key: cord-0721603-8lpb5lo6
authors: Larrea, E; Garcia, N; Qian, C; Civeira, M P; Prieto, J
title: Tumor necrosis factor α gene expression and the response to interferon in chronic hepatitis C
date: 2003-12-30
journal: Hepatology
DOI: 10.1002/hep.510230203
sha: fcd5ccb7b4bdbcf613e747604e2a941583b738c1
doc_id: 721603
cord_uid: 8lpb5lo6

Tumor necrosis factor α (TNF‐α) is a cytokine with pleiotropic properties that is induced in a variety of pathological situations including viral infections. In this work, we analyzed the expression of TNF‐α gene in patients with chronic hepatitis C. Serum TNF‐α levels were found to be elevated in all chronic hepatitis C patients including those cases presenting sustained biochemical remission of the disease after interferon therapy. Untreated patients with chronic hepatitis C showed increased TNF‐α messenger RNA (mRNA) levels in the liver and mononuclear cells as compared with healthy controls. After completion of treatment with interferon, patients experiencing sustained complete response showed values of TNF‐α mRNA, both in the liver and in peripheral mononuclear cells, within the normal range, significantly lower than patients who did not respond to interferon and than those with complete response who relapsed after interferon withdrawal. Pretreatment values of TNF‐α mRNA were lower in long‐term responders to interferon than in cases who failed to respond to the treatment. Values of TNF‐α mRNA in the liver or in mononuclear cells were higher in specimens with positive hepatitis C virus (HCV) RNA than in those samples where the virus was undetectable. Neither the intensity of the liver damage nor the amount of HCV RNA in serum or in cells showed correlation with the levels of TNF‐α transcripts in peripheral mononuclear cells but it was found that high TNF‐α values were associated with genotype 1b. In conclusion, there is an enhanced expression of TNF‐α in HCV infection. High levels of this cytokine may play a role in the resistance to interferon therapy.

50 years) and 47 patients with chronic hepatitis C (27 male dles (Baxter, Deerfield, IL). One third of the specimen was and 20 female; age range, 21 to 75 years). This group of immediately frozen in liquid nitrogen and stored at 080ЊC patients included 16 untreated patients and 31 patients who until the extraction of RNA, and the remaining was used for were studied after completion of treatment with IFN-a (3 histological study. MU daily for 2 months and 3 MU 3 times a week for 8 to 10

