Journalof Hepatology, 1992; 14: 294-299 @ 1992Elsevier Science Publishers B.V. All rights reserved. 016%8278/92/$05.00

294 HEPAT 00932

Interferon-y receptor eq3ression in chronic

Johnson

Y.N.

atitis

ction

Lau’, Alan G. Morris2, Graeme J.M. Alexander’ and Roger Williams’

‘Instituteof Liver Studies, King’s College Hospital and King’s College School of Medicine and Dentistry, London and ‘Departmenr of Biological Sciences, Universityof Warwick, Coventry, United Kingdom

(Received 19January 1991)

It is known that interferon-y (IFNy) is not effective in inducing a sustained inhibition of HBV replication in patients with chronic HBV infection in contrast to interferon-a (IFNa). To determine whether this was related to IFNy receptor (IFNyR) underexpression, binding characteristics of IFNy to peripheral blood lymphocytes were studied in chronic HBV infection using r~dio~~dinated recombinant IFNy. Peripheral blood lymphocytes from patients with chronic HBV infection (n = 20), normal healthy controls (n = 12) and patients with non-viral related chronic liver disease expressed a similar number of IFNy-R (medians (ranges): 1891 (1581-2515); 1916 (1589-2441); 1893 (1692-2104) sites/cell, respectively, p = N.S.) with a similar dissociation constant (Kd = 0.7-2.7 nM). There was no correlation between IFNy-R expression and serum transaminase, serum HBsAg and HBV-DNA titres and liver histology. IFNa therapy in vivo also did not enhance IFNy-R expression (n = 3). There is therefore, no evidence from this data that IFNy-R is underexpressed in patients with chronic HBV infection to account for the difference in clinical response to these two forms of therapy.

Introduction Interferon-a (IFNa) is effective in inducing a sustained inhibition of hepatitis B virus (HBV) replication in 30-50% of the patients with chronic HBV infection (l-5). As measured by serum HBV-DNA, IFNy is also anti-viral in chronic HBV infection (6-8). In clinical trials, the use of IFNy in chronic HBV infection, either alone or in combination with IFNa, has been shown to be not effective in inducing sustained clearance of HBV although transient inhibition of HBV occurs (6-8). IFNs elicit their activities by binding to specific high affinity cell surface receptors (9). Despite similarities in many of their biological activities, IFNa and IFNy bind to different cell surface receptors (9,lO). Recent studies have demonstrated that patients with chronic HBV infection express IFNa receptor (IFNa-R) normally in peripheral blood mononuclear cells (11,12). To determine whether the poor response to IPNr therapy in chronic HBV infection is related to an abnormal expression of IFNy-R, we have studied the expression of IFNy-R on pe-

ripheral blood lymphocytes in patients with chronic HBV infection and compared with normal healthy controls and patients with non-viral-related chronic liver disease.

Matwb-:lsand Methods Twenty patients seropositive for hepatitis B surface antigen (HBsAg) for more than 12 months were studied. Thirteen patients were seropositive for HBV-DNA and all were negattive for hepatitis D virus and human immunodeficiency virus markers. None had received any antiviral or immunotherapy in the preceeding 6 months. Sixteen pat:ents had received IFNa treatment; thirteen of them received the therapy 12 months before the IFNy-R assay and four of them responded and seroconverted to anti-HBe; three of them received the IFNa therapy after the IFNy-R determination and all of them had their IF’NyR redetermined after receiving 4 or 8 weeks of IFNa therapy. The clinical data, HBV markers and liver histology of these patients are tabulated in Table 1.

Correspondence: Dr. Johnson Y.N. Lau, Institute of Liver Studies, King’s College Hospital, Denmark Hill, London SF5 9RS. U.K.

IFN-y RECEPTOR EXPRESSION IN CHRONIC HBV INFECTION

TABLE 1 ~IiniCal

data, HBV ma;kers and liver histology of the study popula-

tions

Number Male/female Median age (yrs) (range) Caucasians/Asians

Normal controls

Chronic HBV initclinn

Liver disease controlsa

12 11:l 32 (26-36) 9:3

20 l&2 34 (24-58) 17:3

8 4:4 41 (32-62) 8:0

IFNa treatment responderb non-responder

16 4 12

Serum HBV-DNA + HBsAg titre (median)

13 1:12 800

Liver histology CPHc CAHd (cirrhosis) not done

7 lO(2) 3

a See text for the diagnosis of these patients. b IF’Naresponder is defined as those who cleared sernm HBV-DNA and seroconverted from HBeAg to anti-HBe. ’ CPH, chronic persistent hepatitis. d CAH, chronic active hepatitis.

