Up-regulation of intercellular adhesion molecule 1 transcription by ...

Proc. Natl. Acad. Sci. USA Vol. 89, pp. 11441-11445, December 1992 Immunology

Up-regulation of intercellular adhesion molecule 1 transcription by hepatitis B virus X protein (immune response/interferon y)

KE-QIN Hu, CHANG-HONG YU, AND JOHN M. VIERLING Department of Medicine, Cedars-Sinai Medical Center and University of California at Los Angeles School of Medicine, Los Angeles, CA 90048-0750

Communicated by Philippa Marrack, July 21, 1992

ABSTRACT Intercellular adhesion molecule 1 (ICAM-1), a counter-receptor for lymphocyte function-associated antigen 1 on T cells, is critically important to a wide variety of adhesion-dependent leukocyte functons, including antigen presentation and target cell lysis. ICAM-1 expression by hepatocytes is increased in areas of inflammation and necrosis during chronic hepatitis B. Whether induction of ICAM-1 is due to the effect of inflammatory cytokines or involves a direct effect of the hepatitis B virus (HBV) remains unknown. In the present study, transfection of the HBV genome into human hepatoma cell lines resulted in enhan expression of ICAM-1 protein and RNA in the absence of inflammation. Results of subgenomic transfections indicated that the HBV X protein (pX) induced ICAM-1 expression. Nuclear run-on assays showed that pX induced the ICAM-1 gene by increasing its rate of transcription. Although both pX and interferon y induced transcription of ICAM-1, addition of interferon y to cells expressing pX did not show an additive or synergistic effect. These results indicate that pX can directly regulate ess of ICAM-1 and may participate in the immunopathogene of HBV infection.

The hepatitis B virus (HBV) genome contains four open reading frames (ORFs): pre-S/S, core/e, X, and Pol (1, 2) (Fig. 1). The X ORF encodes a polypeptide of 16.5 kDa, which is designated as X protein (pX). Expression of pX elicits an antibody (anti-pX) response in HBV-infected individuals (3-6). pX transactivates a variety of viral and human genes, including those of the human immunodeficiency virus and class I and II major histocompatibility complex (MHC) molecules (7-13). Since class I and II MHC genes are involved in antigen-specific immune responses of T cells, their transactivation by pX suggests that HBV may directly participate in the immunologic response of the host to infected hepatocytes. HBV infection results in several distinct outcomes: acute and chronic hepatitis, a chronic carrier state, and development of hepatocellular carcinoma (1, 2). It is generally accepted that hepatocellular necrosis during HBV infection is the result of the host immune response against one or more HBV antigens expressed on the surface of infected hepatocytes (14-18). Extensive studies indicate that cytotoxicity mediated by T cells is the principal mechanism (14-18). The asymptomatic HBV carrier state appears to represent a state of clonal anergy in which the T cells remain functionally unresponsive to infected hepatocytes (13, 19, 20). Following treatment and withdrawal of immunosuppressive medications or chemotherapeutic agents, chronic carriers may develop aggressive hepatitis (20), indicating the presence of previously unresponsive T cells. However, the viral and/or host immunologic mechanisms responsible for different clin-

\xz~ -A~

-Enh I

Cp Enh H

FIG. 1. Genetic organization ofthe HBV genome. Four ORFs are shown: S, pre-S/S gene; C, core/e gene; X, X gene; P, DNA polymerase gene. Black boxes represent the promoters, P1, P2, Xp, and Cp; hatched boxes denote enhancer elements, EnhI and EnhIl.

kb, Kilobases.

