| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Oncology |
Department of Virus-Host Interactions [Q. S., R. B., C. H. S.] and Biostatistics Unit [L. E.], Deutsches Krebsforschungszentrum, 69120 Heidelberg, Germany; Departments of Pathology [Q. S., S-F. W.] and Clinical Diagnostics [T-E. C.], Tangdu Hospital, The Fourth Military Medical University, Xi’an, 710038 China; University of Lübeck, 23538 Lübeck, Germany [H. H.]; and Takegoshi Internal Medicine Clinic, Takaoka, 9330014 Japan [K. T.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: Nucleic acids extracted from sera were analyzed using established PCR and reverse transcription-PCR procedures targeted to HBV x protein gene regions. Amplification products were cloned and sequenced.
Results: Base substitution patterns were determined, which indicated infection stages advanced to different degrees regardless of the transcript type analyzed. HBV full-length RNA (fRNA) showed a high correlation with hepatitis B e antigen and viral DNA, indicative for a replicative infection. In contrast, truncated RNA (trRNA) appeared to be independent of hepatitis B e antigen and showed only a weak association with circulating viral DNA. No correlation was observed between the levels of trRNA and the apparent liver damage as reflected by alanine transaminase levels. An age-dependent representation of fRNA and trRNA was observed: fRNA decreased progressively to low levels, whereas trRNA remained at comparably high values. trRNA and RNA not polyadenylated at either of the two polyadenylation signals were detected even in the absence of any other conventional HBV seromarker, including viral DNA. This was shown for patients with cryptogenic cirrhosis and hepatitis C virus carriers.
Conclusions: The identification of HBV RNA in human serum has a diagnostic potential for apparent and for inapparent infection stages.
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Chronic infection proceeds from replicative to nonreplicative stages (8)
and may be asymptomatic for many years or result in slight liver damage. Alternatively, it leads to progressive liver disease. Regardless if causing progressive liver disease or not, the chronic infection appears to be linked to the development of HCC, although to a different extent (1
, 9)
. New cases of HCC (>400,000) are found each year worldwide with a 5-year survival rate <3% (10)
. HBV is considered to be the main causative factor of liver cancer, possibly accounting for 60% of cases worldwide and 
70% of cases in endemic areas (11)
. Although vaccination programs in several countries have been successful (2
, 3
, 12
, 13)
, for those already infected, it is important to develop discriminative screening procedures as a basis for targeted therapy of critical infection stages. The present study introduces circulating viral transcripts as a potential parameter.
The four overlapping ORFs on viral DNA, i.e., PreC/C, polymerase, PreS1/PreS2/S, and X, are transcribed into genomic and subgenomic RNA molecules. The translation product of the smallest viral transcript, HBx, is regarded as a hepatocarcinogenic factor. Properties assigned to HBx include the stimulation of transcription and signal transduction (14 , 15) , interference with DNA repair (16 , 17) , induction of apoptosis (18, 19, 20, 21) , and malignant transformation both in vitro and in vivo (22, 23, 24, 25, 26) . Although the mechanism of HBV-associated hepatocarcinogenesis is not established, integration of viral DNA into chromosomes of hepatocytes (27, 28, 29) , preferential preservation of HBx ORF sequences (30) , and selective, albeit weak expression of HBx in preneoplastic (31) and neoplastic human hepatocytes (29 , 31, 32, 33) may be critical.
During replication of HBV, all viral transcripts mature at a unique poly(A) signal downstream of the HBx ORF, namely a UAUAAA motif at position 1789 (34 , 35) . Collectively, these transcripts are referred to as full-length transcripts in this study. Chromosomally integrated DNA usually does not allow generation of full-length transcripts. Truncations affecting the 3'-end region of the HBx ORF are frequent (28 , 29 , 36, 37, 38) and lead to fusion of HBx sequences with adjacent cellular sequences. Respective HBx/cell hybrid transcripts may direct the synthesis of functionally active HBx/cell fusion proteins (25 , 36 , 39, 40, 41) . Other transcripts, designated as truncated, use for maturation a cryptic poly(A) signal, a CAUAAA motif within the HBx ORF (position 1661). They are present in HCC and surrounding liver parenchyma (36) . The existence of respective truncated HBx proteins with growth-regulating properties has been postulated (42) . In a recent study, using liver tissue samples from HCC patients, trRNA has been shown to be a frequent marker of chronic infection, the representation of which increases with advancing age (43) .