Two control populations wers: assessed simultaneously. The first group consisted of 12 healthy laboratory personel who were negative for HBsAg. The seoond group, the liver disease control group, consisted of eight patients with non-viral related chronic liver diseases (four primary biliary cirrhosis, two primary sclerosing cholangitis, one haemochromatosis and one Budd-Chiari syndrome with no evidence of blood dyscrasia). The demographic data of the controls are also summarized in Table 1. Serology Serum was tested for HBsAg, HBeAg, anti-HBe and total antibodies to hepatitis D virus by radioimmunoassay (Abbott Diagnostics, Maidenhead, U.K.). Antibodies to human immunodeficiency virus were tested by enzymoimmunoassay (Abbott Diagnostics). HBV-DNA was assayed by a modified spot hybridization technique described by Fagan et al. (13). Peripheral blood lymphocyte preparation Peripheral blood mononuclear cel!s were separated from heparinized blood by standard density gradient centrifugation on Dextran-Isopaque (Lymphopaque, Nycorned, Norway). Cells at the interface were carefully aspirated and washed three times with Hepes-buffered RPMI 1640 (Gibco Europe, Uxbridge, U.K.) supplemented with antibiotics (complete medium) and then resuspended in complete medium supplemented with 5%

295

heat-inactivaited fetal calf serum (Gibco Europe). J_ymphocyte-enriched fractions were then propared by depleting mOnOcyks using plastic adherence by incubating PBMC in Petri plastic dishes (Sterilin, Hounslow, U.K.) at 37°C for 1 h. Non-adherent cells were resuspended in complete medium and comprised more than 90% lympbocytes as determined by anti-CD2 (T-11, Dako, U.K.) on ~yt~spins using standard indirect immunocytochemical staining technique employing Naphthol salt and Fsst Red TR (Sigma) as the substrate (14). Viability of the lympbocytes exceeded 95% as assessed by Trypan blue dye exelusion. The cell counts were determined ith a baematocytometer and adjusted to 8 million cells/ml in complete medium. To account for the possibility that prior occupancy of cellular IFNy-R by endogenous IFNy might interfere subsequent radioligand receptor assay, cells from three healthy subjects and three patients with chronic HBV infection were washed with 0.05 M glycine-HCl (pH 3.5) for 3 min to release receptor bound IFNy followed by three washes with complete medium before performing receptor determination. The binding was essentially the same with unwashed cells, indicating that there was no or minimal IFNy binding to the cells in both controls and chronic HBV patients. Zodinationof ZFNyand ZFNy bioassay Recombinant human IFNy was kindly provided by Dr. G.R. Adolf, Boehringer Ingrlheim, Austria (lot No. 802 003/3), specific activity 2 . lo7 U/mg protein). Labelling with radioactive iodine (lzI) To high specificity was carried out using Chloramine T oxidation (15). Briefly, a 100 ,ul volume containing 20 ,ug IFNy, 1 mCi Na’=I (Amersham International, U.K.), 40 ,ug chloramine T in 0.1 M sodium phosphate buffer (pH 7.4) was reacted at 4°C for 5 min. The iodination was terminated by the addition of 80 pg sodium metabisulphite in 10 ~1 of sodium phosphate buffer. The radiolabelled IFNy was separated from the free lzsI by chromatography in a 10 cm Sephadex G-50 column (Phartnacia, U.K.) equilibrated in PBS containing 0.2% bovine serum albumin (BSA, Sigma). This was followed by dialysis with 2 litres of PBS/0.2% BSA overnight at 4°C. The label in these preparations was at least 95% protein-bound as determined by measurement of radioactivity that was precipitable by 10% trichloroacetic acid. The specific activity of [‘251]IFNa was between 30-40 Ci/g as determined by both biological assay (see below) and by competitive binding assay using Daudi cells (16). Radioiodination of IFNy by this method to 30-40 Ci/g did not denature the ligand as determined by cornpetiGve binding assay. Each batch of [12sI]IFNy was used for only 4-5 weeks to avoid radiolysis.

J.Y.N.