ical outcomes ranging from fulminant hepatitis to the asymptomatic carrier state remain undefined (14, 15). Recent evidence indicates that optimal interaction of TCR with antigen-MHC complexes requires concurrent binding of adhesion molecules on T cells with their counter-receptors on either antigen-presenting cells or target cells (21, 22). Lymphocyte function-associated antigen 1 (LFA-1), an adhesion molecule of the integrin family, is expressed on T cells (21, 22). LFA-1 binds to its counter-receptor, intercellular cell adhesion molecule 1 (ICAM-1), a cell surface glycoprotein that is a member of the immunoglobulin superfamily (21-24). Binding of LFA-1 to ICAM-1 is critically important to a wide variety of adhesion-dependent leukocyte functions (24-26). For example, antibody against ICAM-1 can block T-cell cytotoxicity, and cotransfection of fibroblasts with ICAM-1 and MHC class II cDNAs is required for activation of T-helper cells (27, 28). Inflammatory cytokines, such as interferon y (IFN-y), interleukin 1, and tumor necrosis factor, induce ICAM-1 expression and potentiate the immune response (21, 22). Although normal hepatocytes do not express ICAM-1, expression by hepatocytes has been observed in inflamed biopsy specimens from patients with HBV infection (29-31). Whether ICAM-1 expression is induced indirectly by inflammatory cytokines (21, 22) or directly by HBV proteins remains unknown. To study this issue, we assessed the effects of HBV genomic and subgenomic transfection of human hepatoma cell lines on the in vitro expression of ICAM-1 protein and RNA. Our results indicate that pX induces ICAM-1 expression through a transcriptional mechanism in hepatoma cell lines. Abbreviations: HBV, hepatitis B virus; ICAM-1, intercellular adhesion molecule 1; LFA-1, lymphocyte function-associated antigen 1; MHC, major histocompatibility complex; IFN, interferon; ORF, open reading frame; HBsAg, hepatitis B surface antigen.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

11441

11442

Immunology: Hu et al.

MATERIALS AND METHODS Plasmid Vectors. All HBV plasmid vectors used in this study were derived from a cloned HBV adw strain (Fig. 1) (32). As previously described (13, 33), pNER contains the entire HBV genomic DNA. The plasmid pUC13 was used to construct the following subgenomic HBV vectors: pUCSL (containing the OREs for pre-S/S and X) and pUCC (containing the ORFs for core/e and X) (13). The plasmid vector pMNX (containing the X ORF alone) was a gift of Aleem Siddiqui (13). Plasmid pICAM-1, kindly provided by Brian Seed, contains a 1.85-kb cDNA sequence of human ICAM-1 (24). Celi Lines and Transfection. HepG2, HuH-7, and Hep3B human hepatoma cell lines were used. HepG2 and HuH-7 are negative for HBV DNA, whereas Hep3B cells express and secrete hepatitis B surface antigen (HBsAg) (34, 35). Cell culture and DNA transfection were performed as reported (13, 33). Specifically, calcium phosphate precipitation was used for DNA transfection (13, 33). MNX, a stably transformed cell line constitutively expressing pX, was established by transfection of HepG2 cells with pMNX, followed by selection with G418 (13). To assess the stimulation of ICAM-1 by IFN-y, cell lines were cultured for 24 hr in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10%1 fetal bovine serum and recombinant IFN-y (Genzyme) at 200 units/ml. This concentration of IFN-y is optimal for ICAM-1 induction (24). Cytoplasmic RNA was then isolated from these cells and analyzed for ICAM-1 transcripts by slot hybridization. lmnunocytochemical Assays. Expression of HBV peptide antigens and ICAM-1 in transfected cells was detected by an indirect immunoperoxidase technique (13, 17, 18). The specificities of the monoclonal antibodies against HBV peptides have been reported (13, 17, 18). Monoclonal antibody against human ICAM-1 (84H10) was obtained from AMAC (Westbrook, ME). Peroxidase-labeled anti-mouse IgG was purchased from Sigma. As negative controls, HepG2 and HuH-7 cells, either nontransfected or transfected with pUC13 (the plasmid vector used to construct plasmids pUCSL and

pUCC), were employed. Southern Blot Hybridization. Southern analysis was performed as reported (36), using 4 ,ug of genomic DNA isolated from nontransfected HepG2 and MNX cell lines. After isolation, DNA extracts were digested with BamHI or EcoRI endonuclease (GIBCO/BRL), electrophoresed through a 0.8% agarose gel, and transferred to a Nytran membrane (Schleicher & Schuell). ICAM-1 DNA was hybridized with the 32P-labeled ICAM-1 cDNA probe and detected by autoradiography. Assays for ICAM-1 RNA. Cytoplasmic RNA from transiently transfected or stably transformed cell lines was isolated by a modification of the method of Chirgwin et al. (37). RNA slot and Northern blot hybridizations were carried out by loading 10 ,ug and 20 ,ug of total cytoplasmic RNA, respectively. RNA was probed with 32P-labeled ICAM-1 cDNA according to standard methods (38). Nuclear run-on assays for ICAM-1 were performed in HepG2 and MNX cells as described (13, 33, 39), except that 32P-labeled nuclear RNA was extracted by an acid phenol/guanidinium thiocyanate procedure (40) rather than trichloroacetic acid precipitation (39). All hybridization signals were analyzed quantitatively with a computerized densitometer (Helena Laboratories).

