Hepatic expression of proteasome subunit alpha type-6 is upregulated during viral hepatitis and putatively regulates the expression of ISG15 ubiquitin-like modifier, a proviral host gene in hepatitis C virus infection.

The interferon-stimulated gene 15 (ISG15) plays an important role in the pathogenesis of hepatitis C virus (HCV) infection. ISG15-regulated proteins have previously been identified that putatively affect this proviral interaction. The present observational study aimed to elucidate the relation between ISG15 and these host factors during HCV infection. Transcriptomic and proteomic analyses were performed using liver samples of HCV-infected (n = 54) and uninfected (n = 10) or HBV-infected controls (n = 23). Primary human hepatocytes (PHH) were treated with Toll-like receptor ligands, interferons and kinase inhibitors. Expression of ISG15 and proteasome subunit alpha type-6 (PSMA6) was suppressed in subgenomic HCV replicon cell lines using specific siRNAs. Comparison of hepatic expression patterns revealed significantly increased signals for ISG15, IFIT1, HNRNPK and PSMA6 on the protein level as well as ISG15, IFIT1 and PSMA6 on the mRNA level in HCV-infected patients. In contrast to interferon-stimulated genes, PSMA6 expression occurred independent of HCV load and genotype. In PHH, the expression of ISG15 and PSMA6 was distinctly induced by poly(I:C), depending on IRF3 activation or PI3K/AKT signalling, respectively. Suppression of PSMA6 in HCV replicon cells led to significant induction of ISG15 expression, thus combined knock-down of both genes abrogated the antiviral effect induced by the separate suppression of ISG15. These data indicate that hepatic expression of PSMA6, which is upregulated during viral hepatitis, likely depends on TLR3 activation. PSMA6 affects the expression of immunoregulatory ISG15, a proviral factor in the pathogenesis of HCV infection. Therefore, the proteasome might be involved in the enigmatic interaction between ISG15 and HCV.

Performance characteristics of the VERSANT hepatitis C virus RNA 1.0 (kPCR) assay.

HCV RNA assays are of central importance for virological diagnostics and for clinical planning and monitoring of an antiviral combination treatment of chronic HCV infections. The objective of the pre-market evaluation of the VERSANT HCV RNA 1.0 Assay (kPCR) was to collect analytical performance data for this new method of HCV RNA quantification and to compare them with the high standards that exist in this context. The assay exhibited a specificity of 100%. The mean intra- and inter-assay imprecision was 14.1% and 14.6%, respectively. The detection limit was determined to be 16IU/ml (95% confidence interval: 11.9-30.6IU/ml) and consequently corresponded to the manufacturer's claims (i.e. 15IU/ml). The test exhibited linearity for all HCV genotypes in a broad range from 15 to 10(8)IU HCV RNA/ml. Hence, the kPCR assay in general is well suitable for HCV RNA determinations in clinical practice. However, in a methodological comparison, a considerable under-quantification of the concentrations of HCV genotype 2 and 3 isolates was detected. Provided that the assay's manufacturer will quickly remedy this shortcoming, the VERSANT HCV RNA 1.0 (kPCR) can be called a completely reliable technique for HCV RNA quantification in routine virological diagnostics.

Long-term stimulation of Toll-like receptor 3 in primary human hepatocytes leads to sensitization for antiviral responses induced by poly I:C treatment.

Chronic hepatitis C infection is associated with increased expression of interferon-sensitive genes (ISGs) in the liver, which is, paradoxically, correlated with the nonresponse to interferon (IFN)-based therapies. In the present study PHHs were isolated from HCV-infected or uninfected patients and stimulated with the TLR1-9 ligands for 6-24 h. Expression of cytokines and ISGs was determined by ELISA and qRT-PCR. A comparative analysis was performed for TLR3 signalling, which was also correlated with single nucleotide polymorphisms (SNPs) related to HCV pathogenesis. TLR-activated PHHs produced pro-inflammatory and anti-inflammatory cytokines, whereas IFNs were exclusively induced by TLR3 stimulation. Here, IL-29 and IL-28A were significantly highly expressed than IFN-α and IFN-ß. TLR3-induced IFN response was enhanced in hepatocytes isolated from patients with HCV infection. This hyper-responsiveness could be mimicked in naïve PHHs consistently stimulated with low dose of poly I:C, but not Guardiquimod. The higher responsiveness in PHH isolated from HCV-infected patients could be partially explained by higher frequencies of unfavourable SNP alleles of different SNPs associated with HCV progression and treatment outcome. These data suggest that durable activation of TLR3 but not TLR7, by low doses of viral replicative intermediates, increases the sensitivity to viral invasion. These findings shed new light on the relevance of TLR3 in the pathogenesis of HCV and may provide a possible explanation for the increased ISG expression during chronic HCV infection, the so-called IFN paradox.