296 IFNY was assayed in human amnion cells (WISH), challenged with Semliki Forest virus by the method of inhibition of incorporation of [3H]uridine into viral nucleic acids, according to a protocol essentially described by Meager (17). IFNy-R binding assay Binding reactions were carried out in 200 ~1 volumes in duplicates. Briefly, 100 ,ul lymphocyte suspensions in the complete medium (8 - lO?ml) were added to 100~1 of appropriately concentrated radioiodinated IFNy (seven to eight different concentrations) in complete medium with 0.2% BSA. After incubation for 3 h at 4”C, the reaction mixtures were quickly layered on 200~1 of a phthalate oil mixture (2:1, v/v, mixture of dibutyl phthalate and dinonyl phthalate, BDH, U.K.) in a 1.5 ml conical tube and centrifuged for 60 s at 12 000 x g in a MSE microcentrifuge. The medium and the phthalate oil were then aspirated and the tip of the tube which contained the cell pellet was cut and cell-bound radioactivity determined on a high efficiency gamma-counter (Compugamma 1282, LKB, U.K.). Non-specific binding was determined in parallel incubations including a 200-250-fold excess of unlabelled IFNy at each radioligand concentrations. These non-specific counts varied from 30% to 50% of the total counts. Specific bindings were then obtained by subtracting the non-specific counts from the total bound counts. Analysisof binding data With increasing concentrations of [1251]IFNyin the cellular binding reaction, respective specific binding corresponding to each [‘251]IFNy concentrations was determined. These specific binding data were then plotted according to the method of Scatchard (18) and analysed by the LIGAND program of Munson and Rodbard (19). This program resolves the binding isotherms and determine objectively, the number of classes of receptors, the binding sites per cell and the dissociation constants (&). The & is practically defined a* the concentration of IFNy that is required to saturate 50% of the receptor sites on the cell surface (20). Determination of IFNy-R in a hepatoma cell line (PLCIPRFIS) PLC/PRF/S (Alexander) cell line (21), obtained from European Cell Type and Animal Cell Collection (ECACC), was maintained in Hepes-buffered RPM1 1640 supplemented with 10% fetal calf serum. Cells were seeded at approx. 2.5.106/well in 24-well plastic plates (Sterilin) at 37°C in a humidified incubator provided with 5% CO, and grown to confluence for 5 days. Confluent cell IaYers were then incubated with indicated concentra-

LAU

et al,

tions of [1251]IFNyin the absence or presence of 200-times of unlabelled IFNy in triplicate for 3 h at 4°C. The cells were then washed four times with phosphate-buffered saline (PBS) and solubilized in 1 ml of 1% sodium dodecyl sulphate, the suspensions were then counted with the gamma-counter. Cells from six parallel wells were trypsinised and the number of cells determined. Analysis of the binding data were as above. Statistics Results were analysed using Wilcoxon’s signed ranksum test and Spearman’s rank correlation test.

Results Binding characteristicsof [1251]IFNy The specific binding of [1251]IFNyto cells increased rapidly and maximal binding was reached after 90 min at 4°C and remained stable for at least 3 h (data not shown). In

3 2o t

0

Binding Isothwm

0

5

10

‘251-IFNy

15

20

(no/ml)

Scatchard Analysis

0.001

&COO

0

5

10

15

20

25

BOUND ‘251-iFN7 (pg) Fig. 1. Binding of [12Sl]lFNr to peripheral blood lymphocytes in a normal healthy control. (A) Saturable specific binding isotherm of [**sI]IFNy. Each point represents the mean of duplicates at each [1251]IFNy concentration. (B) Scatchard p:ot of the specific binding data as resolved by the LIGAND program. A linear curve indicates a single class of receptor. The number of receptors per cell is estimated by extrapolating the curve and reading the point zf intersection with the abscissa (1920 sites/cell). Dissociation constant is given from the slope of the curve (1.9*10-10 M).