Although it cannot be excluded that free forms of viral DNA contribute to the pool of trRNA, a substantial part of it is likely to be transcribed from chromosomally integrated HBV DNA, which accumulates during chronic infection. Hence, truncated transcripts are attractive markers not only for monitoring progression from replicative to nonreplicative stages but also for recognizing inapparent infection stages. Usually, for analyzing viral RNA, liver tissue samples from biopsies or resection material are required. In the present study, we propose that it is also possible to use serum samples to detect and describe viral RNA related to chronic infection. The feasibility of detecting transcripts from a DNA virus in serum has been demonstrated previously for the EBV (44) .
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Primers and Probes.
Upstream primers: txs3 (1434; Ref. 45
): TCT CAT CTG CCG GAC CGT GT; txs1 (1454): GCA CTT CGC TTC ACC TCT GC; txs (1445): GGA CCG TGT GCA CTT CGC TT; and GAPDH1 (3755; Ref. 46
): CAT CTC TGC CCC CTC TGC TG. Downstream primers: txas5 (1683): (T)15 GCT GG; Rxas2 (1808): (T)15 GAA GC; Rxas4 (1806): (T)15 AGC TC; xas1 (1668): AAT TTA TGC CTA CAG CCT CC; xas2 (1634): TTA ACC TAA TCT CCT CCC CC; and GAPDH2 (4344): GGA TGA CCT TGC CCA CAG CCT. Probes for polarity determination: 1561+: GAC CGA CCT TGA GGC ATA CTT CAA AGA CTG and 1590-: CAG TCT TTG AAG TAT GCC TCA AGG TCG GTC.
PCR for X-region DNA.
For amplification of X-region DNA, 2 µl of nucleic acid extract, 75 ng each of txs3 and xas1 (Fig. 1)
, deoxynucleotide triphosphate (0.2 mM each; Life Technologies, Inc., Gaithersburg, MD), and 1.25 units of Taq DNA polymerase (Life Technologies, Inc.) were incubated in a reaction volume of 50 µl as specified by the supplier. Cycles (35)
were performed (40 s denaturation at 90°C, 50 s annealing at 56°C, and 40 s elongation at 70°C). Amplification efficacy and specificity were demonstrated by positive controls, using plasmids pMT9T40A (200 pg) and pMT9T41A (200 pg) carrying cloned cDNAs of full-length and truncated HBV transcripts, respectively (36)
, and by a negative control (target sequences replaced by water).
|
). Cycles (35) were performed as described in the reaction for viral DNA after reverse transcription (20 min at 50°C). Samples containing cDNA of the plasmids pMT9T40A and pMT9T41A, before and after digestion with DNase I, served as positive and negative controls. In addition, water controls were included. rRNA (1 ng), used as a carrier during the nucleic acid extraction described above, served as an additional negative control. To ensure RNA specificity, the described PCR procedures were performed with and without a reverse transcriptase reaction. RT-PCR for detection of GAPDH transcripts using the primers GAPDH1 and GAPDH2 (46)
served as controls for the presence and integrity of RNA in the sera.
Identification of Full-length and Truncated HBV Transcripts via a Semi-nested PCR After Reverse Transcription.