RESULTS HBV and ICAM-1 Protein Expression in HBV DNATransfected Hepatoma Cells. Expression of HBV peptide antigens following transfection was verified in the culture medium and in cells by radioimmunoassay (RIA) and immu-

Proc. Natl. Acad. Sci. USA 89 (1992)

nocytochemistry, respectively (13). Transfection of HepG2 and HuH-7 cells with the HBV genome resulted in expression of pre-S/S, core/e, and pX. Transfection with pUCSL resuIted in expression of pre-S/S and pX, while transfection with pUCC resulted in expression of core/e and pX. Cells transfected with pMNX expressed pX alone. MNX, a stably transformed HepG2 cell line transfected with pMNX (13), showed greater expression of pX than transiently transfected cells, and pX mRNA was readily detected (data not shown). Native HepG2 and HuH-7 cells or those transfected with the control vector pUC13 expressed a small amount of ICAM-1 on their surface membranes (Fig. 2A). Expression of ICAM-1 on these cells was increased following transfection with whole recircularized HBV genome (Fig. 2B). Transfection of HepG2 cells with the subgenomic vector pUCSL, pUCC, or pMNX also resulted in increased expression of ICAM-1. Hep3B cells, which express HBsAg alone (34), exhibited scant expression of ICAM-1 (Fig. 2C). MNX, a stably transformed HepG2 cell line expressing pX only, expressed a greater density of ICAM-1 than cells transiently transfected with pMNX (Fig. 2D). These data indicate that HBV transfection can directly induce ICAM-1 expression and that pX, the only common product among the genomic and subgenomic vectors, might play an important role. Assays of ICAM-1 Genonc DNA. Following digestion of HepG2 and MNX cellular genomic DNA with BamHI or EcoPJ, the predominant hybridization signals obtained with the ICAM-1 cDNA probe were at 6.3 and 4.5 kb, respectively. Densitometry indicated that native HepG2 and MNX cells expressed comparable amounts of ICAM-1 genomic DNA (data not shown). Thus, the increased ICAM-1 expression observed in MNX cells was not due to increased quantities of ICAM-1 DNA. Assays of ICAM-1 Transcripion. To determine whether regulation of ICAM-1 expression induced by HBV peptide antigens occurred at the level of transcription, a series of RNA assays was performed. In ICAM-1 RNA slot assays (Fig. 3), nontransfected HepG2 and HuH-7 cells showed faint background signals for ICAM-1 RNA. Transfection of these cells with recircularized HBV genome resulted in an 3-fold increase in ICAM-1 RNA expression. Hep3B cells, expressing HBsAg only, showed a scant background expression of ICAM-1 RNA. Transient transfection of HepG2 with pUCSL (pre-S/S and X) or pUCC (core/e and X) increased expression of ICAM-1 RNA. Furthermore, transient transfection of HepG2 cells with pMNX also caused enhanced expression of ICAM-1 RNA. A maximal 5-fold increase in expression of ICAM-1 RNA was observed in the MNX cell line. To evaluate the relative quantities of RNA in slot hybridization assays, membranes were rehybridized with a specific albumin probe. Signals of comparable intensity were observed among these diverse samples (Fig. 3). Northern blots showed that HepG2 cells transfected with the HBV genome and MNX cells expressed two species of ICAM-1 RNA of approximately 3.2 kb and 1.9 kb (Fig. 4). The 1.9-kb species, was partially interrupted by 18S rRNA. Both sizes are consistent with those previously reported for ICAM-1 mRNA (24). The 3.2-kb species represented the major transcript of ICAM-1 (93.3%), whereas the 1.9-kb species represented a minor species (6.7%). In contrast, hybridization signals were very weak in nontransfected HepG2 and Hep3B cells. Expression of ICAM-1 transcripts in HepG2 cells was increased -2-fold in cells transiently transfected with recircularized HBV genome and -4-fold in permanently transformed MNX cells compared with nontransfected HepG2 cell controls. Rehybridization of membranes with a cDNA probe specific for albumin confirmed that comparable amounts of RNA were loaded into each lane (data not shown). Transfection with HBV genomic or subgenomic DNA did not change the molecular size of ICAM-1

Immunology: Hu et aL

Proc. Natl. Acad. Sci. USA 89 (1992)

11443

B

A

.'t.

