- Email: [email protected]
Hepatic expression patterns of the large and middle hepatitis B virus surface proteins in viremic and nonviremic chronic hepatitis B
GASTROENTEROLOGY
1990;98:1017-1023
Hepatic Expression Patterns of the Large and Middle Hepatitis B Virus Surface Proteins in Viremic and Nonviremic Chronic Hepatitis B HANS P. DIENES, WOLFRAM H. GERLICH, MARITA WORSDORFER, GUIDO GERKEN, LEONARDO BIANCHI, GEORG HESS, and KARL-HERMANN MEYER ZUM BtiSCHENFELDE Department of Pathology and First Department of Internal Medicine, University of Mainz, Federal Republic of Germany: Department of Pathology, University of Easel. Switzerland: Department of Medical Microbiology, University of Giittingen, Federal Republic of Germany
The envelope of hepatitis B virus consists of large, middle, and small hepatitis B surface proteins. Recent data from in vitro studies suggest that intracellular expression and distribution of the three polypep tides may be variable. These observations in artificial expression systems prompted this analysis of the occurrence and distribution of the three hepatitis B surface proteins in the liver tissue of substantial viremic (hepatitis B virus DNA- and hepatitis B e antigen-positive) and low-viremic or nonviremic (hepatitis B virus DNA-negative, anti-hepatitis B e antigen-positive) carriers by specific monoclonal antibodies against large, middle, and small proteins. Patients with an active form of viral replication showed a prevalence of middle and small hepatitis B surface proteins in the liver. In nonviremic carriers, the large hepatitis B surface protein was the predominant intrahepatic antigen, a finding that was confirmed at the ultrastructural level by staining of the entire filaments of the viral envelope material in ground glass hepatocytes. The present data are thus consistent with observations in hepatitis B virustransgenic mice and in transfected cell systems, suggesting that the different patterns of the envelope proteins in the liver may be due to different processing at the translational level.
*The Nomenclature and approved by the Harbour Conference 1985 and was referred
LHBs, MHBs. and SHBs has been discussed participants of a meeting at the Cold Spring on Molecular Biology of Hepatitis B Virus in by P. Tiollais in Nature 1985;317:489-495.
T
he hepatitis B virus (HBV] codes for at least three coterminal envelope proteins that have different aminotermini. The large hepatitis B surface protein (LHBs*; P39 or GP 42 in its glycosylated form) contains the pre-S1 domain, a nonglycosylated pre-S2 domain, and a gene S domain (1). The middle hepatitis B surface protein (MHBs*; GP33 or GP36) has a glycosylated pre-S2 domain and the gene S domain, whereas the most abundant small hepatitis B surface protein (SHBs*; P24 or GP27) consists of the gene S domain only. Infected hepatocytes secrete large amounts of surplus hepatitis B surface proteins (HBs) to the blood of virus carriers, where they are found as 20-nm hepatitis B surface antigen (HBsAg) particles. Expression of MHBs and SHBs is controlled by the same transcriptional promoter (21, whereas expression of LHBs may be regulated independently by a second, usually less active, promoter (3). The intrahepatic occurrence and distribution of the three envelope proteins has not yet been completely studied. In hepatitis B, LHBs is mainly found in viremic sera (44. This would suggest that viremic carriers should have more of LHBs in the liver than Abbreviations used in this paper: ABC, avidin-biotin complex; DAB, diaminobenzindine; GGH, ground-glass hepatocytes; HBcAg, hepatitis B core antigen; HBeAg, hepatitis B e antigen; HBs, hepatitis B surface protein; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HRPO, horseradish peroxidase; LHBs, large hepatitis B surface protein; MHBs, middle hepatitis B surface protein; SHBs, small hepatitis B surface protein. 0 1990 by the American Gastroenterological Association 0016-5085/90/$3.00
1018 DIENES ET AL.
GASTROENTEROLOGY Vol. 98, No. 4
nonviremic carriers. However, recent findings in transfected cell lines suggest a more complicated pattern. Particles are formed by LHBs in the lumen of the endoplasmic reticulum that are not secreted (6-11). It requires an excess of the small proteins to be secreted as a component of HBs particles (6-11). It is known that HBsAg is present in the so-called ground-glass hepatocytes (GGH) (12,13). First studies on the presence of the three HBs proteins in liver biopsy samples are inconsistent. They were based on immunofluorescence technique (14-177, which does not allow an exact localization of antigens in the liver cells regarding GGH. To achieve deeper insight into the hepatocellular distribution on the three HBs proteins at the light-microscopic and ultrastructural level, liver biopsy samples of viremic and nonviremic HBV carriers were studied by a sensitive avidin-biotin technique using monoclonal antibodies. Tissue sections were also examined by immunoelectron microscopy.
