRNA synthetase), and protein translocation (EF-2). Chaperonin 10, which was also relatively underexpressed in the ACR group, is involved in ATPdependent protein folding and may have been downregulated secondary to diminished protein synthesis.36 Ubiquitin and ubiquitin activating enzyme E1 were relatively overexpressed in the ACR group. Enhanced protein ubiquitination, secondary to overexpression of ubiquitin and ubiquitin activating enzyme E1, would be expected to stimulate proteolysis.37,38 Differential expression of these genes may have been secondary to relatively greater local cytokine levels. Stimulation of proteolysis and inhibition of protein synthesis by cytokines (e.g., TGF-beta and TNF) has been well described.39-41 Heat shock protein 70, which is thought to be involved in protection of the cytoskeleton during stress,42 was relatively overexpressed in the ACR group, possibly reflecting higher oxidative stress in patients with ACR. The transcriptional activation of stress response gene (HSP 70) and decreased expression of chaperonin 10 and protein disulfide-isomerase that process damaged or misfolded proteins during the ACR suggest a central role for protein modifications associated with ACR. Transcript levels of genes coding for reactive oxygen species scavenging enzymes, metallothionine 1 and glutathione peroxidase, were relatively underexpressed in the ACR group. Both these genes play a protective role against oxidative damage by neutralizing reactive oxygen species.43-45 Removal of the superoxide radical by superoxide dismutate and hydrogen peroxide by glutathione peroxidase prevents formation of reactive hydroxyl radicals, which are postulated to be responsible for oxidative cellular injury. Although the mecha-
820
Sreekumar et al
nism is unclear, HCV infection is known to induce oxidative stress and DNA damage.46-48 Finally, three genes involved in glucose metabolism, glycogen synthase, phosphoenolpyruvate carboxykinase, and GLUT2, were differentially underexpressed in the ACR group. Glycogen synthase catalyzes the rate-limiting step in glycogen synthesis, whereas phosphoenolpyruvate carboxykinase is a key enzyme in gluconeogenesis. The gene transcript levels of both these genes were relatively higher in the ACR group. In contrast, GLUT2 (a low-affinity transporter of glucose present in the plasma membrane of hepatocytes) and insulin-like growth factor-1 (IGF-I) in hepatocytes, a stimulator of glycolytic metabolism through activation of the calcium calmodulin-dependent phosphatase calcineurin,49-52 were relatively overexpressed in the ACR group. The finding of relatively increased expression of IGF-1 among patients with ACR is of particular interest because IGF-1 stimulates calcineurin expression, opposing the effects of calcineurin inhibitors. In light of this observation, we are currently studying whether higher posttransplant IGF-1 levels are associated with increased risk of ACR in general. IGF-1 induces expression of the transcription factor GATA-2, which associates with calcineurin and a specific dephosphorylated isoform of the transcription factor NF-ATc1.50 Thus, IGF-1 induces calcineurin-mediated signaling. In summary, ACR and recurrence of HCV are associated with distinct mRNA expression patterns. In all likelihood, this reflects mechanistic differences in these two events. The differential expression of intrahepatic mRNA raises the possibility that gene expression, e.g., of immunomodulatory peptides, may be useful in the distinction of ACR from recurrence of HCV. In addition, these experiments provide evidence that alloimmunity, as indicated by expression of T-cell activation and apoptosis-inducing genes, is less important in recurrence of HCV than in ACR.
References 1. Charlton M, Seaberg E, Wiesner R, Everhart J, Zetterman R, Lake J, et al. Predictors of patient and graft survival after liver transplantation for hepatitis C. Hepatology 1998;28:823-830. 2. Sreekumar R, Gonzalez-Koch A, Maor-Kendler Y, Batts K, Moreno-Luna L, Poterucha J, et al. Early identification of recipients with progressive histologic recurrence of hepatitis C after liver transplantation. Hepatology 2000;32:1125-1130. 3. Berenguer M, Ferrell L, Watson J, Prieto M, Kim M, Rayon M, et al. HCV-related fibrosis progression after liver transplantation: increase in recent years. J Hepatol 2000;32:673-684. 4. Gane EJ, Naoumov NV, Qian KP, Mondelli MU, Maertens G, Portmann BC, et al. A longitudinal analysis of hepatitis C virus
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