.,*A;A

D

C

FIG. 2. Indirect immunoperoxidase staining of ICAM-1 in hepatoma cells transfected with HBV DNA sequences. Nontransfected HepG2 cells exhibited scant ICAM-1 (A), HepG2 cells transfected with recircularized HBV genome showed increased ICAM-1 expression (B), Hep3B cells showed no ICAM-1 expression (C), and MNX cells showed dense ICAM-1 expression (D). (x220.)

transcripts observed in nontransfected HepG2 or Hep3B cells. These data indicate that pX can induce the expression of both mRNA species of ICAM-1. To test whether enhanced expression of ICAM-1 RNA was due to a transcriptional or posttranscriptional mechanism, nuclear run-on assays were performed with nuclei from MNX

A Hep3B HepG2(HBV)

B .,

HepG2

HepG2 (S& X)

_ww

HepG2(C&X)

*iI_

HepG2( X )

a'.:;

MN X

IWFr

C

and nontransfected HepG2 cells. This technique determines the relative RNA polymerase loading on a gene and hence its transcriptional activity (13, 39). As expected, transcription of the HBV X gene was observed exclusively in MNX cells (Fig. 5). Transcription of the albumin gene was similar in both the MNX and the HepG2 cells. The transcriptional rate of the ICAM-1 gene was increased >2-fold in MNX cells compared with nontransfected HepG2 cells. These data demonstrate that transcription of the ICAM-1 gene is increased selectively in the MNX cell line, which expresses pX. Thus, pX appears to enhance transcription of the ICAM-1 gene. Regulation of ICAM-1 Transcription by pX and IFN-y. A number of inflammatory cytokines, including IFN-y, regulate the expression of ICAM-1 at the transcriptional level (21, 22). To compare the effects of pX and IFN-y on the expression of ICAM-1, nontransfected HepG2, HuH-7, and MNX cells were cultured for 24 hr in medium supplemented with IFN-y at 200 units/mi. Cellular RNA was then extracted for slot assays. ICAM-1 RNA expression was increased by the addition of IFN-y to the culture medium of HepG2 and HuH-7 cells. In contrast, the addition of IFN-y did not

D

A

B

C

D

HuH-7 HuH -7(HBV) n

4-".

FIG. 3. RNA slot assays of ICAM-1. Samples (10 pzg) of total cellular RNA were loaded on Nytran membrane (Schleicher & Schuell), and ICAM-1 cDNA was labeled with [a-32P]dCTP and used as the probe. (A) From top to bottom: Hep3B; HepG2 transiently transfected with recircularized HBV genome; nontransfected HepG2; HepG2 cells transiently transfected with pUCSL (S & X); pUCC (C & X), or pMNX (X alone); and MNX cells, which are permanently transformed with pMNX. (C) Nontransfected HuH-7 cells and HuH-7 cells transiently transfected with recircularized HBV genome. (B and D) Membranes were rehybridized with a 32P-labeled cDNA probe specific for albumin as an internal control for the quantity of RNA.

28S

18S-

FIG. 4. Northern blot assays for ICAM-1. Samples (20 j&g) of total cellular RNA from Hep3B cells (lane A), nontransfected HepG2 cells (lane B), HepG2 cells transfected with recircularized HBV genome (lane C), and MNX cells (lane D) were electrophoresed through 1% agarose gel, transferred to Nytran membranes, and hybridized with the 32plabeled ICAM-1 cDNA probe.