Analysis of Serum TNF-a Levels. To analyze the serum conmonths); of these, 15 did not respond to the therapy and 16 centration of TNF-a we used an ELISA (Cytoscreen; Bioexhibited sustained biochemical remission of the disease for source International, Camarillo, CA) using specific anti-humore than 6 months after stopping IFN-a administration. man TNF-a antibodies. The assay was performed following In addition we analyzed TNF-a mRNA levels in 71 PBMC the instructions of the manufacturers. The sensitivity limit samples from 61 patients with chronic hepatitis C (44 male of the assay was 1 pg/mL. and 17 female; age range, 21 to 72 years) and 17 healthy Analysis of TNF-a mRNA Levels. To analyze TNF-a mRNA controls (11 male and 6 female; age range, 25 to 38 years). levels we used a procedure based on the reverse transcrip-Liver biopsy results showed chronic active hepatitis in all of tion-polymerase chain reaction (RT-PCR). Either PBMC or them, and cirrhosis was present in 4 cases. Histological activliver total RNA (0.75 mg) was reverse-transcribed (60 minutes ity index in liver samples was determined according to Knoat 37ЊC) with 250 U of M-MuLV reverse transcriptase (BRL, dell's scoring system. 24 Twenty patients had not received Gaithersburg, MD) in 25 mL volume of 51 RT buffer (0.25 treatment (untreated group [UT]), and 36 were studied at mol/L Tris-HCl pH 8.3, 0.375 mol/L KCl, 15 mmol/L MgCl 2 ), the end of 10 to 12 months of IFN-a therapy. These patients supplemented with 5 mmol/L dithiothreitol, 0.4 mmol/L deoxwere classified according to the response to IFN-a into three ynucleoside triphosphate, ribonuclease inhibitor (30 U), and groups: sustained response (SR), response with relapse (Rr), random hexamers (250 ng) (Boehringer Mannheim, Mannand no response (NR). The SR group (n Å 12) included all heim, Germany). After heating (95ЊC, 5 min) and quick-chillpatients with normal alanine aminotransferase (ALT) values ing on ice, 6.7-or 10-mL aliquots of the complementary DNA at the end of therapy with persisting normal values during (cDNA) pool (PBMC or liver respectively) were used for PCR more than 1 year of follow-up. The NR group (n Å 11) included amplification in 50 ml of 101 PCR buffer (100 mmol/L Tristhose patients with elevated ALT levels at the end of therapy. HCl pH 9.3, 500 mmol/L KCl, and 1% Triton X-100) con-The Rr group (n Å 13) included all patients with normal ALT taining 0.02 mmol/L deoxynucleoside triphosphate, 50 mCi/ levels at the end of treatment in whom serum ALT levels mL of (a-32 P)-deoxyeitidine triphosphate (Amersham, Buckincreased after interferon withdrawal. In 5 patients, the inghamshire, England), upstream and downstream primers study was performed on two occasions, before initiation the (40 ng each), 1.5 mmol/L MgCl 2 and 2 U of Taq DNA polymertreatment with IFN-a and at the end of the therapy. All these ase (Promega Corporation, MD). Blank reactions with no cases showed a sustained response to the treatment. Fifteen RNA were performed in all experiments. PCR amplifications additional patients showing persistent normal serum ALT of a fragment of b-actin cDNA (using 10 mL of the cDNA levels after IFN-a therapy were studied 4 to 15 months after pool) were also performed as internal control for each sample. completion of the treatment (these cases are referred to as the TNF-a cDNA fragments were amplified by 25 or 29 cycles SR-PT group). Five of these patients had also been studied at (PBMC or liver, respectively) (94ЊC, 59ЊC, and 72ЊC, 1 minute the end of treatment. each step), and b-actin was amplified by 20 cycles (94ЊC, Samples from liver biopsies from 35 patients (26 males and 55ЊC, and 72ЊC, 1 minute each step) a protocol that avoided 9 females; age range, 16 to 71 years) with chronic hepatitis C interference of the plateau effect. Oligonucleotides (5-3) (4 with cirrhosis) were studied. This group of patients (of d(GTCAGATCATCTTCTCGAACC) and d(CAGATAGATGGwhom 14 also had PBMC analyzed) included 10 UT and 25 GCTCATACC) were the upstream and downstream primers, treated subjects in whom liver biopsy specimens were obrespectively, used for amplification of 360-base-pair (bp) tained at the end of 12 months of therapy (8 NR, 9 SR, and fragment (nucleotides 315-673) from human TNF-a cDNA. 26 8 Rr). Normal liver samples were obtained from 6 controls (1 d(TCTACAATGAGCTGCGTGTG) and d(GGTGAGGATCTTmale and 5 female; age range, 41 to 66 years) at laparotomy CATGAGGT) were the primers used to amplify a 314-bp frag-(in 5 cases because of gallstones and in 1 because of colon ment from human b-actin cDNA that is located between nucancer without apparent liver metastasis).

cleotides 1319-2079 in the reported human b-actin gene All patients included in the present study were anti-HCVsequence. 27 After PCR amplifications, 20 mL aliquots of the positive using enzyme-linked immunosorbent assay (ELISA) PCR reactions were electrophoresed in 2% agarose gel and (Ortho Diagnostic Systems, second-generation, Raritan, NJ) bands were visualized by ethidium bromide. Equal size bands and all of them were or had been (before treatment) serum were excised and radioactivities were determined. Obtained HCV RNA positive using the reverse transcription polymervalues were corrected with background radioactivity from ase chain reaction. Other causes of chronic hepatitis were blank reactions with no RNA. Finally, values corresponding excluded.