As anchored oligo(dT) primers ensure selective amplification of RNA (43
, 45)
, pretreatment of samples with DNase I was not required when studying 3'-end structures. Nucleic acid extracts, 10 µl each, were used for the first-round PCR (anchored RT-PCR) with a one-tube system in a reaction volume of 50 µl as described above. The upstream primer (150 ng; txs3 for fRNA and txs for trRNA; Fig. 1
) and 300 ng of the downstream anchored oligo(dT) primer (a Rxas2/Rxas4 mix for fRNA and txas5 for trRNA; Fig. 1
) were used for the first-round PCR. Amplification products, 2 µl each, were then subjected to the second-round PCR with txs1 as the upstream inner primer for both reactions (full-length and truncated transcripts), while maintaining the anchored oligo(T) primers used in the respective first-round PCR (Fig. 1)
. Amplification conditions were the same as described above except for the reduction of the annealing temperature to 53°C (43
, 45)
.
Aside from water controls, in a series of 15 samples, efficacy and specificity of the reaction for full-length transcripts were assessed using 50-pg plasmid pMT9T40A as a positive control and 50-pg plasmid pMT9T41A as a negative control. Inversely, for truncated transcripts, plasmid pMT9T41A was used as a positive control, and plasmid pMT9T40A was used as a negative control. The inclusion of the positive controls also allowed to assess signal intensities at identical staining conditions. In some cases, RNA extracted from HBV-infected liver shown previously to contain only full-length transcripts (43)
and another extract from HCC tissue containing only the truncated transcripts (43)
were included as positive controls. In addition, 1 ng of rRNA was used as a negative control. Special attention was paid to avoiding contamination of the samples, as described by Kwok and Higuchi (47)
. Assays were carried out at least twice in independent series.
DNA Probe Preparation.
A DNA probe for Hyb was prepared by PCR with Dig-11-dUTP (Roche), pMT9T40A as a template, and the primers txs1 and xas1 (Fig. 1)
. A low-incorporation ratio Dig-11-dUTP:dTTP = 1:20 was used to verify visible (Eb) DNA products as HBx specific rather than to increase the sensitivity.
Electrophoresis, Southern Blotting, and Hyb.
Amplification products (15 µl) were separated on a 2% agarose gel in the presence of Eb (0.2 µg/ml) and visualized under UV light. Subsequently, the gels were blotted onto Hybond-N+ nylon membranes (Amersham, Buckinghamshire, England) and hybridized to the Dig-labeled DNA probe. Hyb signals were visualized via consecutive application of a sheep anti-Dig antibody (Roche) and alkaline phosphatase-labeled rabbit antisheep IgG (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA). PCR involving the anchored oligo(dT) primers usually yielded two products, one with the predicted migration behavior and a faster migrating one. The strand composition of each product was determined via Hyb with 32P-labeled probes of opposing polarity, 1651+ and 1590-.
Cloning and Sequencing of Amplified cDNA.
Selected amplification products of full-length and truncated transcripts were ligated into the pCR2.1-TOPO vector (Invitrogen, Groningen, the Netherlands). Sequencing of both strands was performed by the Sanger di-deoxy-chain-termination-method.
Statistical Analysis.
Cases positive for HBV seromarkers were analyzed for correlation between age, conventional seromarkers (determined at a 1:200 dilution of the serum), viral DNA, fRNA, and trRNA. The following marker combinations were used to establish subgroups: HBs-positive/HBe-positive, HBs-positive/HBe-negative, HBs-negative/viral nucleic acid-positive, and negative for all markers. Statistical evaluations were carried out using the software S-PLUS (48)
and StatXact 3 for Windows (Cytel Software Corp., Cambridge, MA). Pearson’s correlation coefficient (r) was used to describe relations between circulating viral transcripts or DNA and immunological seromarkers for HBV infection, serum ALT values, or ages of donors, within a 95% CI. To compare ages of different patient groups, the Kruskal-Wallis rank-sum test was used for an overall difference, and the Wilcoxon’s rank-sum test (49)
was used for pairwise comparisons. Fisher’s exact test (two-sided; Ref. 50
) was applied to compare frequencies of viral transcripts and DNA, in sera positive and negative for anti-HBc and in cases with or without HCC. Each respective contingency table was assessed for heterogeneity or equivalently for dependency of the two factors generating the table using Fisher’s exact test in the case of smaller samples (n < 50). In the case of larger samples (n 
50), the 
2 test or an approximate version of StatXact software was used. All correlations are given as absolute (positive) values in this paper. P < .05 was regarded as being statistically significant.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Proceeding from top to bottom, Fig. 1
outlines HBx target sequences for PCR analyses: (a) PCR for the detection of DNA carrying sequences between map positions 1434 and 1668; (b) RT-PCR for detection of RNA spanning the same sequence; (c) RT-PCR recognizing the junction structure at the poly(A) addition site of full-length transcripts [GCUUC(A)n1]; and (d) RT-PCR with anchored oligo(dT) primers recognizing the corresponding structure [CCAGC(A)n] on truncated transcripts. In the first-round amplification for fRNA, the sense primer txs3 (1434) was used, and for trRNA, the sense primer txs (1445) was used. In the second-round PCR, uniformly the sense primer txs1 (1454) was used. A Dig-labeled probe spanning map positions 1454 and 1668 was used to confirm the HBx-sequence specificity.