11444

Proc. Nati. Acad. Sci. USA 89 (1992)

Immunology: Hu et al. HepG2 A b

.j

..i, i;

!CAM - 1

=:

pX

FIG. 5. Nuclear

tant roles in the immunopathogenesis of hepatocellular necrosis by altering either the afferent or the efferent components of the immune response. Thus, we used an in vitro expression system for HBV (13, 33) to study the effect of HBV proteins on the expression of ICAM-1. Our results clearly demonstrate that HBV gene products can induce ICAM-1 expression in vitro by human hepato-

MN X

..

:. .:.I

run-on assays

of ICAM-1

gene.

The DNA

fragments [albumin (Alb) cDNA, ICAM-1 cDNA, and HBV X-gene DNA (encoding pX)] were loaded on Nytran membranes and probed with 32P-labeled rmn-on-generated RNA from nontransfected HepG2 and MNX cells.

augment the already enhanced expression ofICAM-1 RNA in

MNX cells (Fig. 6).

DISCUSSION Extensive studies have indicated that HBV peptide antigens, such as pre-S/S and core/e, are expressed on the surface of hepatocytes during HBV infection and that membranous expression is associated with inflammatory infiltrates predominantly composed of T cells (16-18). It is widely believed that the host immune response against HBV antigen(s) mediates hepatocellular necrosis, since HBV does not appear to be directly cytopathic in vitro or in the chronic HBV carrier state (14, 15). Immunohistochemical studies of liver biopsy samples from patients with HBV infection and hepatocellular necrosis have demonstrated increased expression of class I MHC antigens and de novo expression of both class II MHC antigens (14, 15) and ICAM-1 (15, 30, 31). In contrast, normal human hepatocytes express small amounts of class I MHC molecules, whereas class II MHC molecules and ICAM-1 are absent (15, 29). Since inflammatory cytokines can induce in vitro synthesis of class I and II MHC molecules and ICAM-1 (15, 21, 22), expression of these molecules in inflamed liver biopsy samples has been attributed to cytokines. Recent studies demonstrating that pX can transactivate both class I and II (12, 13) MHC genes in vitro has provided an alternative explanation. Optimal interaction of CD4 and CD8 T cells with either antigen-presenting cells or target cells requires concurrent binding of T-cell antigen receptors to antigen-MHC complexes and binding of adhesion molecules to their counterreceptors (24-26). Of particular importance is ICAM-1, which serves as a counter-receptor for LFA-1 expressed by mature T cells (21, 24). Coexpression of ICAM-1 and class II MHC molecules is required to activate T-helper cells (28). Moreover, not only does induction of ICAM-1 on target cells increase susceptibility to lysis by cytotoxic T cells, but lysis can be blocked by anti-ICAM-1 antibody (27). Expression of ICAM-1 by HBV-infected cells, therefore, could play imporHepG2

HuH-7

MNX

IFN-'Y (-)

IFN--Y (+)

*qwj+l

FIG. 6. RNA slot hybridization of ICAM-1. Samples (10 pg) of total cellular RNA from nontransfected HepG2, nontransfected HuH-7, and MNX cells cultured in medium without (upper slots) or with IFN-y (lower slots) were loaded on Nytran membranes and hybridized with the 32P-labeled ICAM-1 cDNA probe.

blastoma and hepatocellular carcinoma cell lines in the absence of inflammatory cells or their cytokines. Nontransfected HepG2 and HuH-7 cells showed a low background expression of both ICAM-1 protein and ICAM-1 RNA. In contrast, transfection of either cell line with recircularized whole HBV genome significantly increased expression of ICAM-1 protein and RNA. Transfection of the subgenomic vectors pUCSL (containing pre-S/S and X ORFs) and pUCC (containing core/e and X ORFs) also significantly increased expression of ICAM-1 protein and RNA. Since the subgenomic vectors shared only the X ORF in common, the data suggest that pX may have directly induced ICAM-1 RNA, leading to enhanced expression of ICAM-1 protein. Indeed, the permanent cell line MNX, created by transfection of HepG2 cells with pMNX (X ORF alone), showed the greatest increase in ICAM-1 protein and RNA expression. Scant background expression of ICAM-1 by Hep3B cells (expressing only HBsAg) indicated that the transactivation potential of the pre-S/S ORF (41) was unlikely to be involved in those vectors with an intact pre-S/S ORF. To exclude the possibility that pX merely altered the amount ofICAM-1 genomic DNA, Southern blot assays were performed with nontransfected HepG2 and MNX cells. Since hybridization signal intensities for ICAM-1 were comparable, induction of ICAM-1 by pX could not be attributed to an alteration of ICAM-1 genomic DNA. However, the molecular size of ICAM-1 DNA in both the nontransfected HepG2 cells and the MNX cells differed from that reported by Staunton et al. (36). The reason for this difference remains