replication after liver transplantation. Gastroenterology 1996; 110:167-177. Charlton M, Seaberg E. Impact of immunosuppression and acute rejection on recurrence of hepatitis C: Results of the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database. Liver Transpl 1999; 5(suppl 1):S107-S114. Rosen HR, Shackleton CR, Higa L, Gralnek IM, Farmer DA, McDiarmid SV, et al. Use of OKT3 is associated with early and severe recurrence of hepatitis C after liver transplantation [see comments]. Am J Gastroenterol 1997;92:1453-1457. Sheiner PA, Schwartz ME, Mor E, Schluger LK, Theise N, Kishikawa K, et al. Severe or multiple rejection episodes are associated with early recurrence of hepatitis C after orthotopic liver transplantation. Hepatology 1995;21:30-34. Demetris AJ, Seaberg EC, Batts KP, Ferrell LD, Ludwig J, Markin RS, et al. Reliability and predictive value of the National Institute of Diabetes and Digestive and Kidney Diseases Liver Transplantation Database nomenclature and grading system for cellular rejection of liver allografts. Hepatology 1995;21:408416. Si L, Whiteside TL, Van Thiel DH, Rabin BS. Lymphocyte subpopulations at the site of “piecemeal” necrosis in end stage chronic liver diseases and rejecting liver allografts in cyclosporine-treated patients. Lab Invest 1984;50:341-347. Lockhart DJ, Dong H, Byrne MC, Follettie MT, Gallo MV, Chee MS, et al. Expression monitoring by hybridization to highdensity oligonucleotide arrays. Nature Biotechnology 1996;14: 1675-1680. Fodor SP, Read JL, Pirrung MC, Stryer L, Lu AT, Solas D. Light-directed, spatially addressable parallel chemical synthesis. Science 1991;251:767-773. Knodell RG, Ishak KG, Black WC, Chen TS, Craig R, Kaplowitz N, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981;1:431-435. Batts KP, Ludwig J. Chronic hepatitis: An update on terminology and reporting. Surg Pathol 1995;19:1409-1417. Mahadevappa M, Warrington JA. A high-density probe array sample preparation method using 10- to 100-fold fewer cells. Nature Biotechnology 1999;17:1134-1136. Lowenthal JW, Zubler RH, Nabholz M, MacDonald HR. Similarities between interleukin-2 receptor number and affinity on activated B and T lymphocytes. Nature 1985;315:669-672. Levy DE, Kessler DS, Pine R, Darnell JE Jr. Cytoplasmic activation of ISGF3, the positive regulator of interferon-alpha-stimulated transcription, reconstituted in vitro. Genes Devel 1989; 3:1362-1371. Akalin E, Hendrix RC, Polavarapu RG, Pearson TC, Neylan JF, Larsen CP, et al. Gene expression analysis in human renal allograft biopsy samples using high-density oligoarray technology. Transplantation 2001;72:948-953. Polyak SJ, Khabar KS, Paschal DM, Ezelle HJ, Duverlie G, Barber GN, et al. Hepatitis C virus nonstructural 5A protein induces interleukin-8, leading to partial inhibition of the interferon-induced antiviral response. J Virol 2001;75:6095-6106. Trapani JA, Browne KA, Smyth MJ, Jans DA. Localization of granzyme B in the nucleus. A putative role in the mechanism of cytotoxic lymphocyte-mediated apoptosis. J Biol Chem 1996; 271:4127-4133. Krams SM, Villanueva JC, Quinn MB, Martinez OM. Expres-
Differential Allograft Gene Expression
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
sion of the cytotoxic T cell mediator granzyme B during liver allograft rejection. Transpl Immunol 1995;3:162-166. Tannapfel A, Kohlhaw K, Ebelt J, Hauss J, Liebert U, Berr F, et al. Apoptosis and the expression of Fas and Fas ligand (FasL) antigen in rejection and reinfection in liver allograft specimens. Transplantation 1999;67:1079-1083. Lipman ML, Stevens AC, Strom TB. Heightened intragraft CTL gene expression in acutely rejecting renal allografts. J Immunol 1994;152:5120-5127. Sharma VK, Bologa RM, Li B, Xu GP, Lagman M, Hiscock W, et al. Molecular executors of cell death— differential intrarenal expression of Fas ligand, Fas, granzyme B, and perforin during acute and/or chronic rejection of human renal allografts. Transplantation 1996;62:1860-1866. Smith CA, Farrah T, Goodwin RG. The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death. Cell 1994;76:959-962. Musgrave BL, Phu T, Butler JJ, Makrigiannis AP, Hoskin DW. Murine TRAIL (TNF-related apoptosis inducing ligand) expression induced by T cell activation is blocked by rapamycin, cyclosporin A, and inhibitors of phosphatidylinositol 3-kinase, protein kinase C, and protein tyrosine kinases: Evidence for TRAIL induction via the T cell receptor signaling pathway. Exp Cell Res 1999;252:96-103. Wu GS, Burns TF, McDonald ER III, Meng RD, Kao G, Muschel R, et al. Induction of the TRAIL receptor KILLER/DR5 in p53-dependent apoptosis but not growth arrest. Oncogene 1999;18:6411-6418. Ruuls SR, Sedgwick JD. Unlinking tumor necrosis factor biology from the major histocompatibility complex: lessons from human genetics and animal models. [Review] [38 refs]. Am J Hum Genet 1999;65:294-301. Fukuda K, Kojiro M, Chiu JF. Induction of apoptosis by transforming growth factor-beta 1 