204 Journal ofHepatology, 1993; 19:204-210 © 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. 0168-8278/93/$06.00

HEPAT 01380

Identification of hepatitis B virus-DNA in the liver by in situ hybridisation using a biotinylated probe. Relation to HBcAg expression

and histology

Nikolai V. Naoumov, Helena M. Daniels, Fergus Davison, Adrian L.W.F. Eddleston, Graeme J.M. Alexander and Roger Williams

Institute of Liver Studies. King's College Hospital and School of Medicine and Dentistry. Denmark Hill. London SE5 9RS. UK

(Received 7 August 1990)

The cellular iocalisation of hepatitis B virus (HBV)-DNA in liver tissue was studied by in situ hybridisation using biotinylated and radiolabelled probes on samples from HBsAg carriers with a spectrum of disease and related to the presence of HBV-DNA in serum and intrahepatic HBcAg expression. Sixteen of the 31 patients studied were seropositive for HBV-DNA; nine had chronic active hepatitis and seven had chronic persistent hepatitis. HBV-DNA was detected in the liver tissue in seven of these patients. In each, HBV-DNA was detected in both cytoplasm and nuclei. All seven also had nuclear and/or cytoplasmic HBcAg which in six was associated with chronic active hepatitis. HBcAg (without tissue HBV-DNA) was detected in the remaining nine patients with an exclusively nuclear pattern in two. Fifteen patients were seronegative for HBV-DNA. HBV-DNA was not detected in the tissue of any of these. Three of these were HBcAg positive but in each this was confined to occasional nuclei and each had inactive disease. The close association between the presence of detectable HBV-DNA in tissue, cytoplasmic HBV-DNA expression and chronic active hepatitis in one group and a failure to detect HBV-DNA in those with nuclear HBcAg and benign disease suggests that there may be two distinct patterns of HBV replication in chronic HBV carriers which may influ- ence the development of liver damage.

Key words: Chronic hepatitis B; In situ hybridisation; HBV-DNA; HBcAg

Inflammatory changes in the liver in relation to chronic hepatitis B virus (HBV) infection vary from minimal damage to chronic active hepatitis. This varia- tion is a consequence of the complex interactions be- tween viral replication, tissue expression of viral gene products and the host immune response. It is generally agreed that the HBV is not cytopathic but that hepato- cyte necrosis occurs as a consequence of the immune re- sponse to viral antigens and is therefore limited to the duration of active viral replication (1,2). HBcAg the main antigen against which the cytotoxic response is

directed, is also a sensitive indicator of active viral repli- cation correlating with the presence of HBV-DNA in serum (3,4). Recent studies have emphasised that the in- trahepatic distribution of HBcAg is important since

cytoplasmic HBcAg is more likely to be associated with chronic active hepatitis while exclusively nuclear HBcAg is associated with low grade inflammation, usually con- fined to the portal tract (5).

The mechanism that underlies the differential in- trahepatic expression of HBcAg at different stages of chronic HBV infection has not been determined but it

Correspondence to: Nikolai V. Naoumov, Institute of Liver Studies, King's College Hospital and School of Medicine and Dentistry. Denmark Hill. London SE5 9RS, UK.

MECHANISMS OF LIVER DAMAGE IN CHRONIC HBV INFECTION 205

may reflect the pattern of viral replication. In situ hybridisation for viral DNA in liver tissue using a radiolabelled DNA probe has revealed HBV-DNA in a predominantly cytoplasmic location in patients with chronic active hepatitis (6,7) in association with cyto- plasmic HBcAg expression as well as HBeAg in serum. The presence of HBV-DNA in tissue was interpreted as indicative of active viral replication. More recently chemically labelled probes, including biotin probes, have attracted attention, largely because of improved microscopic resolution and, using these, viral DNA se- quences have been identified in non-hepatic tissues (8-10); it has proved more difficult to identify DNA se- quences in liver tissue with biotin labelled probes because of endogenous hepatic biotin, although with ap- propriate predigestion this is less of a problem (11).