unexplained. To assess whether the induction of ICAM-1 RNA by pX involved a transcriptional or a posttranscriptional mechanism, nuclear run-on assays were performed using nuclei from nontransfected HepG2 and MNX cells. As expected, only MNX cells transcribed X. Although the transcriptional rates for the albumin gene were similar for both cell lines, the transcriptional rate of the ICAM-1 gene was increased nearly 3-fold in the MNX cells. Thus, pX induced ICAM-1 expression through a transcriptional mechanism. Although recent studies have shown that influenza and respiratory syncytial viruses can induce or suppress ICAM-1 expression in human mononuclear leukocytes (42), to our knowledge, the present study is the first demonstration that a viral protein, pX, can induce ICAM-1 expression through the mechanism of transactivation. HBV pX acts as a transactivator to stimulate expression of several viral and cellular genes (7-13). Studies of animal inoculation and transgenic mice also have shown that pX can induce hepatocellular carcinoma (43, 44). Although specific mechanism(s) of pX transactivation are incompletely understood, it appears that pX functions primarily by binding to intracellular response elements. For example, NF-KB and AP2 sequence motifs have been proposed as sites for pX transactivation (45, 46). pX has been shown to alter the DNA binding specificity of the CREB and ATF-2 transcription factors through protein-protein interaction (47). Binding of pX to the large tumor antigen of simian virus 40 can prevent transactivation by pX (48). Evidence that pX has protein kinase activity (49) and amino acid sequence homology to a seine protease inhibitor (50) suggests that enzymatic activity may also be involved in the transactivation function of pX. Thus, pX may activate gene transcription by different mechanisms depending on the availability of specific factors pres-

Proc. Natl. Acad. Sci. USA 89 (1992)

Immunology: Hu et al. ent in the infected cells. Further studies will be required to determine which cellular factor(s) are involved in pX trans-

activation of ICAM-1. Since IFN-y also can induce ICAM-1 expression at the transcriptional level (24) and stimulate expression of both class I and II MHC molecules (12, 13), we studied the effect of IFN-y on the expression of ICAM-1 in nontransfected HepG2 and HuH-7 cells and the MNX cell line. Binding of IFN-y to its receptor induces synthesis of IFN-stimulated gene factor 3y (ISGF3y), which, in turn, transactivates IFN-stimulated response elements present in IFN-inducible genes (51, 52). In contrast, pX transactivation appears to involve formation of intracellular pX-protein complexes capable of binding to specific gene response elements (47, 48). As expected, optimal concentrations of recombinant human IFN-y (24) induced increased ICAM-1 RNA expression in both nontransfected HepG2 and HuH-7 cells. However, IFN-y did not augment the increased level of ICAM-1 RNA expression in MNX cells. There are two possible interpretations of these results. First, it is possible that transactivation of the ICAM-1 gene by pX was maximal and that IFN-y could not exert an additive effect. Second, pX might have inhibited the formation or function ofISGF3y, a phenomenon observed with adenovirus ElA protein (52). Elucidation of the specific mechanism(s) of pX induction of ICAM-1 will be required to evaluate these possibilities. In view of the putative immunopathogenesis of hepatocellular necrosis in HBV infection, pX transactivation of class I and II MHC molecules (12, 13) and ICAM-1 in hepatocytes could play an important role in the initiation of the host immune response to HBV infection. Experimentally, two signals appear to be necessary for clonal expansion and differentiation of functional T cells (19, 53). The absence of a costimulatory second signal results in functional clonal inactivation or clonal anergy (19, 53). Thus, it is interesting to speculate that concurrent hepatocyte expression of ICAM-1 and HBV antigen(s) complexed with class II MHC molecules could produce distinct clinical outcomes. If infected hepatocytes provided both first and second signals, immune clearance of infected cells would result and infection would be transient, as observed in most clinical cases (1, 2). In contrast, if infected hepatocytes provided only a first signal, clonal anergy could result in a chronic HBV carrier state as a reservoir of infection. Indeed, available evidence indicates that the chronic HBV carrier state involves clonal anergy, since anergy can be terminated by transient immunosuppression followed by spontaneous reconstitution of the host immune system (20). Further investigation is required to evaluate the potential role of pX transactivation of class II MHC and ICAM-1 genes in the initiation of the host immune response to HBV infection and the capacity of infected hepatocytes to provide a costimulatory second signal. We thank Dr. Aleem Siddiqui for providing plasmid pMNX, Dr. Brian Seed for providing plasmid pICAM-l, Dr. Barbara B. Knowles for providing HepG2 and Hep3B cell lines, and Dr. John Prehn for his critical discussion of this work. 1. Tiollais, P., Pourcel, C. & Rejean, A. (1985) Nature (London) 317, 489-495. 2. Ganem, D. & Varmus, H. E. (1987) Annu. Rev. Biochem. 56, 651-693. 3. Moriarty, A. M., Alexander, H., Lerner, R. A. & Thornton, G. B. (1986) Science 277, 429-433. 4. Siddiqui, A., Jameel, S. & Mapoles, J. (1987) Proc. Nat!. Acad. Sci. USA 84, 2513-2517. 5. Hu, K. Q., Hao, L. J. & Will, H. (1988) Chinese Med. J. 101, 671-674. 6. Wang, W., London, W. T., Lega, L. & Feitelson, M. A. (1991) Hepatology 14, 29-37. 7. Twu, J. S. & Schloemer, R. H. (1987) J. Viro!. 61, 3448-3453.