in the rat hepatoma cell line McARH7777: A possible association with tissue transglutaminase expression. Hepatology 1993;18:945-953. Singbartl K, Forlow SB, Ley K. Platelet, but not endothelial, P-selectin is critical for neutrophil-mediated acute postischemic renal failure. FASEB J 2001;15:2337-2344. Coughlan AF, Berndt MC, Dunlop LC, Hancock WW. In vivo studies of P-selectin and platelet activating factor during endotoxemia, accelerated allograft rejection, and discordant xenograft rejection. Transplant Proc 1993;25:2930-2931. Kawano M. Complement regulatory proteins and autoimmunity. [Review] [52 refs]. Archivum Immunologiae et Therapiae Experimentalis 2000;48:367-372. Pratt JR, Abe K, Miyazaki M, Zhou W, Sacks SH. In situ localization of C3 synthesis in experimental acute renal allograft rejection. Am J Pathol 2000;157:825-831. Barrington R, Zhang M, Fischer M, Carroll MC. The role of complement in inflammation and adaptive immunity [Review] [42 refs]. Immunol Rev 2001;180:5-15. McLean AG, Hughes D, Welsh KI, Gray DW, Roake J, Fuggle SV, et al. Patterns of graft infiltration and cytokine gene expression during the first 10 days of kidney transplantation. Transplantation 1997;63:374-380. Renna ME, Puppo F, Brenci S, Scudeletti M, Cinti P, Alfani D, et al. Serum HLA class I soluble antigens: A marker of acute rejection after liver transplantation. Transplant Proc 1995;27: 1155-1156. Todd MJ, Viitanen PV, Lorimer GH. Dynamics of the chapero-
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
821
nin ATPase cycle: Implications for facilitated protein folding [Review] [57 refs]. Science 1994;265:659-666. Zwickl P, Voges D, Baumeister W. The proteasome: A macromolecular assembly designed for controlled proteolysis [Review] [139 refs]. Philos Trans R Soc Lond Series B: Biological Sciences 1999;354:1501-1511. Ciechanover A. Ubiquitin-mediated degradation of cellular proteins: why destruction is essential for construction, and how it got from the test tube to the patient’s bed [Review] [40 refs]. Isr Med Assoc J: Imaj 2001;3:319-327. Charters Y, Grimble RF. Effect of recombinant human tumour necrosis factor alpha on protein synthesis in liver, skeletal muscle and skin of rats. Biochem J 1989;258:493-497. Llovera M, Garcia-Martinez C, Lopez-Soriano FJ, Argiles JM. The effect of chronic tumour necrosis factor-alpha treatment on urinary nitrogen excretion in the rat. Biochem Mol Biol Int 1994;33:681-689. Tayek JA. Effects of tumor necrosis factor alpha on skeletal muscle amino acid metabolism studied in-vivo. J Am Coll Nutr 1996;15:164-168. Wang TT, Chiang AS, Chu JJ, Cheng TJ, Chen TM, Lai YK. Concomitant alterations in distribution of 70 kDa heat shock proteins, cytoskeleton and organelles in heat shocked 9L cells. Int J Biochem Cell Biol 1998;30:745-759. Esposito LA, Kokoszka JE, Waymire KG, Cottrell B, MacGregor GR, Wallace DC. Mitochondrial oxidative stress in mice lacking the glutathione peroxidase-1 gene. Free Radic Biol Med 2000; 28:754-766. Gobel H, van der Wal AC, Teeling P, van der Loos CM, Becker AE. Metallothionein in human atherosclerotic lesions: a scavenger mechanism for reactive oxygen species in the plaque? Virchows Arch 2000;437:528-533. Ren Y, Smith A. Mechanism of metallothionein gene regulation by heme-hemopexin. Roles of protein kinase C, reactive oxygen species, and cis-acting elements. J Biol Chem 1995;270:2398823995. Cardin R, Saccoccio G, Masutti F, Bellentani S, Farinati F, Tiribelli C. DNA oxidative damage in leukocytes correlates with the severity of HCV-related liver disease: Validation in an open population study. J Hepatol 2001;34:587-592. Vendemiale G, Grattagliano I, Portincasa P, Serviddio G, Palasciamo G, Altomare E. Oxidative stress in symptom-free HCV carriers: relation with ALT flare-up. Eur J Clin Invest 2001;31: 54-63. Sumida Y, Nakashima T, Yoh T, Nakajima Y, Ishikawa H, Mitsuyoshi H, et al. Serum thioredoxin levels as an indicator of oxidative stress in patients with hepatitis C virus infection. J Hepatol 2000;33:616-622. Gooch JL, Tang Y, Ricono JM, Abboud HE. Insulin-like growth factor-I induces renal cell hypertrophy via a calcineurin-dependent mechanism. J Biol Chem 2001;276:42492-42500. Musaro A, McCullagh KJ, Naya FJ, Olson EN, Rosenthal N. IGF-1 induces skeletal myocyte hypertrophy through calcineurin in association with GATA-2 and NF-ATc1. Nature 1999;400:581-585. Pons S, Torres-Aleman I. Insulin-like growth factor-I stimulates dephosphorylation of ikappa B through the serine phosphatase calcineurin (protein phosphatase 2B). J Biol Chem 2000;275: 38620-38625. Semsarian C, Wu MJ, Ju YK, Marciniec T, Yeoh T, Allen DG, et al. Skeletal muscle hypertrophy is mediated by a Ca2⫹-dependent calcineurin signaling pathway. Nature 1999;400:576-581.