The aims of the present study were first to compare findings using biotinylated and radiolabelled HBV- DNA probes and second, to investigate the cellular io- calisation of HBV-DNA in patients with a wider spec- trum of chronic HBV infection than hitherto examined, including those with minimal liver damage.

Materials and Methods

Patients Thirty-one males seropositive for HBsAg for at least

12 months in whom a liver biopsy had been performed for diagnostic purposes were studied (see Table 1). Thir- teen were seropositive for HBeAg, 18 were seropositive for anti-HBe. Sixteen were seropositive and 15 were seronegative for HBV-DNA. The liver histology was minimal liver damage in nine, chronic persistent hepati- tis in seven, chronic active hepatitis in three, chronic ac- tive hepatitis with cirrhosis in six and inactive cirrhosis in six. Sixteen patients were from areas where HBV is endemic and of those born and resident in the UK, eight were homosexual. Liver tissue from five patients with chronic active liver disease who were serologically nega- tive for all markers of HBV infection served as control material.

Serology Hepatitis markers were sought using standard

radioimmunoassays, 'Ausria II' for HBsAg, 'HBeAg test' for HBeAg and anti-HBe and anti-delta for total antibody to the hepatitis delta virus (all Abbott Diagnostics, Wokingham, UK). Serum HBV-DNA was quantified by a spot hybridisation technique as describ- ed previously (12).

In situ hybridisation Liver tissue fixed in formalin for 18 h and embedded

in paraffin was used for these experiments and analysis was performed blind to serological findings. The tech- nique is described in detail elsewhere (11). Briefly, sec- tions were mounted on slides freshly coated with poly-L-lysine. The probes used were: biotinylated HBV- DNA (ENZO, New York, USA) as the test probe; pHY 2.1 (a Y chromosome probe) and total human DNA kindly provided by Dr. K.A. Fleming (Oxford) and Dr. F. Farzaneh (London), respectively, as positive control probes; hygromycin B-phosphotransferase (Hygro) and chloramphenicol acetyltransferase (Cat), both probes for Escherichia coli genes, also kindly donated by Dr F. Farzaneh as negative control probes. The latter four probes were biotinylated with biotin I I-dUTP (Gibco- BRL, UK) by nick translation. For the radiolabelled HBV probe a 3.2 kb EcoRI fragment covering the whole HBV genome derived from the recombinant plasmid pHBV 130 (kindly donated by Dr. K. Murray, Edin- burgh) was 32p labelled using the Multiprime DNA labelling system (Amersham, UK).

Endogenous peroxidase activity was blocked with methanol containing 1% H20 2 for 30 min. Following hydration and washing, liver sections were digested with 1 mg/ml Protease (type XXIV, Sigma, UK) for 30-33 min and subsequently dehyrated. Ten microliters of the hybridisation mixture were added to the prepared sec- tions. The mixture comprised 50% formamide, 10% dex- tran sulphate, 2x standard saline citrate, 400 #g/ml sheared herring sperm DNA (type XIV, Sigma, UK) and 1 #g/ml biotinylated probe. Tissue and probe DNA were denatured at 95°C for 10 min and hybridisation was carried out at 37°C for 15-17 h.

After washings, sections were then coated with 10% normal goat serum and 2% bovine serum albumin to block non-specific binding and then with rabbit anti- biotin IgG (ENZO, New York, USA) 1:400 for 45 min, biotin-labelled goat anti-rabbit antibody (Vector Labo- ratories, UK) 1:200 for 30 min and finally with preform- ed avidin-biotinylated peroxidase complex (Vectastain ABC kit, Vector Laboratories, UK) 1:200 for 30 min or streptavidin biotinylated complex (Amersham, UK) 1:200. Peroxidase activity was revealed using freshly prepared nickel diaminobenzidine (DAB) as substrate. Fifty microlitres of 8% NiCI2 in distilled water were added to 10 ml TBS with 0.5 mg/ml DAB and sections incubated for 5 min followed by the same substrate plus 0.01% H202 for 5 min. Staining was enhanced by silver amplification (13). Sections were counterstained with haematoxylin and mounted in DPX.