Materials and Methods Patients Forty patients with chronic hepatitis B were included in this study. All patients were positive for HBsAg and anti-HB,Ag (antibody to hepatitis B core antigen) in the serum. Patients were divided in one group with viremia (20 cases, positive for hepatitis B e antigen [HB,Ag] and HBV DNA in the serum) and in another group without massive viremia (sera being negative for HBeAg and HBV DNA). From all patients, a percutaneous liver biopsy sample was obtained after written consent. Eight otherwise healthy patients undergoing cholecystectomy whose biopsy samples were taken during surgery served as negative controls. All these patients were negative for HBV markers in the serum. Histopathological diagnosis of hepatitis was performed according to the recommendations of an international group (18). Serological
and Biochemical
Techniques
lmmunoelectron
Microscopy
Eight biopsy samples (4 from each group) were according to a protocol published previously (23). However, the fixative periodate-lysine-paraformaldehyde was replaced by 2% paraformaldehyde alone when it was learned that the staining for LHBs was reduced considerably by the periodate. Briefly, after incubation with the same antibodies as for light microscopy, an indirect technique with peroxidasecoupled secondary antisera was used. The sections were embedded in EPON on the slides, removed, and cut on an LKB ultramicrotome. The ultrathin sections were stained with uranyl acetate and viewed with a Philips EM 401. prepared
Tests
The sera were tested for activities of aminotransferases, alkaline phosphatase. and y-glutamyltranspeptidase. Bilirubin level was also determined. The HBV-associated antigens and antibodies HBsAg, anti-HBs, anti-HBc, HBeAg, and anti-HBeAg were assayed by radioimmunoassays (all purchased from Abbott Laboratories, North Chicago, Ill.). The presence of HBV DNA was detected by blotting technique according to Scotto (19). Tests for pre-S polypeptides were performed according to recently published methods (5,201.
Immunohistological Preparation
patts, F.R.G.; dilution 1:50); LHBs antigen, monoclonal mouse anti-pre-Sl, sequence 38-47 (MA 18/7; biotinylated, dilution 1:400 on cryostat and 1:lOO on paraffin sections) (1); and MHBs antigen, monoclonal mouse antibody against pre-S2 glycopolypeptide (QlWll; biotinylated, dilution 1:400 on cryostat and 1:lOO on paraffin sections) (21). All antisera were incubated on the tissue sections for 4 h. Secondary antiserum for SHBs was goat anti-mouse immunoglobulin G biotinylated at a dilution of 1:~ (Dakopatts), and for the HBcAg swine anti-rabbit biotinylated at a dilution of 1:50 (Dakopatts) was used. After application of the biotinylated antisera, the avidin-biotin complex (ABC) technique (22) was performed with streptavidin-horseradish peroxidase (HRPO) (Dakopatts) and DAB as disclosure solution (Sigma). Tissue preparation. From all patients, formaldehydefixed specimens were available, and routine paraffin sections were cut and dewaxed before incubation with the given antisera was performed. Additionally, frozen biopsy specimens were available from 13 of the 40 patients. Serial sections were used for the incubation with SHBsAg, LHBs, and MHBs antibodies to obtain an exact comparison of the staining patterns. After final incubations, the sections were counterstained with Mayer’s hemalum. The slides were interpreted by two of the authors (L.B. and H.P.D.) independently, each ignoring the viral replication status.