to TNF-a mRNA were normalized to those of b-actin mRNA. Preparation of PBMC, Serum, and Liver Samples. For iso-Results are expressed as counts-per-minute (cpm) ratio of lation of PBMC, fresh blood diluted with 1 volume of 0.9% TNF-a to b-actin. When TNF-a mRNA values were normal-NaCl was overlaid on Lymphopred (Nycomed Pharma AS, ized to those of b-actin mRNA, it was assumed that b-actin Oslo, Norway) with a density of 1.077 g/mL and centrifuged mRNA was constantly expressed in patients with chronic 600g for 30 minutes. The mononuclear cell layer was harhepatitis C, both in patients who did not receive treatment vested at the interphase and washed twice in 0.9% NaCl. and in those treated with IFN-a. To confirm this, we analyzed PBMC were lysed in 4 mol/L guanidinium thiocyanate and simultaneously 15 PBMC and 15 liver samples (5 healthy stored at 080ЊC until extraction of total RNA, which was controls, 5 untreated patients, and 5 treated patients). No performed according to the method of Chomczynski and differences in the cpm values corresponding to b-actin mRNA Sacchi. 25 were observed between the groups. In PBMC samples, cpm To obtain serum samples, venous blood was collected into values were 14808. 8 mL of the first PCR product was amplified (30 cycles) using inner primers (200 ng each) and the same conditions as in the first PCR. After amplification, 15-mL aliquots of the final PCR reaction were electrophoresed in 2% agarose gel and the bands were visualized by ethidium bromide staining. The sequences for oligonucleotide primers were chosen from five noncoding regions of the HCV genome. 28 The primers for HCV cDNA synthesis and PCR amplification were: the outer sense, GTATCTCGAGGCGACACTCCACCATAGAT, and outer antisense, ATACTCGAGGTGCACGGTCTACGAGACCT; and inner sense, CCACCATAGATCTCTCCCCTGT, and inner antisense, CACTCTCGAGCACCCTATCAGGCAGT. 28a For HCV RNA analysis in serum, 50 mL aliquots were extracted. HCV RNA was reverse transcribed with random primers and PCR was performed using the above conditions. HCV RNA was quantified by a competitive PCR technique. Briefly, RNA from 200 mL of serum or 1.5 mg of total RNA from PBMC was reverse transcribed with random primers in Validation experiments of PCR assays using known quantities of total RNA (from 0 to 1 mg) were performed (Fig. 1) .

The cpm values corresponding to the used amount of either TNF-a or b-actin mRNA (0.2 and 0.3 mg for TNF-a in PBMC and liver, respectively, and 0.3 mg for b-actin in both) were in the linear range of the curve. To analyze whether changes of TNF-a mRNA levels relative to those of b-actin resulted in changes of the ratio of TNF-a PCR product cpm/b-actin PCR product cpm, we performed a RT-PCR with 0.1 or 0.3 mg of total PBMC RNA from a healthy control (for TNF-a or b-actin, respectively) and added serial dilutions of known quantities of TNF-a riboprobe (from 25 pg to 0.19 pg). TNF-a riboprobe was obtained by run-off transcription of a plasmid which contained a 760-bp insert (nucleotides 156-916) from the human TNF-a cDNA. 26 As shown in Fig. 2 , with increasing amounts of TNF-a mRNA used in the reaction, we obtained a progressive increase of the TNF-a/b-actin ratio (the increase was linear up to 1.56 pg of the TNF-a riboprobe).