206 N.V. NAOUMOV et al.

Comparison with radiolabelled probe Results with the biotinylated probe were directly com-

pared with 32p-labelled HBV-DNA in 16 sections. Prehybridisation, denaturation and hybridisation condi- tions were identical except that after digestion with Pro- tease, instead of blocking endogenous peroxidase activity, tissue was acetylated with fresh 0.1 M triethanolamine-HCI with 0.25% acetic anhydride for 10 min. Two nanograms of radiolabelled HBV-DNA were applied to each section in a hybridisation mix containing 50% formamide, 10% dextran sulphate, 2× standard saline citrate solution, 1 x Denhardt's solution and 400 /~g carrier DNA. After hybridisation sections were air dried and coated with Ilford K-5 emulsion melted at 40°C. Sections were developed after 3-6 days in the dark and, after fixation, were counterstained with haematoxylin.

Positive samples were those with more than 10 grains per cell, where clear differentiation from background could be made and where digestion with DNAse reduc- ed the signal.

Detection of HBcAg Sections were prepared according to previous recom-

mendations (14). In particular, formalin fixation did not exceed 24 h, sections were heated prior to deparaffinisa- tion (56°C for 10 min) and in xylene for only 10 s'. After blocking endogenous peroxidase activity and non- specific absorption, sections were incubated with rabbit anti-HBcAg (Dako, UK) 1:200 for 45 min at 37°C and subsequently biotin-labelled goat anti-rabbit antibody

(Vector Laboratories, UK) 1:200 for 30 min at 37°C and strepavidin biotinylated peroxidase complex (Amer- sham, UK). The enzyme reaction was developed with nickel-complexed DAB and 0.01% H.~O_, for 5 min. Controls for specificity included omission of rabbit anti- HBcAg, use of non-specific normal rabbit serum and use of HBV negative tissues. The proportion of HBcAg positive hepatocytes was quantified as follows: single positive cells 1+; < 25% 2+; 25-50% 3+; _> 50% 4+.

R e s u l t s

In situ hybridisation using the biotinylated probe HBV-DNA was detected in seven of 31 liver biopsies

with two characteristic patterns. The distribution was either focal, in which groups of hepatocytes with diffuse

Fig. I.

eu~,

. .- : : , • , .., ::2

Liver tissue from a chronic HBV carrier hybridised with a biotinylated probe and developed with immunoperoxidase, showing a cluster of cells with predominantly cytoplasmic staining• for HBV-DNA (arrowed).

M E C H A N I S M S O F L I V E R D A M A G E IN C H R O N I C HBV I N F E C T I O N 207

granular cytoplasmic staining were seen (Fig. 1), or scat- tered hepatocytes with a random distribution (Fig. 2). In all seven biopsies in which HBV-DNA was detected the distribution was predominantly cytoplasmic staining. In some cells only the nucleus was positive for HBV-DNA and in others in association with cytoplasmic staining (Fig. 2).

Specificity was ensured first by using negative con- trols in every experiment (hybridisation fluid alone, or containing the Hygro or Cat probes); these were always negative. The second approach was to seek HBV-DNA in liver sections of patients serologically negative for HBV; these were also negative. Finally, no positive sig- nal was obtained when detection of DNA probes was sought with ABC alone. Positive signals were consistent- ly obtained using the pHY 2.1 and total human DNA probes with typical nuclear staining in all sections studied (11). When HBV genomic material was detected,

• ? ~ ..+ . , , . + ; , .+ ~ + . ' " ~

l b + ++ , + , + + " i , * ~ ' + "+ I + - ' , ~ + + - + - w + -+.,+

• ' . " +:"+ l i P + - . , + . / ' - ' , I ~ Q .+ • + ,+++ -++-+++++,,+:. +: . , , +,. if-+ + + ++ i# / , i

.+ '~ .~ +++ ' ,~ 4 , ~.-

,+ .++. + o + . v + - , .