11445

8. Siddiqui, A., Gaynor, R., Srinivasan, A., Mapoles, J. & Farr, R. W. (1989) Virology 169, 479-484. 9. Zahm, P., Hofschneider, P. H. & Koshy, R. (1988) Oncogene 3, 169-177. 10. Spandau, D. & Lee, C. H. (1988) J. Virol. 62, 427-434. 11. Twu, J. S., Chu, K. & Robinson, W. (1989) Proc. Nati. Acad. Sci. USA 86, 5168-5172. 12. Zhou, D. X., Tarboulos, A., Ou, J. H. & Yen, T. S. B. (1990) J. Virol.

64,4025-4028.

13. Hu, K. Q., Vierling, J. M. & Siddiqui, A. (1990) Proc. Nat!. Acad. Sci. USA 87, 7140-7144. 14. Mondeili, M., Manns, M. & Ferrari, C. (1988) Arch. Pathol. Lab. Med. 112, 489-497. 15. Peters, M., Vierling, J., Gershwin, M. E., Milich, D., Chisari, F. V. & Hoofnagle, J. H. (1991) Hepatology 13, 977-994. 16. Eggink, H. F., Houthoff, H. F., Huitema, S., Wolters, G., Poppema, S. & Gips, C. H. (1984) Clin. Exp. Immunol. 56, 121-128. 17. Hu, K. Q., Song, P. H. & Hao, L. J. (1987) Chinese J. Pathol. 16, 86-89. 18. Hu, K. Q., Hao, L. J., Zhang, Y. Y. & Wang, Y. K. (1989) Am. J. Gastroenterol. 84, 1538-1542. 19. Mueller, D. L., Jenkins, M. K. & Schwartz, R. H. (1989) Annu. Rev. Immunol. 7, 445-480. 20. Hoofnagel, J. H., Dusheiko, G. M., Schafer, D. F., Jones, E. A., Micetich, K. C., Young, R. C. & Costa, J. (1982) Ann. Intern. Med. 96, 447-449. 21. Springer, T. A. (1990) Nature (London) 346, 425-434. 22. Dustin, M. L. (1990) BioEssays 12, 421-427. 23. Makgoba, M. W., Sandars, M. E., Luce, G. E. G., Dustin, M. L., Springer, T. A., Clark, E. A., Mannoni, P. & Shaw, S. (1988) Nature

(London) 331, 86-88.