The coefficient of interassay variation for TNF-a/b-actin was 13.58%. The identity of the PCR products from TNF-a cDNA amplification was further verified, first, by Southern blot analysis, using a-32 P-labelled PstI/BamHI fragment (1.6 kb-long) of TNF-a cDNA 26 as hybridization probe and, second, by digestion with BglII, PvuII and EcoRI. The two former enzymes yielded the predicted restriction fragments while EcoRI did not digest the amplified PCR product (no restriction site for this endonuclease is present in the amplified region). 

feron Treatment. Serum TNF-a levels were higher in untreated patients with chronic hepatitis C (166.28 { 33.21 pg/mL) and in those who failed to respond to IFN after 10 to 12 months of treatment (163.05 { 36.72 pg/mL) than in healthy controls (19.28 { 5.20 pg/mL; P õ .001). Patients who made a complete and sustained biochemical response to IFN-a therapy also showed se- FIG 3 . Transcriptional expression of TNF-a gene (as a ratio to rum TNF-a levels (117.14 { 24.46 pg/mL) significantly that of b-actin) in PBMC in healthy controls (C) and patients with higher than control values (P õ .001). Although in this chronic hepatitis C without treatment (UT) or after 10 to 12 months of interferon alfa treatment. NR, patients with no response to intergroup of patients serum TNF-a tended to be lower than feron studied at the end of the treatment; Rr, patients with complete in nonresponders to IFN-a, this difference did not atresponse but with post-therapy relapse studied at the end of the tain statistical significance. treatment; SR, patients with complete response without post-therapy TNF-a mRNA levels in UT, NR, and Rr patients were relapse (sustained response) studied at the end of the treatment; SR-PT, patients with sustained response studied 4 to 15 months after higher than in healthy controls both in PBMC (Fig. 3) completion of interferon treatment. *P õ .01 vs. UT or NR or Rr; and in the liver (Fig. 4) . In contrast, in SR patients, **P õ .05 vs. SR or SR-PT; ***P õ .05 vs. SR.

TNF-a mRNA values in PMBC and in liver tissue were similar to controls and significantly lower than in UT, NR, and Rr patients. No differences in hepatic and 40 mL or 60 mL of volume, respectively. After reverse tran-PBMC TNF-a mRNA levels were observed among these scription, 5 mL of cDNA from serum or 10 mL of cDNA from last three groups of patients. TNF-a gene expression PBMC was mixed with serial dilutions of a competitor DNA in PBMC in SR-PT group was similar to that of the SR (6 1 10 5 , 6 1 10 4 and 6 1 10 3 molecules for serum, and 3 1 10 5 , 6 1 10 4 , 3 1 10 4 , and 6 1 10 3 for PBMC) that differed group (Fig. 3) .

from the cDNA of interest by a small delection (45 bases).

Fifteen patients from the UT group underwent IFN-This competitor DNA was produced from cloned DNA from a therapy. Of these, 5 experienced sustained response 5 nonconding region of one HCV-infected patient. Samples to the treatment, and 10 did not respond to IFN-a. were amplified by nested PCR with two sets of primers (5-Pretreatment levels of TNF-a mRNA in PBMC were CCTGTGAGGAACTACTGTCT-3 and 5-CTATCAGGCAGTsignificantly higher in the patients who failed to re-ACCACAAG-3 for outer primers; 5-ACTGTCTTCACGCAGspond to IFN-a than in those patients who exhibited a AAAGC-3 and 5-GACCCAACACTACTCGGCTA-3 for insustained response to the therapy (0.75 { 0.1 vs. 0.28 ner primers). For both PCRs, 30 cycles were performed as follows: denaturation for 15 seconds at 95ЊC, annealing for 12 seconds at 52ЊC, and extension for 20 seconds at 72ЊC with a final extension for 1 minute at 72ЊC, in a Perkin Elmer (Norwalk, CT) Gene Amp PCR System 9600. First PCR were performed in 40-mL mixtures and 2 mL of the first PCR product was transferred to the second PCR mixture performed in 20 mL, and containing 25 mCi/mL of (a-32 P)-deoxycitidine triphosphate. After second PCR amplifications, 15-mL aliquots of the PCR reactions were electrophoresed in 2.5% agarose gel, and bands were visualized by ethidium bromide. Equal-sized bands from cDNA and competitor DNA were excised and radioactivities were determined and compared. The cpm corresponding to competitor DNA was corrected by the number of citosines in which competitor DNA differs from the studied cDNA. The copy number of HCV RNA in the samples is considered equal to the copy number of the competitor when the cpm of the studied cDNA/cpm of the competitor is equal to 1.