RNA specificity of the RT-PCR for X-region transcripts was ensured via parallel PCR testing of samples with and without DNase I pretreatment, as well as with and without reverse transcription. RNA specificity of reactions involving anchored oligo(dT) primers for full-length and truncated transcripts has been demonstrated previously (43
, 45)
. In this study, fRNA and trRNA were identified even in the absence of detectable viral DNA (Fig. 2
, cases M and S–U and Fig. 3
, cases B, D, and H). These data provide further evidence for the RNA specificity of the reactions conducted. For all 437 sera examined, a GAPDH amplification product of 305 bp was obtained, as expected for GAPDH transcripts, but not of 590 bp, as expected for GAPDH-DNA amplification (46)
.
|
|
and 3
illustrate representative PCR patterns. In the case of viral DNA and GAPDH mRNA, only the Eb-stained gels are shown; in the case of HBx RNA, the Hyb data are shown; and in addition, in the case of fRNA and trRNA, the Eb-stained gels are shown. HBx RNA signals (Figs. 2
and 3
, RNA) are present in most of the cases selected but displayed a wide range of intensities. fRNA signals visualized both by Eb staining and Hyb appear as a double band, with the upper one migrating at the expected position of 370-bp DNA and a faster migrating lower band, which represented single-stranded minus-strand DNA generated by asymmetric amplification as revealed by strand-polarity analysis. A similar tendency for asymmetric amplification was also observed for trRNA (data not shown).
The amplification of the 3'-end region of trRNA gave rise to the expected 245-bp product. In all cases positive for the fRNA, an additional 370-bp amplification product was obtained. Cloning and sequencing of this 370-bp cDNA segment confirmed its identity with that of the authentic fRNA amplification product, except for its very 3'-end on which the txas5 anchor was connected to positions ranging from 1803 to 1806. The presence of the trRNA-specific 245-bp product was independent of the presence or absence of the 370-bp product (Fig. 2
, compare cases A–C). Apparently, the RT-PCR procedure adapted for the analysis of trRNA from sera recognized both fRNA and trRNA.
PCR signal intensities for fRNA and trRNA obtained with a nucleic acid equivalent of 40 µl of serum were graded into weak (1+), moderate (2+), and strong (3+) and related to values obtained with 1 fg, 10–100 fg, and 1 pg of cDNA (pMT9T41A). These estimations do not take into account the reduction in sensitivity by background RNA, as described by Schutz et al. (45)
, and, therefore, represent minimum values.
On the basis of presence and intensities of signals for full-length and truncated transcripts, five patterns were distinguished: (a) fRNA dominance at high (3+; Fig. 2
, cases B, C, E, F, H–J, O, and Q) and (b) at low levels (2+ or 1+; Fig. 2
, case K and Fig. 3
, case B); (c) trRNA dominance (Fig. 2
, cases G and T); (d) trRNA exclusively (ranging from 1+ to 3+; Fig. 2
, cases A, M, N, S, and U and Fig. 3
, cases D and H); and (e) absence of both RNA types (-/-; Fig. 2
, case D and R and Fig. 3
, cases A, E, and I).