Il=N-y RECEPTOR EXPRESSION IN CHRONIC HBV INFECTION

297

all subsequent experiments, incubations were carried out for 3 h at 4°C. With increasing concentrations of [‘*‘I]IFNy in the absence or presence of 200-fold excess unlabelled IFNy, specific binding curves were generated (Fig. 1A). The Scatchard plot analysis of the cell fractions from all patients were linear, indicating a single class of receptor (Fig. 1B). %FNy-R expression on peripheral blood lymphocyres

Comparison of the expression of IFNy-R on lymphocytes from normal controls (n = 12, median 1916 Sites/cell, range: 1589-244811, patients with chronic HBV infection (n = 20, median 1891 sites/cell, range: 1581-2515), and liver disease controls (n = 8, median 1893 sites/cell, range: 1692-2104) showed no difference between patients with chronic V infection and the two control groups (Fig. 2). When the patients with chronic HBV infection were stratified according to their serum HBV-DNA status, there was also no difference between the two groups (HBV-DNA +: median 1924 sites/cell, range: 1581-2515; HBV-DNA-: median 186Clsites/cell, range: 1764-2311; p = N.S.). All patients showed a similar Kd of l.9.10-‘” M (range: 0.7-2.7-10-‘” M) and no difference was observed between the groups. There was also no association between the IFNy-R expression (receptor numbers and Kd values) and either the clinical parameters (age, sex), the biochemical parameters (transaminase, albumin), the titre of serum HBsAg/HBV-DNA and the histological findings. IFNy-R expression was also unrelated to their previous clinical response to lFNa therapy. In three patients, IFNy-R expression was re-evaluated after receiving 4 (n = 1) or 8 (n = 2) weeks of lymphoblastoiq ir?terferon therapy 10 MIU thrice a week (Wellcome, U.K.), there was no change in the IFNy-R expresmedian 1966 sites/cell, range: sion (pre-treatmeEt:

i 9 f c m

3ooo1 25w-0

zooo--

u i

1500--

I& -

I

1rXB-l

COHTROL

HRV-DNA+ CHRONIC

HBV-DNA-

HRV

OTHER CLDS

Fig. 2. Expression of If N-y-R(sitesl’cell) in peripheral blood lymphocytes in normal healthy controls, patients with chronic HBV infection (serum HBV-DNA + /-) and patients with non-viral related chronic liver disease.

IO

BOUND ‘zsI-IFN7

(pg)

Fig. 3. Scatchard analysis of the binding isotherm of [lZ5I]IFNr fo PLC/PRF/j. The numbsr of I%Ny-R per cell was estimated to be 4260 and the Kd was 3.2~1W M.

1921-2112; 1816-2224).

post-treatment:

1910

sites/cell,

range

IFNy-R expression on PLCIPRFIS The average number of cells in six wells was 0.8-106

cells/well. Scatchard Analysis of the binding data is shown in Fig. 3. PLC/I’RF/S cells express a single class of receptor with approx. 4260 receptor sites/cell and a Kd of 3.2.10-10 M similar to that of peripheral blood lymphocytes.

iscossion The demonstration in the present study that patients with chronic HBV infection express IFNy-R normally in their peripheral blood lymphocytes indicates that the chronicity of viral infection is unlikely to be related to an abnormal IFNy-R status and is more likely to be caused by other factors like impaired production of IFNa, impaired post-receptor response or other immunological abnormalities. The observation of decreased expression of IFNy-R in patients with chronic active hepatitis and liver cirrhosis of various aetiologies reported by Kukumu et al. was not observed in our study (median IFNy-R expression 1996 sites/cell, n = 12) (22). The possible explanations for this difference are firstly, that various non-viral causes of liver diseases were also included in that study; and secondly, that some of the patients with liver cirrhosis in that study might have had decompensated liver diseases and high serum II+Jy levels which in turn downregulated the IFNy-R expression. The normal expression of IFNy-R also suggests that the failure of IFNy ther(.ipy in patients with chronic HBV infection i,s not related to an underexpression of IFNy-R. Furthermore, we have shown that treating patients with IFNa did not enhance the expression of IFNy receptor ex-

J.Y.N. LAU et al.