HCV Genotypes. Genotyping was performed by means of a hybridization procedure using specific probes for HCV genotypes 1a, 1b, 2a, 2b, and 3a, according to Simmonds et al., 29 FIG 4. Transcriptional expression of TNF-a gene (as a ratio to and the amplified nested PCR product of the HCV core region. that of b-actin) in the liver in patients with chronic hepatitis C. (C) Probes, primers, and technique were as described by Viazov normal liver. Other initials as in Fig. 2 (Fig. 5) . These data suggest that high type. NR, no response to interferon; Rr, complete response with postpretreatment TNF-a mRNA values are associated with therapy relapse; SR, sustained complete response. lack of response to IFN-a.

In five patients who presented sustained response to IFN-a, TNF-a mRNA levels were determined in PBMC cases and in 19 of 20 PBMC samples. In patients who before and at the end of the therapy. Of these cases, in had received IFN-a treatment, HCV RNA was present one there was a substantial reduction in the levels of in 19 of 51 PBMC samples and in 16 of 25 liver biopsy the cytokine transcripts (from 0.44 before treatment to specimens (see Table 1 ). 0.09 after treatment), in two there was a slight de-As shown in Fig. 7 , when the patients studied at the crease (from 0.39 to 0.30 and from 0.13 to 0.06), and end of 10 to 12 months of treatment with IFN-a were in the remaining two there was a slight increase (from divided according to the presence or absence of HCV 0.31 to 0.41 and from 0.13 to 0.24). Overall, IFN-a RNA in the liver or in PBMC, it was shown that TNFtreatment did not induce significant changes in TNFa mRNA values in the corresponding tissue were a mRNA values in this group of patients who already higher in those cases with detectable HCV RNA than presented low basal levels of TNF-a transcripts. in cases where HCV RNA was negative (0.70 { 0.10 HCV genotyping was performed in 18 UT patients vs. 0.28 { 0.03, P õ .002 in PBMC; 0.79 { 0.06 vs. 0.49 and in 36 patients studied at the end of IFN-a therapy. { 0.09, P õ .05 in liver). In UT patients, genotype 1b was observed in 13, genotype 1a in 2 patients and genotype 3 in 3 patients. Among patients studied at the end of IFN-a therapy, observed that in both untreated patients (Fig. 5 ) and SR* 2/9 0/12 in the patients studied after IFN-a treatment (Fig. 6) , Considering all patients, no correlation was observed between TNF-a mRNA values, measured in the liver or in PBMC and the levels of serum aminotransferases. Also, there was no correlation between TNF-a mRNA values and the Knodell's score of the liver biopsy results (similar results were obtained when total Knodell index or when only inflammatory parameters were considered).

This study and previous reports 20, 21 show that HCV can be observed not only in serum and hepatocytes but also in lymphoid cells, and our results indicate that HCV infection is associated with increased transcriptional expression of the TNF-a gene, in both the liver and PBMC and with high serum levels of TNF-a.

Induction of TNF-a gene by viruses has been shown to occur in a variety of in vitro models, 4,6-9,31 as well as in diverse viral diseases, including HBV-and HCVinduced hepatitis. 11, [16] [17] [18] [19] 32 It has been recently shown FIG 7 . Levels of TNF-a mRNA (as a ratio to b-actin) in PBMC that production of TNF-a in the liver takes place not and in the liver in patients with chronic hepatitis C at the end of only in nonparenchymal cells but also in hepatocytes. 33 10 to 12 months treatment with interferon. Patients were divided The precise stimulus responsible for enhanced TNF-a according to the presence (HCV/) or absence (HCV0) of HCV RNA gene expression in chronic viral hepatitis has not been in the corresponding specimen.