" ' ' ' 1 1 ' + i I d'++l +, . L+

"~ ] . . . , , . _+dip -+ ', •

- • • • , ~ + , : . ?

. + .' + .+z e + . . . , + +

.11,, -' . W ? " " +~:~- ,+ I I +~,+ql , , ,

Fig. 2. Liver tnssue from a chronic HBV carrier hybridised with a biotinylated probe and developed with immunoperoxidase, showing a

random hepatocyte with nuclear and cytoplasmic HBV-DNA.

these experiments were repeated to ensure specificity and in addition sections were pretreated with ribonuclease (Sigma, UK) 100/zg/ml for I h at 37°C, to determine if this was DNA or RNA. In none of the cases was HBV genomic material digested by ribonuclease, in- dicating that the material detected in these experiments contained HBV-DNA. Three sections staining positive with the HBV-DNA probe were retested after digestion with DNAse (type I, Sigma, UK) which effectively removed positive staining.

Correlation with histology (Tables 1 and 2) The identification of HBV-DNA in tissue using a

biotinylated probe correlated closely with the presence of piecemeal necrosis; six of the nine patients with chronic active hepatitis and active viral replication ac- cording to the presence of HBV-DNA in serum had HBV-DNA in hepatocyte cytoplasm. Only one of the seven patients with chronic persistent hepatitis and ac- tive viral replication according to the presence of HBV- DNA in serum had HBV-DNA in liver tissue. Of the seven patients with HBV-DNA in tissue, six also had cytoplasmic HBcAg, although all seven had nuclear and/or cytoplasmic HBcAg.

Correlation of tissue with serum HBV-DNA (Tables 1 and 3)

All seven patients with HBV-DNA detectable in tis- sue were also seropositive for HBV-DNA while, in con- trast, only nine of 24 in whom HBV-DNA was not detected in liver were seropositive for HBV-DNA. There was no correlation between the presence of HBV-DNA in liver sections and the quantity of HBV-DNA in serum.

Comparison between biotinylated and radiolabelled pro- bes

The sensitivity of these two probes in the detection of tissue HBV-DNA by in situ hybridisation was compared in 14 of the 31 patients and two controls. Controls were negative by both techniques, while in 13/14 patients there was agreement between the two methods. Thus sections from five patients were positive for HBV-DNA with a similar distribution by both techniques and sec- tions from eight patients were negative for HBV-DNA by both techniques. In the remaining patient tissue was positive for HBV-DNA with the radiolabelled probe

alone.

Statistics Direct comparison between groups was made using

Fisher's exact test.

208

TABLE I Tissue HBcAg and HBV-DNA in relation to patient subgroups

N.V. NAOUMOV et al.

Serum HBV-DNA H BeAg/ Liver HBcAg Histology (pg per 40 #1 serum) anti-HBe HBV-DNA (nuclear/

cytoplasmic)

Group A - - _/+ __ a MLD (n = 9)

Group B 281 +/- + 3+/2+ CPH (n = 7) 180 +/- - - 4+/4+ CPH

115 +/- - - 3+11+ CPH 23 +/ - - - I +/I + CPH

250 +/- - - 3+/- CPH 125 +/- - - 2+/- CPH 120 +/- - - 4+/4+ CPH

Group C 5 + / - - - I +/2+ CAH/C (n = 6) 38 +/- - - 2+/1+ CAH

36 +/- + 4+/4+ CAH/C 25 +/- + 4+/4+ CA H/C 35 +/- + 4+/4+ CAH 81 +/- + 4+/4+ CAH

Group D 49 - /+ + 2+/- CAH/C (n = 3) 63 - /+ + 4+/4+ CAH/C

90 - /+ - - 4+/3+ CAH/C

Group E - - _/+ _ a/_ Inactive (n = 6) cirrhosis

Groups A and E were negative for tissue and serum HBV-DNA. aThree patients in groups A and E had occasional hepatocytes with nuclear HBcAg.