Toll-like receptor-mediated immune responses are attenuated in the presence of high levels of hepatitis B virus surface antigen.

It has been recently shown that Toll-like receptor (TLR) signalling in murine nonparenchymal liver cells (NPCs) is suppressed in the presence of Hepatitis B virus surface antigen (HBsAg). It is not clear, however, whether this is also relevant for the adaptive immune responses and how this effect is mediated. Peripheral blood mononuclear cells (PBMCs) from Hepatitis B virus (HBV) patients and controls were stimulated by TLR ligands in the absence or presence of autologous serum. Interestingly, TLR-mediated cytokine expression (Interleukin-6 and -10) as well as TLR3-induced interferon (IFN) expression in PBMCs of HBV patients was significantly higher than in the healthy volunteers, showing a negative correlation with the levels of HBsAg. In addition, TLR3-mediated IFN-γ production was inhibited in the presence of HBV-containing serum. To mechanistically analyse this observation, murine Kupffer cells (KCs) and sinusoidal endothelial cells (LSECs) were stimulated with TLR3 ligands in the presence or absence of HBsAg. Mixed lymphocyte reactions were performed to study T-cell activation induced by TLR-stimulated NPCs. Gene expression of cytokines and TLR3 was analysed by quantitative rt-PCR, and activation of transcription factors was assessed by Western blot or reporter gene assays. TLR-induced expression of interferon γ, interferon sensitive genes and proinflammatory cytokines in murine KCs and LSECs was efficiently suppressed in the presence of HBsAg, whereas the expression of anti-inflammatory cytokines was enhanced. Activation of NFκB, IRF-3 and MAPKs in these liver cells was potently suppressed by HBsAg. T-cell activation mediated through TLR3-stimulated KCs or LSECs was suppressed by HBsAg which could be reverted by anti-IL-10 antibodies. These findings may, at least in part, explain how HBV evades innate and adaptive immune responses to maintain a persistent infection.

MicroRNA-155 controls Toll-like receptor 3- and hepatitis C virus-induced immune responses in the liver.

The hepatitis C virus (HCV) establishes persistent infections despite strong activation of the innate immune system through TLR3 and other sensors. Therefore, we analysed regulatory mechanisms of TLR3-induced immune responses in nonparenchymal liver cells (NPCs). Effects of Interleukin-10 (IL-10), transforming growth factor beta (TGF-ß) and immunoregulatory miR-155 on poly I:C-activated murine (C57BL/6) Kupffer cells (KC) and sinusoidal endothelial cells (LSEC) were assessed in vitro. NPCs were assayed for inflammatory and antiviral cytokines and T-cell (Balb/c)-activating factors. Gene expression of miR-155, IL-10, TGF-ß and interferon sensitive genes (ISGs) in biopsies of patients with HCV was determined by qrt-PCR. TLR3-induced antiviral activity in murine NPCs was potently suppressed by IL-10 and TGF-ß which correlated with decreased TLR3 expression and inhibition of NF-κB and IRF-3 activation. T-cell activation, induced by TLR3-activated NPCs, was also suppressed by IL-10 and TGF-ß, which was associated with a down-regulation of CD80 and CD86. Pretreatment with IL-10 or TGF-ß suppressed TLR3-induced miR-155 expression, which itself positively regulated poly I:C-mediated immune responses, thus counteracting IL-10 or TGF-ß-induced immunosuppression. In addition, hepatic expression of miR-155 was elevated in chronically infected patients with HCV, was associated with an IL-28B SNP (rs12979860) and was inversely correlated with HCV serum load and ISG expression levels. As miR-155 is a key regulator of anti-inflammatory mechanisms that control innate and adaptive hepatic immune responses during HCV infection, miR-155 based therapies may represent a novel mechanism to control HCV in the future.

CCL5 mRNA is a marker for early fibrosis in chronic hepatitis C and is regulated by interferon-α therapy and toll-like receptor 3 signalling.