defined. Although TNF-a may be generated during the inflammatory reaction that follows the immune recognition of viral antigens, the mere intracellular presence To evaluate whether TNF-a gene expression might of viral compounds may also stimulate TNF-a gene exbe related to the quantity of the viral load, we quantipression as a built-in defense program of the cell to fied HCV RNA in serum in all of the 20 untreated activate neighboring leukocytes or macrophages 7 or to patients and in PBMC in 15 of these subjects using undergo apoptosis, 34 thus limiting the spread of infeca competitive PCR method. We found no correlation tion. between the levels of TNF-a mRNA in PBMC and the In this study, we did not find a correlation between levels of HCV RNA in serum (P Å .74) or in PBMC (P the activity of the liver disease and the level of TNF-a Å .84). Although the levels of HCV RNA before treatgene expression in the liver or PBMC. However, we ment tended to be higher in NR than in SR patients, observed a relationship between the presence of HCV the differences were not significant either in serum (3.4 RNA in the cells and the values of TNF-a transcripts. 1 10 7 copies/mL vs. 5.4 1 10 6 copies/mL; not significant Thus, in UT patients TNF-a mRNA levels were shown [NS]) or in PBMC (9.6 1 10 4 copies/mg RNA vs. 6.3 to be increased in the liver and in PBMC, and HCV 1 10 4 copies/mg RNA; NS).

RNA was detectable in both sites; whereas in SR pa-TNF-a Gene Expression and Liver Damage. TNF-a tients, HCV RNA was absent from both the liver and transcripts in the liver and in PBMC were similar in PBMC in the majority of cases, and TNF-a mRNA valpatients with and without cirrhosis (0.75 { 0.1 vs. 0.67 ues were comparable to those observed in normal sub-{ 0.04, and 0.31 { 0.08 vs. 0.53 { 0.06, NS). We comjects. On the other hand, in NR subjects and in Rr pared TNF-a mRNA values in liver samples from the patients, HCV RNA persists in the liver and lymphoid eight Rr patients (all of them with low activity of the cells at the end of the treatment in most cases, and liver lesion at the end of the treatment) with those this is associated with an enhanced transcriptional exfound in 10 untreated patients with high activity index pression of TNF-a in these two groups of patients. in the liver biopsy specimens. In the former group the The idea that the transcriptional expression of TNF-Knodell index in liver specimens obtained at the end a is related more to the intracellular presence of the of treatment was significantly lower than in the latter virus than to the activity of the liver disease is supgroup of patients (6.75 { 0.8 vs. 11.5 { 0.9; P õ .005), ported by the following observations. At the end of IFNbut hepatic TNF-a mRNA values were similar in both a therapy, serum aminotransferase levels were equally groups (0.79 { 0.08 vs. 0.68 { 0.04; NS). Furthermore, normal in Rr and SR patients, whereas TNF-a levels among untreated patients, there were five subjects in PBMC and in the liver were significantly increased with minimal changes in the liver biopsy results (mean in the former group. The main difference between these histological activity index, 4); in these cases, TNF-a two groups of patients was viral persistence in patients mRNA levels in PBMC were similar to those observed who eventually relapsed. On the other hand, UT patients showed a more active liver lesion (as estimated in untreated patients with active disease (mean histo-m4730$0017 01-18-96 17:06:42 hepal WBS: Hepatology by serum transaminase levels and histological activity tients with chronic hepatitis C previously resistant to IFN-a alone. 44 From our data it could be speculated index) than Rr patients, but both groups of patients had similarly increased TNF-a mRNA levels, and in that high TNF-a levels produce oxidative stress, which in turn stimulates TNF-a synthesis and causes both, HCV RNA was detectable in liver tissue and lymphoid cells.

IFN-a resistance. This would explain the higher TNFa values observed in patients insensitive to IFN-a Although the presence of the virus appears to determine an enhanced TNF-a gene expression, neither the treatment. Further studies are needed to determine whether pretreatment analysis of TNF-a gene expres-level of viremia nor the quantity of HCV RNA in PBMC were shown to be related to the amount of TNF-a tran-sion can be useful to predict the response to IFN-a in chronic hepatitis C. scripts. Interestingly, the serum concentration of TNFa was observed to be increased not only in untreated Acknowledgment: We thank C. Asensio for excelpatients and in patients who did not respond to IFNlent technical assistance. a treatment but also (although to a lesser extent) in those who exhibited sustained complete response to the