D i s c u s s i o n

T h e s e resu l t s i n d i c a t e t h a t H B V - D N A c a n be iden-

t if ied in l iver t i s sue by in s i tu h y b r i d i s a t i o n to a l l ow tis-

sue loca l i s a t i on . H o w e v e r , t i s sue l o c a l i s a t i o n is

u n d o u b t e d l y a t the expens e o f r e d u c e d sens i t iv i ty

because o f the need fo r t i s sue d iges t i on , f irst to e x p o s e

g e n o m i c m a t e r i a l a n d s e c o n d to r e m o v e e n d o g e n o u s

h e p a t i c b io t in . F o r the l a t t e r r e a s o n , b i o t i n is p e r h a p s

n o t the idea l c h e m i c a l labe l fo r in s i tu h y b r i d i s a t i o n

w i t h i n liver. F o l l o w i n g d i g e s t i o n the c h e m i c a l a n d

TABLE 2 Relation between histology, liver HBcAg and liver HBV-DNA

Histology Liver HBcAg positive Liver HBcAg negative

Nuclear Nuclear plus cytoplasmic

Piecemeal • • • • • necrosis • O O O present (n = 9)

Piecemeal O O O O • O O O O O O O necrosis O O O O O O absent O O O O (n = 22)

HBV-DNA was detected more frequently in tissue of those with piecemeal necrosis and HBcAg, than those with HBcAg without piecemeal necrosis, P = 0.0345. O, Tissue positive for HBV-DNA; O, tissue negative for HBV-DNA.

r a d i o l a b e l l e d p r o b e s h a d s im i l a r s ens i t i v i ty a n d whi l e

use o f the r a d i o l a b e l l e d p r o b e w i t h o u t d i g e s t i o n to

e l i m i n a t e e n d o g e n o u s b i o t i n w o u l d u n d o u b t e d l y h a v e

r e su l t ed in i n c r e a s e d sens i t iv i ty , t h i s w o u l d n o t h a v e per -

m i t t e d the s u b c e l l u l a r l o c a l i s a t i o n o f H B V - D N A , w h i c h

was the p r i m a r y a i m o f the p r e s e n t s tudy . P r e v i o u s s tud -

ies in t i ssues o t h e r t h a n l iver u s i n g b o t h t e c h n i q u e s h a v e

s h o w n s imi l a r sens i t iv i ty b u t s u p e r i o r ce l lu l a r r e s o l u t i o n

w i th the i m m u n o c y t o c h e m i c a l a p p r o a c h (15).

T h e t i ssue l o c a l i s a t i o n o f H B V - D N A in th i s ser ies ac-

c o r d s w i th p r e v i o u s r e p o r t s u s i n g a r a d i o l a b e l l e d p r o b e

TABLE 3 Relation between tissue and serum HBV-DNA and liver HBcAg

Histology Liver HBcAg positive Liver HBcAg negative

Nuclear Nuclear plus cytoplasmic

HBV-DNA • O O • • • • positive • • O O (n = 16) O O O O

O

HBV-DNA O O O O O O O negative O O O O (n = 15) O O O O

HBV-DNA was detected in tissue exclusively in those seropositive for HBV-DNA, P = 0.0087. O, Tissue HBV-DNA positive; O, tissue HBV-DNA negative.

MECHANISMS OF LIVER DAMAGE IN CHRONIC HBV INFECTION 209

in which HBV-DNA in a predominantly cytoplasmic pattern was identified in patients with chronic active hepatitis and severe liver disease (6,16). The pattern in more benign histological lesions is less well documented although HBV-DNA was identified in the nucleus in a single patient with chronic persistent hepatitis (17). One previous report of the use of a biotinylated probe in chronic HBV infection also noted a cytoplasmic pattern (18) but in that study the confounding effect of endogen- ous biotin was not considered and staining may have been non-specific, as recently demonstrated (I1).