Sequence Variation on the HBx-ORF Segment Amplified by RT-PCR Involving Anchored Oligo(dT) Primers.
Amplification products obtained in RT-PCR assays were cloned and sequenced. Table 1
presents base substitution patterns for individual cases of fRNA and trRNA predominance, respectively. Positions different from a putative wild-type sequence (at least in two cases) are indicated by bold letters. Considering 1635T/1637A (XhoI coordinates; corresponding to EcoRI coordinates 1762/1764), 1526T, and 1626C or G as indicators of an elapsed period of chronicity (reviewed by Kramvis and Kew in Ref. 51
), trRNA (Table 1
, trRNA) can be assigned to conserved as well as less conserved templates of viral DNA. Hence, trRNA originates early and late during chronic infection. Similar base substitution patterns were gained for fRNA (Table 1
, fRNA), although fRNA tended to be less conserved than trRNA.
|
. An HBx RNA signal, i.e., a signal for sequences shared by both fRNA and trRNA, was obtained for almost all samples with full-length and/or truncated viral transcripts. If present, intensities of the HBx RNA signal were in accordance with those for full-length transcripts (Fig. 2)
; in cases with dominance or exclusive presence of truncated transcripts, HBx RNA signals (one round of RT-PCR) usually appeared weaker than those for truncated transcripts after the seminested (double round) RT-PCR (Fig. 2
, cases A, G, N, and S). In some of the HBs-positive samples in which neither full-length nor truncated transcripts were identified (39 of 237), a weak signal was obtained for HBx RNA (Fig. 2
, case D), indicating the presence of transcripts not polyadenylated at either of the two viral poly(A) signals.
|
, a close correlation existed between the levels of full-length transcripts and viral DNA (r = 0.79, P < .05). It is noteworthy that subgroups with full-length transcripts levels 1+-3+ were found in the group showing the highest levels of viral DNA (3+). In HCC patients, the combination of high DNA levels with full-length transcripts at lower levels appeared to be more frequent compared with HCC-free carriers (data not shown). In summary, most of the sera with high levels of full-length transcripts contained high amounts of viral DNA. By contrast, the correlation between truncated transcripts and circulating viral DNA was very weak (r = 0.14, P < .05; Fig. 4B
).
|
). A similar pattern was observed for full-length viral transcripts (r = 0.60, P < .05; Fig. 5B
). By contrast, no significant change was observed for trRNA during progression to HBs-negative stages (r = 0.008, P > .05; Fig. 5C
). It can be concluded that there is a sequential change in the ratio of the two forms of viral transcripts.
|
) and in parallel with the levels of the viral DNA (r = 0.39, P < .05; Fig. 6A
). However, the levels of truncated transcripts were not significantly correlated with ALT values (r = 0.04, P > .05; Fig. 6C
). Apparently, the presence of truncated transcripts is not related to liver damage, whereas this applies to a certain degree for viral DNA and fRNA.
|
) in parallel with a decline of viral DNA (r = 0.31, P < .05; Fig. 7A
). An age-dependent decline of truncated transcripts was not found (r = 0.07, P > .05). In the first decade (samples from only 9 patients were available), its level was lower than observed for full-length transcripts. Sustained high levels were observed starting with the second decade (Fig. 7C)
. In the age group 30–50 years, fRNA dominance at high levels gradually decreased to dominance at lower levels. This was mainly found in cases seronegative for HBe with high levels of viral DNA (see Fig. 5, A and B
, middle columns). In summary, a gradual decrease in fRNA dominance accompanied by a relative increase in the fraction of trRNA was observed (Fig. 7D)
.
|
Viral Transcripts in HBs-negative Sera from HCC Patients.
Even in HBs-negative HCC patients (Table 2)
, serum HBV nucleic acids were detected (29 of 49, 49.2%). Their prevalence was higher in sera positive (19 of 26, 73.1%) than in sera negative (10 of 23, 43.5%) for anti-HBc and anti-HBs (P < .05). Among the 29 positive cases, 20 were positive only for transcripts, 5 were positive for both transcripts and viral DNA, and 4 showed a weak signal only for viral DNA.