Mechanisms causing liver fibrosis during chronic hepatitis C virus infection (cHCV) are not sufficiently understood. This study was aimed to identify biomarkers for early fibrosis (EF) and to investigate their potential role in cHCV-related fibrogenesis. To this end, peripheral whole blood (PB) samples from 36 patients with cHCV recruited from two independent cohorts were subjected to microarray analysis 12 h before initiation of peginterferon-alpha (Peg-IFN-α) and ribavirin therapy. Liver biopsies were evaluated using the Batts-Ludwig staging (BL-S) classification system for fibrosis. We showed that gene expression profiles (N = 8) distinguished between EF (BL-S: 0,1) and late fibrosis (LF; BL-S: 2,3,4) with 88.9% accuracy. Fibrosis-related functional annotations for chemokine-'C-C-motif'' ligand 5 (CCL5) provided foundation for focused investigation, and qRT-PCR confirmed that CCL5 mRNA levels (PB) reliably discriminate EF from LF (accuracy: 86.7%). Positive correlations (P < 0.05) with CCL5 mRNA levels and EF discovered gene expression profiles (PB) reflecting stable expression of IFN-α receptor 1, negative regulation of the MyD88-dependent toll-like receptor (TLR) pathway and decreased expression of TLR3 in vivo. Remarkably, Peg-IFN-α suppressed CCL5 mRNA levels (PB) in EF in vivo. These findings along with results from parallel in vitro investigation into the effect of IFN-α or poly I:C (TLR3-agonist) on CCL5 gene expression in hepatic stellate cells (HSC) attest to the multi-site involvement of these pathways in regulating fibrogenesis. In conclusion, we identified novel, reliable biomarkers for EF and exposed functional properties of the molecular network regulating CCL5 biosynthesis in peripheral or hepatic cell types with key roles in cHCV-related liver and/or immune pathogenesis.

Tri-iodothyronine as a stimulator of liver regeneration after partial and subtotal hepatectomy.

BACKGROUND: Tri-iodothyronine (T3) has been shown to be a hepatic mitogen. We investigated whether exogenous application of T3 improves liver regeneration after 70% partial hepatectomy (PH) and confers a survival advantage after 90% subtotal hepatectomy (SH) in rats and whether this is associated with the stimulation of angiogenesis. METHODS: Rats were subjected to PH or SH 10 days after injection of a single dose of T3. Liver body weight ratio (LBR), hepatic proliferation (Ki-67), biochemical markers as well as vascular endothelial growth factor (VEGF) expression were assessed by immunohistochemistry. Gene expression of pathogenic relevant genes was determined by customized cDNA arrays and quantitative RT-PCR. RESULTS: T3-treated rats showed an increased LBR and Ki-67 index after PH and SH, which reached statistical significance compared to placebo-treated rats (p < 0.05). On the transcriptional level, T3-treated rats had an increased expression of VEGF as demonstrated by immunohistochemistry, which was associated with a higher expression of its receptor Flt-1. CONCLUSIONS: Exogenous administration of T3 ameliorates liver regeneration after 70% PH and 90% SH, possibly due to stimulation of angiogenesis. Therefore, its clinical use might be of interest due to its excellent general practicability.

Selection of hepatitis B virus variants with aminoacid substitutions inside the core antigen during interferon-alpha therapy.

The hepatitis B virus (HBV) core antigen carries many epitopes relevant for B and T cell response that show aminoacid variation during viral infection. In a longitudinal analysis, sequential serum samples of 15 patients that suffered from chronic HBV infection were collected before, during, and after high-dose IFN-alpha treatment. The HBV preCore/Core (preC/C) sequence of the selected samples in each patient was determined and analysed for sequence variations compared to the pretreatment sample. The positions of HBV core aminoacid substitutions were assigned to immunodominant B, CD4(+) and CD8(+) cell epitopes. Seventy-five percent of all aminoacid substitutions were found within immunodominant T and B cell epitopes (12.5% were inside known HBV core mutation cluster regions) that show an increased number of clustered aminoacid substitutions during chronic HBV infection and overlap partially with the immunodominant epitopes. Only 12.5% of the detected core antigen aminoacid substitutions could not be assigned to any epitope or mutation cluster region. Stable HBV core antigen aminoacid substitutions, which were found between pretreatment sequence and the last sequence analysed during therapy, were found most frequently inside T helper cell epitopes. This longitudinal analysis of aminoacid substitutions inside the HBV core antigen in patients with chronic HBV infection shows that core aminoacid variations occur most frequently inside immunodominant HBV core epitopes, possibly due to an immuneselective pressure of the host against the virus. The data also suggest that stable HBV variants with aminoacid substitutions in immunodominant core epitopes can be selected during high-dose IFN-alpha therapy and persist after the end of treatment.