This study also confirmed the value of intrahepatic HBcAg as a marker of viral replication; patients seropositive for HBV-DNA or HBeAg are usually posi- tive for HBcAg (4,5,7,19), although as shown in this study, the converse is not always true. The importance of the distribution of HBcAg, where cytoplasmic anti- gen is associated with active hepatitis and exclusively nuclear antigen with more benign disease, (as reported in a large series (5)) was confirmed in the present study. This observation is consistent with in vitro immunologi- cal studies which suggest that the target for cytotoxic T- lymphocytes is HBcAg expressed on the plasma mem- brane of hepatocytes (20). These findings are also in ac- cord with electron microscopy studies demonstrating that when HBcAg is present in cytoplasm it also express- ed on the plasma membrane (21); cells expressing the target antigen on the plasma membrane would be more susceptible to immune lysis than those in which the tar- get remained intranuclear.

The close association between tissue HBV-DNA and tissue HBcAg is in accord with a previous study (22) and strengthens the view that there are two distinct phases of chronic HBV infection associated with ongoing viral replication (23). The earlier phase is associated with nu- clear HBcAg, the absence of tissue damage and serum markers indicative of very active viral replication, despite which HBV-DNA is detected only rarely in tis- sue. The subsequent phase is characterised by cytoplas- mic expression of HBcAg, tissue damage and lower levels of serum HBV-DNA, yet also by the presence of cytoplasmic HBV-DNA. The finding of tissue HBV- DNA in those with lower levels of viral replication (see groups B and C, Table 1) was therefore unexpected.

The reasons underlying the differential localisation of HBcAg and the change in iocalisation with time in chronic HBV infection are not known. One factor may be subtle changes in the molecular biology, analagous to those with the pre-core mutant (24-26). An alternative explanation is that the liver damage itself results in reduced export of HBcAg from the heptocyte and that

this is associated with accumulation of HBV-DNA within cytoplasm.

Acknowledgement

N.V. Naoumov was supported by the Welicome Trust.

References

I Realdi G, Alberti A, Rugge M, et al. Seroconversion from hepa- titis B e antigen to anti-HBe in chronic type B virus infection. Gastroenterology 1980; 79: 195-9.

2 Hoofnagle JH, Dusheiko GM, Seeff LB, et al. Seroconversion from hepatitis B e antigen to antibody in chronic type B hepatitis. Ann Intern Med 1981; 94: 744-8.

3 Bonino F, Hoyer B, Nelson J, et al. Hepatitis B virus DNA in the sera of HBsAg carriers: a marker of active hepatitis B virus repli- cation in the liver. Hepatology 1981; l: 386-91.

4 Hadziyannis S J, Liberman HM, Karvountzis GG, et al. Analysis of liver disease, nuclear HBcAg, viral replication and hepatitis B virus DNA in liver and serum of HBeAg vs. anti-HBe positive carriers of hepatitis B virus. Hepatology 1983; 3: 656-62.

5 Hsu HC, Su I J, Lai M, et al. Biologic and prognostic significance of hepatocyte hepatitis B core antigen expression in the natural course of chronic hepatitis B virus infection. J Hepatol 1987; 5: 45-51.

6 Gowans EJ, Burrell CJ, Jilbert AR, et al. Detection of hepatitis B virus DNA sequences in infected hepatocytes by in situ hybridisation. J Med Virol 1981; 8: 67-78.

7 Gowans E J, Burrell C J, Jilbert AR, et al. Cytoplasmic (but not nuclear) hepatitis B virus (HBV) core antigen reflects HBV DNA synthesis at the level of the infected hepatocyte. Intervirology 1985; 24: 220-5.

8 Brigati DJ, Myerson D, Leary J J, et al. Detection of viral genome in cultured cells and paraffin embedded tissue sections using bio- tin labelled hybridisation probes. Virology 1983; 126: 32-50.

9 Myerson D, Hackman RC, Meyers JD. Diagnosis of cytomegaloviral pneumonia by in situ hybridisation. J Infect Dis 1984; 150: 272-7.

10 Burns J, Redfern DRM, Esiri MM, et al. Human and viral gene detection in routine paraffin embedded tissue by in situ hybridisation with biotinylated probes: viral Iocalisation of her- pes simplex. J Clin Pathol 1986; 39: 1066-73.