and Tissue
Antibodies. The antibodies used were: HBsAg (SHBs), monoclonal mouse anti-HBsAg (Dynatech, F.R.G.; dilution l:lOO), HBcAg, polyclonal rabbit anti-HBc (Dako-
Results The patients were grouped into 20 highly viremic and 20 nonviremic or lowviremic carriers on the basis of HBeAg and HBV DNA in the serum. One of the patients of the latter group, however, was HBVDNA positive without HBeAg. Serological findings of surface antigens in the two groups are given in Table 1. Histopathology
showed chronic active hepatitis in 10 (out of 20) viremic and in 6 [out of 20) nonviremic
carriers. The other viremic patients had chronic persistent hepatitis. In the nonviremic carrier group, 9 displayed chronic persistent hepatitis, and 5 of the patients had only minimal changes of liver tissue. Results of 15 biopsies of the viremic group were positive for HBcAg. In all cases it was present in the
EXPRESSION OF PRE-S POLYPEPTIDES OF HEPATITIS B VIRUS IN THE LIVER
April 1990
Table 1.
1019
Serological Titers of Surface Proteins LHBs
Nonviremic group (n = 201 Viremic group (n = 20)
MHBs
SHBs
Negative
1:lO
1:lOO
Negative
1:lO
1:lOO
Negative
1:lO
1:lOO
14 3
4 1
2 16
18 6
1 1
1 12
0 0
6 2
14 18
nuclei of hepatocytes, and in 5 biopsies an additional staining in the cytoplasm of few hepatocytes was noted. In 4 biopsies from the nonviremic group, only nuclear HBcAg was detected, with a faint but distinct staining in a few hepatocytes. The staining patterns of the HBV surface polypeptides were identical with those of the cryostat and paraffin sections when the same biopsy results were compared (13 cases tested]. However, there was a difference in titers of antibodies that were applied: the “working” dilutions of the antisera were higher on cryostat sections (anti-SHBs, 1:300; anti-LHBs and anti-MHBs, 1:500) than on paraffin sections (antiSHBs, 1:lOO;anti-LHBs and anti-MHBs, 1:200). There was no significant difference in the staining behavior of SHBs when viremic and nonviremic cases were compared. Results were positive in 17 cases of the viremic group and in 15 cases of the nonviremic group. Positive staining was more frequently found in groups of cells than in single cells. Some hepatocytes showed a membranous pattern of SHBs which, however, could not be correlated with the presence of HBcAg and the degree of inflammatory activity. Ground-glass hepatocytes were not always positive for SHBs staining [Figure 1). The predominant antigen in the liver biopsy results was LHBS (pre-Sl] for the nonviremic patients. Eightypercent were positive for this antigen, whereas only 62% of the patients showed positive staining for
MHBs in this group (Table 2). The quantitative evaluation of hepatocytes containing LHBs and MHBs by counting positive cells in serial sections revealed considerable differences: in the nonviremic group, LHBs-positive hepatocytes were approximately 5 times more frequent than MHBs-positive cells, and in 4 biopsies only LHBs could be detected (Figure 2). In the viremic patients, the differences were not as conspicuous. Some of the LHBs-negative specimens showed only very few (approximately 10-20) MHBspositive hepatocytes. Ground-glass hepatocytes were always positive for LHBs, but not for SHBs or MHBs. Non-GGH positive for the HBs proteins showed a variable staining pattern: membranous decoration was observed as well as patchy cytoplasmic positive reactions. In HBeAg-positive carriers, MHBs antigen was always detectable in the hepatocytes, ranging from few isolated cells to large confluent clusters of liver cells. Ground-glass hepatocyte-containing MHBs were also found in areas of inflammatory activity (Figure 3). The monoclonal antibody Q19/10 against MHBs seemed to be the most reactive reagent for intrahepatic HBs proteins. The antibody against SHBs did not react in 22% of the MHBs positive samples. The presence or absence of the HBs proteins did not correlate with the disease activity. Half of the patients (10 of 20) had a chronic active hepatitis. There was no staining for HBV-associated antigens (HBc, HBs) by our technique in the negative control group. By immunoelectron microscopy, the staining of LHBs was almost ubiquitous [Figure 4) in biopsy samples from patients with inactive viral replication. It correlated well with the appearance of filaments within the endoplasmic reticulum. Most of the filaments also reacted with anti-SHBs, but some areas Table 2. Presence
of Hepatitis in the Liver
B Virus-Associated
Viremic patients
Figure 1. Ground-glass hepatocytes in the liver of a nonviremic chronic carrier with chronic persistent hepatitis: some GGHs (arrows] are spared from staining with an antiserum directed at SHBs (original magnification x 475).