Performance of the COBAS AMPLICOR HCV MONITOR test, version 2.0, an automated reverse transcription-PCR quantitative system for hepatitis C virus load determination.

A clinical evaluation of an automated quantitative PCR assay, the COBAS AMPLICOR HCV MONITOR test, version 2.0 (v2.0), was carried out to assess the performance of this test in comparison with that of the previous, manual version, the AMPLICOR HCV MONITOR test, and with that of nested PCR. Serial dilutions of serum samples infected with genotype 1b, 2a, or 3, as well as synthetic RNA transcripts and serum samples derived from 87 patients with chronic hepatitis C and infected with genotype 1a, 1b, 2a, 2b, 3a, 3b, 4, or 5, were analyzed to determine the ability of the system to efficiently quantify various hepatitis C virus (HCV) genotypes. These experiments showed that the COBAS AMPLICOR HCV MONITOR test, v2.0, has mean intra-assay, interassay, and interoperator coefficients of variation that range from 22 to 34.5% and a 3-logarithm dynamic range, which spans from 10(3) to 10(6) copies/ml. Compared to the previous, manual version of the test, the COBAS AMPLICOR HCV MONITOR test, v2.0, showed an improved efficacy for all genotypes, especially genotypes 2, 3, and 4, whose estimated concentrations were on average 1 logarithm higher. When used to monitor patients under treatment, however, both versions showed the same patterns of viremia, indicating that the COBAS AMPLICOR HCV MONITOR test, v2.0, and the AMPLICOR HCV MONITOR test were equally effective at detecting relative viremia changes in serial samples. As expected, the automated test was less sensitive than nested PCR; among specimens from a cohort of patients treated with interferon, nested PCR identified three more viremic specimens, which probably contained very low concentrations of HCV RNA.

HBV viral load within subpopulations of peripheral blood mononuclear cells in HBV infection using limiting dilution PCR.

Extrahepatic viral load in peripheral blood mononuclear cells (PBMCs) of patients with hepatitis B virus (HBV) is still under debate. In this study, HBV infection rates and viral titers were examined within all PBMC subpopulations using limiting dilution-PCR (LD-PCR). PBMCs of patients with acute or chronic hepatitis B were separated by magnetic beads in monocytes, B-cells, CD4+ T-cells, CD8+ T-cells, and NK cells. Using two-round nested PCR, HBV-DNA sequences were detected in all patients examined within each PBMC subpopulation. The frequencies of HBV-positive cells and viral loads were calculated by Poisson analysis of HBV PCR results from serial dilutions of cells and cell lysates. Highest infection rates were found in monocytes and B-cells followed by CD8 + T-cells, NK cells, and CD4+ T-cells. Concerning all subsets, frequencies of HBV-positive cells were 50- to 500-fold higher in chronic than in acute hepatitis B. Viral loads were mostly estimated at about one HBV genome per HBV-positive cell. Moreover, slightly elevated HBV titers were seen in B-cells, CD4+ T-cells, and NK cells in both acute and chronic hepatitis B. It was demonstrated that beside a generally more latent HBV infection in PBMCs, elevated HBV titers point to replication or selective viral uptake within particular PBMC subsets. Therefore, the data suggest that HBV-infected PBMCs may participate in persistence of HBV.

Clinical evaluation and applications of the Amplicor HBV Monitor test, a quantitative HBV DNA PCR assay.