24. Simmons, D., Makgoba, M. W. & Seed, B. (1988) Nature (London) 331, 624-627. 25. Dustin, M. L. & Springer, T. A. (1988) J. Cell Biol. 107, 321-330. 26. Dustin, M. L. & Springer, T. A. (1989) Nature (London) 341, 619-624. 27. Makgoba, M. W., Sanders, M. E., Lunce, G. E., Gugel, E. A., Dustin, M. L., Springer, T. A. & Shaw, S. (1988) Eur. J. Immunol. 18, 637-640. 28. Altmann, D. M., Hogg, N., Trowsdale, J. & Wilkinson, D. (1989) Nature (London) 338, 512-514. 29. Simith, M. E. F. & Thomas, J. A. (1990) J. Clin. Pathol. 43, 893-900. 30. Volpes, R., Van den Oord, J. J. & Desmet, V. J. (1990) Hepatology 12, 148-154. 31. Malizia, G., Dino, O., Pisa, R., Caltagirone, M., Giannuoli, G., Marco, V. D., Aragona, E., Calabrese, A., Raiata, F., Craxi, A. & Pagliaro, L. (1991) Gastroenterology 100, 749-755. 32. Sninsky, J., Siddiqui, A., Robinson, W. S. & Cohen, S. N. (1979) Nature (London) 312, 639-641. 33. Hu, K. Q. & Siddiqui, A. (1991) Virology 181, 721-726. 34. Knowles, B. B., Howe, C. C. & Aden, D. P. (1980) Science 209, 497-499. 35. Nakabayashi, H., Taketa, K., Miyano, K., Yamane, T. & Sato, J. (1982) Cancer Res. 42, 3858-3863. 36. Staunton, D. E., Marlin, S. D., Stratowa, C., Dustin, M. L. & Springer, T. A. (1988) Cell 52, 925-933. 37. Chirgwin, J. M., Przybyla, A., MacDonald, R. & Rutter, W. J. (1978) Biochemistry 18, 5294-5299. 38. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) Molecular Cloning:A Laboratory Manual (Cold Spring Harbor Lab., Cold Spring Harbor, NY), 2nd Ed. 39. Clayton, D., Harrelson, A. & Darnell, J. E. (1985) Mol. Cell. Biol. 5, 2623-2632. 40. Celano, P., Berchtold, C. & Casero, R. A. (1989) BioTechniques 7, 942-943. 41. Caselmann, W. H., Meyer, M., Kekule, A., Lauer, U., Hofschneider, P. H. & Koshy, R. (1990) Proc. Nat!. Acad. Sci. USA 87, 2970-2974. 42. Salkind, A., Nicjols, J. E. & Roberts, N. J. (1991) J. Clin. Invest. 88,

505-511.

43. Shirakata, Y., Kawada, M., Fujiki, Y., Sano, H., Oda, M., Yaginuma, K., Kobayashi, M. & Koike, K. (1989) Jpn. J. Cancer Res. 80, 617-621. 44. Kim, C. M., Koike, K., Saito, I., Miyamura, T. & Jay, G. (1991) Nature (London) 351, 317-320. 45. Twu, J. S., Chu, K. & Robinson, W. (1989) Proc. Natl. Acad. Sci. USA

86, 5168-5172.

46. Seto, E., Mitchell, P. & Yen, T. S. B. (1990) Nature (London) 344, 72-74. 47. Maguire, H. F., Hoeffler, J. P. & Siddiqui, A. (1991) Science 252, 842-844. 48. Seto, E. & Yen, T. S. B. (1991) J. Virol. 65, 2351-2356. 49. Wu, J. Y., Zhou, Z. Y., Judd, A., Cartwright, C. A. & Robinson, W. S. (1990) Cell 63, 687-695. 50. Takada, S. & Koike, K. (1990) Jpn. J. Cancer Res. 81, 1191-1194. 51. Bandyopadhyay, S. K., Kalvakolanu, D. V. R. & Sen, G. C. (1990) Mol. Cell. Biol. 10, 5055-5063. 52. Kalvakolanu, D. V. R., Bandyopadhyay, S. K., Harter, M. L. & Sen, G. C. (1991) Proc. Natl. Acad. Sci. USA 88, 7459-7463. 53. Clark, E. A. & Lane, P. J. L. (1991) Annu. Rev. Immunol. 9, 97-127.