298

pression. Recently, the gene encoding IFNy-R on chromosome 6q21-22 was found to require the presence of a component encoded in human chromosome 21 (which contains the genes for IFNa-R and the transducer for IFNU action) for its activity suggesting that IFNa and 1FNy probably share a common transducer or that IPNY employs IFNa-R as a transducer (9,23). The absence of IFNy-R upregulation by IFNa and the fact that they may share a common transducer suggest that combination therapy employing both IFNa and IFNy may not be more effective than just using IFNa alone. Although most studies demonstrated only a single class of IFNy-R, Aiyer et al. showed that there may be two types of IFNy-R (24). Recent purification and cloning of the IFNy-R, however, indicates that there is probably only one single class of IFNy-R (25-27). The number of IFNy-R also varies between studies (28). Our estimation of IFNy-R (both numbers and & values) were in accord with most published studies (28). Furthermore, as the present study was a comparative study with two control groups, the assay itself should not alter the conclusions that we have drawn. It is possible that the expression of IFNy-R on peripheral blood lymphocytes may not represent the expression of IFNy-R on hepatocytes. However, study of IFNpR expression on hepatocytes is extremely difficult: (i) routine liver biopsy is insufficient to give the large number of cells required for binding assays; (ii) it is unethical to do surgical wedge biopsy on normal human controls and asymptomatic chronic HBV patients; (iii) the collagenase required during perfusion of wedge liver biopsy for the isolation of hepatocytes contains proteinase activity which will remove IFNy-R; and (iv) mechanical isolation of hepatocytes will disrupt the hepatocyte membranes, making the results of the binding assay unreliable. Nevertheless, we have demonstrated that there are specific IFNy-R on

1

.2lexanderGJM, BrahmJ, Fagan EA, et al. Loss of HBsAg with interferontherapyin chronichepatitisB virusinfection.Lancet

1987; ii%-69.. 2 Lok ASF, Novick DM, Karayiannis P, et al. A randomised study of the effects of adenine arabiuoside 5’ -monophosphate (short or long courses) and lymphoblastoid interferon on hepatitis B virus replication. Hepatology 1985; 5:1132-B. 3 Dusheiko G, Di-Bisceglie A, Bowyer S, et al. Recombinant leukocyte interferon treatment of chronic hepatitis B. Hepatology 1985; 5:556-60. 4 Scully LJ, Shein R, Karayiannis P, et al. Lymphoblastoid interferon therapy of chronic HBV infection. A comparison of 12 vs 24 weeks of thrice weekly treatment. J Hepatoll987; 4:51-B. 5 Hoofnagle JH, Peters M, Mullen KD, et al. Randomized controlled trial of recombinant human a-interferon in patients with chronic hepatitis B. Gastroenterology 1988; 95:1318-25.

the human hepatoma cell line PLC/PRP/S. Very recently, Volpes et al. using monoclonal antibody to IFNy-R and immunohistochemical technique, demonstrated that the expression of IFNy-R was increased in the liver tissue of patients with chronic HBV infection, especially over those inflamed areas (29). Combining with the data in the present study, these data indicated that the clinical ineffectiveness of IFNy in chronic HBV infection is not related to an abnormal expression of IFNy-R, either on peripheral blood lymphocytes or in liver tissue. The reason for the discrepancies between the clinical efficacy of IFNa and IFNy in chronic HBV infection remains.unresolved. Recently, employing primary hepatocyte culture using HBV-infected hepatocytes isolated from patients with chronic HBV infection, Bain et al. showed that IFNa treatment in vitro, in a concentration similar to that achieved in clinical studies, differentiakj augmented the production of intracellular HBsAg and pre-Sz and switched the localisation of HBsAg from diffuse cytoplasmic to membraneous/submembraneous, in a way that may enhance immune recognition (30). Using a similar technique, IFNy suppresses all the HBV antigens tested and it is suggested that this difference may be a factor explaining their different clinical effects on chronic HBV infection (31).

Acknowledgements

The authors would like to thank Dr. G.R. Adolf, Boehringer Ingelheim, Austria, for supplying the recombinant human interferon-y and Dr. P.J. Munson, National Institute of Health, Bethesda, U.S.A. for providing the LIGAND program. J.Y.N. Lau was a Croucher Foundation Fellow.

6 Bissett J, Eisenberg M, Gregory P, Robinson WS, Merigan TC. Recombinant fibroblast interferon and immune interferon for treating chronic hepatitis B virus infection: patients’ tolerance and the effect on viral markers. J Infect Dis 1988;157:1076-80. 7 Gomez C, LaBanda F, Porres JC, et al. Combined recombinant alpha and gamma interferon treatment of chronic hepatitis B virus infection. In: Zuckerman AJ, ed. Vial Hepatitis and Liver Disease. New Y&: Alan R. Liss, 1988:872-4. 8 Di-Bisceglie AM, Rustgi VK, Kassianides C, et al. Therapy of chronic hepatitis B with recombinant human alpha and gamma interferon. Hepatology 1990; 11:266-70. 9 Langer J, Pestka S. Interferon receptors. Immunol Today 19BB; 9:393-400. 10 BraI?ca AA, Baglioni C. Evidence that type I and II interferons have different receptors. Nature 1981; 294768-70. 11 Dooley JS, Vergalla J, Hoofnagle JH, Zoon KC, Munson PJ, Jones EA. Specific binding of human alpha interferon to high affinity cell surface binding sites on peripheral blood mononuclear