Viral load has emerged recently as a reliable marker of disease progression and therapeutic efficacy in chronic infections, including AIDS and hepatitis C. The clinical management of type B hepatitis could also be improved by monitoring viremia levels in patients with chronic liver disease undergoing anti-viral treatment. To address this question we evaluated the performance of a newly developed, quantitative PCR assay (Amplicor HBV Monitor test, Roche Diagnostic Systems) in the assessment of viremia changes over time in a group of 45 patients with chronic active hepatitis (CAH) who received interferon treatment. Of the 45 patients, 14 were HBsAg and anti-HBeAg positive and 31 HBsAg, HBeAg positive. Follow-up extended up to 24 months. An average of ten samples per patient were analyzed for levels of ALT, IgM anti-HBc (Abbott Laboratories), HBV DNA by in-house dot-blot hybridization and hybridization-capture assays (HBV-DNA hybrid capture kit, Murex Diagnostics) and by Amplicor HBV Monitor. A sustained biochemical response was observed at the end of treatment in 12 HBeAg-positive and in seven anti-HBeAg positive patients. This was accompanied by the disappearance of HBeAg and of HBV DNA (hybridization assays) in all cases and by the clearance of IgM anti-HBc in 70% of the cases. Viremia (quantitative PCR assay) became undetectable only in 25-30% of cases and was associated with the loss of HBsAg. A good correlation was observed between the time course of IgM anti-HBc, quantitative PCR and dot-blot hybridization although the latter missed 33% of viremic samples. Together, these results indicate that the Amplicor HBV Monitor test is a robust and standardized assay for quantifying HBV viremia levels in the range from 10(2) to 10(7) copies/ml. Compared to other current markers, viral load may provide additional clinical information by predicting long term virologic response and HBsAg clearance in patients with normal ALT at the end of interferon therapy.

Interleukin-10 expression is autoregulated at the transcriptional level in human and murine Kupffer cells.

Interleukin 10 (IL-10) is known to downregulate immune responses. The regulation of IL-10 gene expression therefore determines the outcome of local immune reactions. We investigated time course and downregulation of IL-10 production in primary Kupffer's cells (KC), which are known to secrete IL-10 in response to endotoxin challenge. Human and murine KC were isolated by centrifugal elutriation and investigated for IL-10 gene expression by a two-step amplification procedure (reverse transcriptase-polymerase chain reaction [PCR] followed by T7-polymerase chain reaction). We show that IL-10 messenger ribonucleic acid (mRNA) showed a >450 fold increase in KC 2 hours after endotoxin challenge. IL-10 protein release from KC strictly depended on de novo protein synthesis. Endotoxin mediated increase in IL-10 gene expression was downregulated by exogenous (>350-fold reduction of IL-10 mRNA level), as well as endogenous IL-10 protein, showing a negative autoregulatory feedback loop. IL-10 receptor expression was found to be constitutive and functional in KC. Early expression of IL-10 in KC may be of functional relevance to the outcome of immune and inflammatory reactions in the liver sinusoid. The negative autoregulation of IL-10 expression may represent a mechanism to regain a state of functional responsiveness in the microenvironment towards new proinflammatory stimuli. In conclusion, autoregulatory downregulation of IL-10 expression in KC may account for important regulatory steps of local immune response in the liver sinusoid.

Chronic hepatitis C: the viral load per liver cell before treatment as a new marker to predict long-term response to IFN-alpha therapy.

BACKGROUND/AIMS: So far, there are no reliable parameters that can predict the long-term sustained response to treatment with interferon-alpha in patients with chronic hepatitis C. In this study, we have developed a semi-quantitative method to determine the viral load per liver cell and have correlated this factor with the outcome of hepatitis C patients treated with interferon-alpha. METHODS: Hepatitis C virus RNA levels were measured in serum, peripheral blood mononuclear cells and liver cells of randomly chosen hepatitis C patients before treatment with interferon-alpha (n=37). The number of cells present in the liver biopsies was determined by a polymerase chain reaction-based quantitation of the housekeeping gene beta-globin. The patients were divided into a responder ("R", n=15, 41%) and a non-responder ("NR", n=22, 59%) group, as defined by normal liver enzymes and negative hepatitis C virus-polymerase chain reaction 6 months after treatment. RESULTS: Long-term sustained responders had a significantly lower viral load per liver cell (median: 5 vs. 650 copies/1000 liver cells, p-value: 0.0001), lower age (median: 32 vs. 54 years, p-value: 0.006) and lower percentage of geno- or serotype 1 (46% vs. 91%). Regarding viral load in serum and peripheral blood mononuclear cells, alanine aminotransferase levels, gamma-globulin levels and histological changes, no statistically significant differences were observed. CONCLUSIONS: In chronic hepatitis C infection, a high viral load per liver cell represents a new marker to predict long-term response to therapy with IFN-alpha.

Ligase chain reaction (LCR) assay for semi-quantitative detection of HBV DNA in mononuclear leukocytes of patients with chronic hepatitis B.