Among the HBs-negative sera from HCC patients, 9 were HCV positive. Only 3 of these were anti-HBc positive, whereas 7 were shown to be positive for the HBV viral nucleic acids. Among the 7 cases, 5 were positive for HBx RNA with 2 being positive for both HBx RNA and truncated transcripts. One case of 7 showed only full-length transcripts, one only viral DNA (Fig. 3)
.
Viral Transcripts in HBs-negative Sera from Patients with Diseases Apparently Not Related to HBV.
HBs-negative sera from 9 patients with cryptogenic cirrhosis and 75 patients with extrahepatic chronic diseases were examined as a reference group (Table 2)
. Of the cirrhosis patients, 5 were positive for viral RNA (3 for truncated, 2 for full length), whereas no case was positive for viral DNA. Of the patients with extrahepatic diseases, 11 were positive for viral RNA. Among these, transcripts were polyadenylated exclusively (7 cases; Fig. 2
, cases S and U) or predominantly (2 cases; Fig. 2
, case T) at the cryptic poly(A)-signal in 2 cases only at the major poly(A)-signal.
Viral Transcripts in HBs-negative Sera from Apparently Healthy Individuals.
Samples from 37 urban inhabitants from Xi’an, China and 30 blood donors from Lübeck, Germany served as reference groups. In no case was viral DNA present (Table 2)
. In the Chinese group, 4 cases (10.8%) were positive for HBx transcripts. In the German group, no transcript signal was obtained.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Thus far, it is generally accepted that, after HBs clearance, viral DNA cannot be detected in serum (54) . However, viral DNA has been repeatedly detected in the liver (29 , 33 , 55, 56, 57, 58) . In addition, viral RNA, specifically X-region RNA, has been found in hepatic tissues including HCC from HBs-negative patients (29 , 33 , 59, 60, 61) . Transcription from integrated DNA (28, 29, 30) and low-level replication as observed after HBs seroconversion (57 , 62, 63, 64, 65) may contribute to the pool of viral RNA. Superinfection with HCV (66 , 67) has been related to an inhibition of HBs production, possibly creating conditions where viral RNA is found in the absence of seromarkers indicating replicative processes. Studies on RNA from tissues of chronically infected liver have provided insights into discernible species of HBV transcripts (36 , 43 , 59) . Less is known about changes in the patterns of HBV transcripts during chronic infection. A release of RNA into the blood as observed by Chen et al. (68) for telomerase RNA in breast cancer patients should facilitate studies addressing this subject.
In the present study, X-region RNA was identified in 90% of sera from HBs carriers. In line with its preferential maintenance in liver tissue (59) , X-region RNA was also present in 40% of sera from HCC patients negative for HBs. Our findings thus render HBx-region RNA a potential seromarker for late nonreplicative carrier states.
Full-length transcripts were closely related to serum DNA and HBe and also to the injury of liver parenchyma as reflected by ALT values. In the course of chronic infection, levels of the full-length transcripts declined in parallel to those of viral DNA. These observations confirm the link between full-length viral transcripts and virus replication, as suggested by Kairat et al. (43) on the basis of the analysis of tissue samples.
No pronounced association was observed between circulating truncated viral transcripts and replication parameters or ALT levels. Of all HBV seromarkers examined, trRNA appeared to be the most persistent one (see Fig. 7
). Its fraction of total X-region RNA (HBx RNA) increased with increasing age. Similarly, the fraction of HBx RNA polyadenylated at neither the standard nor the cryptic terminating site (HBx RNA found in the absence of full-length and truncated transcripts) increased during chronic infection, whereas DNA is absent in serum of many cases. It is tempting to assume that these X-region RNA species represent virus/cell RNA hybrid molecules and that they are cotranscribed with trRNA from chromosomally integrated HBV DNA at late stages of infection.