IFN-y RECEPTOR

EXPRESSION

IN CHRONIC

MBV INFECTION

cells. J Lab Clin Med 1989; 113623-31. 12 Lau JYN, Sheron N, Morris AG, Bomford AB, Alexander GJM, Williams R. Interferon-alpha receptor expression and regulation in chronic hepatitis B virus infection. Hepatology 1991; 13:322-g. 13 Fagan EA, Guarner P, Perera SDK, et al. Quantitation of hepatitis B virus DNA in serum using the spot hybridisation technique and scintillation counting. J Virol Methods 1985; 12:251-62. 14 Cordell JL, Falini B, Erber WN, et al. Immunoenzymomatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase. J Histochem Cytochem 1954; 32:219-28. 15 Hunter WM, Greenwood FC. Preparation of Iodine-131 labelled human growth hormone of high specific activity. Nature 1962; 194495-6. 16 Calve JC, Radicella JP, Charreau EH. Measurement of specific radioactivities in labelled hormones by self-displacement analysis. Biochem J 1983; 212:259-64. 17 Meager A. Quantification of interferons by anti-viral assays and their standardization. In: Clemens MJ, Morris AG and Gearing AJH eds. Lymphokines and Interferons: a Practical Approach. Or ford: IRL press, 1989; 129-47. 18 Scatchard G. The attraction of protein for small molecules and ions. Ann NY Acad Sci 1949; 51660-72. 19 Munson PJ, Rodbard D. LIGAND: a versatile computerized approach for characterization of ligand binding systems. Anal Bio&em 1980; 107: 220-39. 20 Lau AS, Read SE, Williams BRG. Downregulation of interferon Q.but not y receptor expression in vivo in the acquired immunodeficiency syndrome. J Clin Invest 1988; 82: 1415-21. 21 Alexander JJ, Bey EM, Geddes EW, Lecatsas G. Establishment of a continously growing cell line from primary carcinoma of the liver. S Afr Med J 1976; 50:2124-g.

22 Kakuma S, Yosbioka K, Fuji A, 5’ahara . Interferon-y ret-q_ tors on T cells in patients with ch: onic liver disease. Heptogastroenterology 1988; 35:158-61. 23 Jung V, Rashidbaigi A, Jones C, et al. Human chromosomes 6 and 21 are required for sensitivity to human interferon-y. Proc Nat1 Acad Sci USA, 1987; 84:4151-5. 24 Aiyer RA, Serrano LE, Jones PP. Interferon-y binds to high and low-affinity receptor components on murine macrophages. J Pmmunol1986; 136:3329-34. 25 Stefanos S, Ahn YH, Pestka S. Characterisation of human interferon-gamma receptor purified from placenta. J Interferon Res 1989; 9:719-30. 26 Hansen BD, Finbloom DS. Characterisation of the interaction between recombinant human interferon-gamma and its receptor on human polymorphonuclear leukocytes. J Leukoc Biol 1990; 4764-g. 27 Jung V, Jones C, Kumar CS, Stefanos S, O’Connell S, Pestka S. Expression and reconstruction of a biologically active human interferon-gamma receptor in hamster cells. J Biol Chem 1990; 265: 1827-30. 28 Rubinstein M, Orchansky P. The interferon receptors. CRC Crit Rev Biochem 1986; 21:249-75. 29 Volpes R, van der Oord JJ, De Vos R, Delpha E, De Ley M, Desmet VJ. Expression of interferon-gamma receptor in normal and pathological HBV-infected human liver tissues. J Hepatol 1991; 12:195-202. 30 Bain VG, White YS, Naoumov NV, et al. Differential effect of ainterferon (IFN) on HBV antigen expressed in cultured hepatocytes. Hepatology 1989; 10642 (Abstr.). 31 Lau JYN, Bain VG, Naoumov NV, Smith HM, Alexander GJM, Williams R. Effects of interferon-y on hepatitis B virus (HBV) antigens expression contrast that with interferon-a in primary hepatocyte culture. J Hepatoll990; (Abstr.) 538.