The molecular action of tumor necrosis factor alpha

Clin Immunol Immunopafor increased TNF-a production (and perhaps delayed thol

In vivo administration of tumor necrosis factor-alpha is associated with antiviral activity in human peripheral mononuclear TNF-a has been shown to increase the production cells

In vitro antiin turn, can stimulate tor-alpha and interferon-gamma

Cytokine (tumor necrosis factor, GSH depletion can activate cell proteases involved in IL-6, and IL-8) production by respiratory syncytial virus-infected NF-kB activation, a factor that increases TNF-a gene human alveolar macrophages

Nain expression. 38 In chronic HCV infection, GSH concentra-M, Gemsa D. Coxsackievirus B3-induced production of tumor tion has been shown to be markedly reduced both in necrosis factor-alpha, IL-1beta, and IL-6 in human monocytes. plasma and in PBMC. 39 Interestingly, GSH levels in

Influenza A virus infection of macrophages the cells and normalized when HCV RNA was cleared enhanced tumor necrosis factor-alpha (TNF-a) gene expression of

HCV infection and that TNF-a might con

Infectribute to the GSH depletion observed in chronic hepation of peripheral blood mononuclear cells by herpes simplex and Epstein-Barr viruses. Differential induction of interleukin 6 and titis C. tumor necrosis factor-alpha

Involvement of tumor necrosis known about the mechanisms determining the refactor-alpha in the pathogenesis of activated macrophage-mediated hepatitis in mice

sponse, or the lack of response

Enexhibiting sus-1992

Tumor necrosis factor and endotoxin in the pathothe rat

Alexander response to IFN-a therapy

Increased production of tumour necrosis factor oxidants, such as N-acetyl-cysteine, together with IFNalpha in chronic hepatitis B virus infection

Typing of hepatitis C virus isolates by DNA enzyme immunoassay

blood mononuclear cells induce HIV-1 replication via a tumor necrosis factor-alpha-mediated mechanism

Tumor necrosis factor alpha production by peripheral blood mononu

liams R. Spontaneous production of tumour necrosis factor alpha and interleukin-1beta during interferon-alpha treatment of

Serum levels of cytokines in chronic liver diseases. chronic HBV infection

Induction of tumor necrosis factor-alpha production by human hepatocytes in extrahepatic HCV RNA strands in chronic hepatitis C: different patterns of response to interferon treatment

Apoptosis in cultured rat hepatocytes: the effects of tumour necrosis factor alpha and interferon gamma. J tion of hepatitis C virus in peripheral blood mononuclear cells. effect of alpha-interferon therapy

Tumor necrosis factor induced oxida

Detection of replicative intermediates of hepatitis C viral tive stress in isolated mouse hepatocytes

RNA in liver and serum of patients with chronic hepatitis C

Tumor necrosis factor-alpha-mediated decrease in glutathione increases the sen

Prediction of the response of chronic hepatitis C to sitivity of pulmonary vascular endothelial cells to H 2 O 2

Alterations in macrophage free radical and tumor necrosis factor production by a

Formulation and application of a numerical potassium channel activator

scoring system for assessing histological activity in asymptomatic chronic active hepatitis

Rapid proteolysis of IKB-alpha is necessary for activation of transcription factor NF-kB

Single-step of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction

Elevated circulating levels of tumor necrosis factor predict unresponsiveness to treatment with inter-1985

Kedes feron alpha-2b in chronic myelogenous leukemia

Evolution of the functional human beta-actin gene and its multipseudogene family: conservation of noncoding regions and 41

Tumor necrosis factor to treat hepatitis C virus genome

Enhanced detection by chronic hepatitis B virus infection

Interleukin-2 and alpha/ PCR of hepatitis C virus RNA

Dow beta interferon down-regulated hepatitis B virus gene expression in vivo by tumor necrosis factor-dependent and -independent BC

Civeira use of type-specific peptides to serologically differentiate infections with HCV types 1, 2 and 3