The existence of viral RNA in serum is difficult to explain in view of the known presence of RNases (69) . Capsids as the principal protecting elements are not very likely, because viral RNA is found also in the absence of replication markers. Furthermore, nucleocapsids are not released from hepatocytes before hydrolysis of the RNA pregenome during its transcription into +strand DNA (70) . A possible release of viral RNA not mediated by a capsid structure is further indicated by the absence of HBc-region RNA in cases with a high representation of X-region RNA observed previously for hepatic tissues (43) . However, it cannot be totally excluded that there exists some association between viral RNA and circulating virions (71) . It could be conceived that plasma membrane vesicles shed from hepatocytes (72 , 73) contain and protect viral RNA; however, other protective elements may exist.
Viral transcripts in serum constitute new parameters to recognize and differentiate apparent and inapparent stages of HBV infection. They are detectable even in apparently healthy individuals without a history of overt infection. For a more accurate estimation of prevalence of these transcripts and assessing their significance, it is necessary to screen large numbers of sera collected from individuals in areas of high and low endemicity. The clinical relevance of HBV infection in the absence of serological markers has been recognized in liver transplantation where the entry of HBV infection into an acute stage is not rare (74) .
The sustained expression of HBx for several decades, even after reaching nonreplicative stages, may expose the host to an increased risk for HCC and may be more important for the induction of the tumor than for its continuous growth (29 , 43 , 59) . There are reports on the presence of X-region DNA and RNA in liver tissue of HBs-negative HCV-infected patients (55 , 58 , 61 , 75) , indicating also for these cases a role of HBx in HCC development. This notion is further supported by prospective studies which indicate that liver tissue HBx DNA and RNA and the serum marker anti-HBc confer an elevated risk to develop HCC (61 , 76) . In the present study, we identified HBx RNA in sera of a large number of patients with inapparent HBV infection, including HCV-infected HCC patients. Thus, serodetection of HBV RNA potentially substitutes tissue analysis requiring liver biopsies and may even be more informative and representative.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 To whom requests for reprints should be addressed, at Department of Virus-Host Interactions (F0600), Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 242, 69120 Heidelberg, Germany. Phone: 49-6221-42-4850; Fax: 49-6221-42-4852; E-mail: c.schroeder{at}dkfzheidelberg.de 
2 The abbreviations used are: HBV, hepatitis B virus; ALT, alanine transaminase; Dig, digoxigenin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HBc, hepatitis B core antigen; HBe, hepatitis B e antigen; HBs, hepatitis B surface antigen; HBx, hepatitis B virus x protein; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; RT-PCR, reverse transcription-PCR; fRNA, full-length RNA; trRNA, truncated RNA; oligo(dT), oligodeoxythymidylic acid; CI, confidence interval; ORF, open reading frame; Eb, ethidium bromide; Hyb, hybridization.
Received 1/29/01; revised 4/ 9/01; accepted 4/17/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. JORG HACKER, W. ZHANG, M. TOKUS, T. BOCK, and C. H. SCHRODER Patterns of Circulating Hepatitis B Virus Serum Nucleic Acids during Lamivudine Therapy Ann. N.Y. Acad. Sci., June 1, 2004; 1022(1): 271 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Hennig, I. Puchta, J. Luhm, P. Schlenke, S. Goerg, and H. Kirchner Frequency and load of hepatitis B virus DNA in first-time blood donors with antibodies to hepatitis B core antigen Blood, September 18, 2002; 100(7): 2637 - 2641. [Abstract] [Full Text] [PDF]
|
|
10 years, from 11 to 20 years, etc.). Case numbers are in parenthesis (abscissa). D, representation patterns of f/trRNA for the same age groups. -/-, absence of both RNA types; tr2, only trRNA; tr1, trRNA dominance; f1, fRNA dominance at high levels; f2, fRNA dominance at low levels. Viral DNA (A): r = 0.31, P < .05, 95% CI = 0.17–0.44; fRNA (B): r = 0.32, P < .05, 95% CI = 0.20–0.45; trRNA (C): r = 0.07, P > .05, 95% CI = 0.07–0.22; f/trRNA patterns (D): r = 0.31; 95% CI = 0.18–0.43.