11 Naoumov NV, Alexander GJM, Eddleston ALWF, Williams R. In situ hybridisation in formalin fixed, paraffin embedded liver specimens: a method for the detection of human and viral DNA using biotinylated probes. J Clin Pathol 1988; 41: 793-8.

12 Fagan EA, Guarner P, Perera SDK, et al. Quantitation of hepati- tis B virus DNA (HBV DNA} in serum using the spot hybridisa- tion technique and scintillation counting. J Virol Methods 1985; 12: 251-62.

13 Przepiorka D, Myerson D. A single step silver enhancement method permitting rapid diagnosis of cytomegalovirus infection in formalin fixed paraffin embedded tissue sections by in situ hybridisation and immunoperoxidase detection. J Histochem Cytochem 1986; 34: 1731-4.

14 Trevisan A, Gudat F, Busachi C, et al. An improved method for HBcAg demonstration in paraffin-embedded liver tissue. Liver 1982; 2: 331-9.

15 Jirikowski GF, Ramalho-Ortigaeo JF, Lindl T, Seliger H. lmmu- nocytochemistry of 5-bromo-2'-deoxyuridine labelled oligonu- cleotide probes. A novel technique for in situ hybridisation. Histochemistry 1989; 91: 51-3.

16 Burrell CJ, Gowans EJ, Rowland R, et al. Correlation between liver histology and markers of hepatitis B virus replication in in- fected patients; a study by in situ hybridisation. Hepatology 1984; 4: 20-4.

[7 Blum HE, Stowring L, Figus A, et al. Detection of hepatitis B

210 N.V. NAOUMOV et al.

virus DNA in hepatocytes, bile duct epithelium and vascular ele- ments by in situ hybridisation. Proc Natl Acad Sci USA 1983; 80: 6685-8.

18 Negro F, Berninger M, Chiarberge E, et al. Detection of HBV- DNA by in situ hybridisation using a biotin-labelled probe. J Med Virol 1985; 15: 373-82.

19 Omata M, Yokosuka O, Imazeki F, et al. Correlation of hepatitis B virus DNA and antigens in the liver. Gastroenterology 1987; 92: 192-6.

20 Naumov NV, Mondelli M, Alexander GJM, et al. Relationship between expression of hepatitis B virus antigens in isolated hepa- tocytes and autologous lymphocyte cytotoxicity in patients with chronic hepatitis B virus infection. Hepatology 1984; 4: 63-8.

21 De Vos R, Ray MB, Desmet VJ. Electron microscopy of hepatitis B virus components in chronic active liver disease. J Clin Pathol 1979; 32: 590-600.

22 Arakaki Y, Toyodo H, Watanabe S, Seki S, Oda T. Detection of hepatitis B virus-DNA in liver tissue in chronic type B liver

diseases and its relevance to viral antigen: a study by in situ hybridisation using a biotinylated probe and streptavidin- alkaline phosphatase. Acta Histochem Cytochem 1988; 21: 489-96.

23 Chu CM, Karayiannis P, Fowler MJF, et al. Natural history of chronic hepatitis B virus infection in Taiwan; Studies of hepatitis B virus DNA in serum. Hepatology 1985; 5: 431-4.

24 Uy A, Bruss V, Gerlich WH, Kochel HJG, Thomssen R. Precore sequence of hepatitis B virus inducing e antigen and membrane association of the viral core protein. Virology 1986; 155: 89-96.

25 Ou J-H, Laub O, Rutter WJ. Hepatitis B virus gene function: the precore region targets the core antigen to cellular membranes and causes secretion of the e antigen. Proc Natl Acad Sci USA 1986; 83: 1578-82.

26 Jean-Jean O, Levrero M, Will H, Rossignol J-M, Perricaudet M. High level expression of hepatitis B virus C-ORE in human cells. In: Robinson WS, ed. Hepadna Viruses. New York: Alan Liss Inc., 1987: 223-35.