A ligase chain reaction (LCR)-based approach to detect hepatitis B virus (HBV) DNA in peripheral blood mononuclear cells (PBMC) is described. Using this new amplification technique, we determined semi-quantitatively the amount of a short HBV S-gene fragment in PBMC lysates of 25 patients with different forms of chronic hepatitis (group A (n = 8), hepatitis B s antigen (HBsAg)+/hepatitis B e antigen (HBeAg)+; group B (n = 9), HBsAg+/HBeAg-; group C (n = 8), HBsAg-/HBeAg-). The LCR results were compared with the findings obtained with polymerase chain reaction (PCR) amplification of three distinct HBV gene regions (preS1/2, S and C) and related to the serological profiles of the patients. Depending on the primer pair used for PCR amplification, sensitivity of HBV LCR in PBMC was equivalent or slightly superior to PCR. The highest positivity rate for HBV DNA was observed in the HBeAg+ and HBV DNA seropositive group (8/8) and was lower in the other patient groups B (4/9) and C (1/8). Interestingly, HBV gene sequences could also be detected in the lymphocytes of an HBsAg negative patient and in two patients from group B who were both negative for serum viral particles by PCR. The rapid LCR procedure represents a reliable alternative to PCR for the sensitive detection of HBV DNA in PBMC samples. In combination with the automated IMx(TM)-system the new amplification technique may be routinely used for screening for HBV in whole blood samples and thus may help to better evaluate the risk of HBV reinfection in liver transplant recipients.

Rare pre-core stop-codon mutant nt. 1897 predominates over wide-spread mutant nt. 1896 in an unusual course of chronic hepatitis B.

We present a patient with an unusual course of hepatitis B e antigen (HBeAg)-negative chronic hepatitis B who had repeated reactivations of his disease progressing to cirrhosis with terminal liver failure. Each flare up presented like an acute hepatitis with very high titres of hepatitis B virus (HBV) and high inflammatory activity followed by rapid clearance of viraemia. The pre-core genome of HBV isolated from sera during 5 years of follow up was analysed. Direct sequencing of polymerase chain reaction (PCR) products derived from consecutive sera showed a rare pre-core stop-codon mutation at nucleotide (nt.) 1897 G --> A with an accompanying mutation nt. 1857 C --> T as well as a stop-codon mutation nt. 1896 G --> A. By cloning and sequencing of PCR products the mutant strain with mutation nt. 1897 was shown to predominate over viral strains with a mutation nt. 1896 during the course of disease, although the stop-codon mutation nt. 1896 in general is observed more frequently. Both mutations allow viral replication by stabilizing the encapsidation signal 'epsilon'. This allowed HBV replication at a very high level as observed during flare ups. The absence of HBeAg may be responsible for the massive cytotoxic T-cell response towards hepatocytes which might explain the rapid progression to liver cirrhosis although no, or very little, HBV replication was observed for long periods. However, there is no clear explanation as to why the nt. 1897 mutant strain overwhelmed the other virus strains.

Posttransfusional, LKM-1-autoantibody-positive hepatitis C virus infection, cryoglobulinemia, and aplastic anemia.

Aplastic anemia is occasionally caused by viral hepatitis, hepatitis C virus being the most important factor. Pathogenetically, decreased bone marrow function, abnormalities of the bone marrow microenvironment, and immune-mediated suppression of hematopoiesis are important. Hepatitis C virus infection is associated with a variety of extrahepatic manifestations including autoimmune features like cryoglobulinemia, Sjögren's syndrome, and autoimmune hepatitis. Here we report the case of a 42-year-old man with aplastic anemia due to posttransfusional hepatitis C virus infection associated with cryoglobulinemia and LKM-1 autoantibodies. Following a triple immunosuppressive therapy, there was a complete reconstitution of the bone marrow. Serum HCV-RNA as well as plus- and minus-stranded HCV-RNA in peripheral blood mononuclear cells (PBMC) were detected before immunosuppressive therapy. After therapy, serum HCV-RNA persisted. Furthermore, PBMC now were positive for plus-stranded RNA only. However, in bone marrow-derived precursor cells we failed to demonstrate HCV molecules after therapy. This would argue for reconstituted PBMC from newly generated uninfected precursor cells. It remains unclear as to whether the autoimmune character of the disease or the hepatitis C virus infection itself have contributed to the pathogenesis of the aplastic anemia.