Antigens
Nonviremic patients
Antigen
(n = 201
(n = ZO]
LHBs MHBs (glycosylated pre-S2) SHBs HBc
12 (60%) 20 (95%) 17 (85%) 15 (75%)
16 (82%) 12 (62%) 15 (75%) 4 (20%)
“Statistically significant (p < 0.01).
1020
GASTROENTEROLOGY
DIENES ET AL.
Vol. 98. No. 4
Figure 2. Nonviremic carrier with minimal hepatitis. A conspicuous membranous pattern is visible for staining with LHBs (A), whereas the same area (serial sections) is negative for MHBs (B) (original magnification x250). The predominant staining for LHBs (C) is obvious around a portal tract compared with MHBs (D) (original magnification x 250)
filled with filaments did not react. In many MHBspositive cells anti-MHBs stained the filaments only segmentally (Figure 5). Sometimes the plasma membrane of the hepatocytes was positive as well. Moreover, LHBs antigen was found in the membranes of the endoplasmic reticulum and in the plasma membrane. In biopsy samples from viremic patients, MHBs
was the predominant antigen also at the ultrastructural level; it could be shown in the cytoplasm, and, in some hepatocytes, also in the plasma membrane (Figure 6). Fibrillar envelope material was seen rather infrequently, and the staining for MHBs was localized on segments of the membranes of the endoplasmic reticulum. However, HBcAg was never found in the plasma membrane. It was detected in some cells as cytoplasmic foci, but usually it was present only in the nucleus.
Discussion
Figure 3. Viremic carrier with CAH: MHBs-positive ground glass cells (k) in the area of necroinflammatory activity (original magnification x425).
Guided by recent findings in artificial production systems of HBV suggesting a different processing of the three HBs proteins of HBV, the expression patterns of the three HBs antigens were examined in liver biopsy samples of HBV carriers with and without serological markers of viral replication. We could show that their staining patterns were not always identical in the same biopsy, thus confirming previous findings from other laboratories (171, with pre-S antigen being the most prominent protein in nonviremic carriers (15). So far, all studies on the expression of the different HBV surface antigens in the tissue were
April
1990
EXPRESSION
OF PRE-S
Figure 4. Immunoelectron microscopy of a GGH in the liver biopsy sample of a nonviremic carrier. Heavy staining for LHBs in the endoplasmic reticulum covering up all fllamentous structures is regarded as ultrastructural equivalents of surface components (original magnification ~36,500)
performed by immunofluorescence technique, which has the disadvantage that exact localization and association with histopathology of the liver tissue are not possible. Using a sensitive immunohistologic technique (22) on formalin-fixed, paraffin-embedded liver biopsy samples, it could be shown that the presence of LHBs protein, especially in GGHs, indicate a storage of HBs proteins when viral replication is absent. Large hepatitis B surface protein is a necessary component of filaments (1,241 and probably also of virions, but secretion of these particles requires an excess of small proteins in vitro or in transgenic mice (10,25). In agreement with these previous findings, most liver samples from viremic carriers contained
Figure 5. Same biopsy sample as in Figure 4. Only a few segments of the abundant tllamentous material (arrows) are positive for MHBs (original magnification x 36,500).
POLYPEPTIDES
OF HEPATITIS
B VIRUS
IN THE LIVER
1021
some LHBs antigen, but its prevalence and its staining intensity was usually weaker than that of smaller proteins. The relative intensities of the three proteins in viremic carriers are consistent with the finding that the viral HBs messenger RNAs will encode predominantly the two smaller proteins, and that the pre-Sl promoter may be weak. The situation is strikingly different in nonviremic carriers. The usually lower serum titer of HBsAg in these carriers is in agreement with the lower prevalence of the intrahepatic MHBs-antigen. The high prevalence of intrahepatic LHBs found by the authors and other investigators (14,15) contrasts with the low titer or even absence of LHBs antigen in the serum (26) (Table 1). Western blots of the liver biopsy material confirmed the prevalence of LHBs antigen in the tissue of nonviremic carriers (27). The current ultrastrucural observations suggest that LHBs still forms filaments within the endoplasmic reticulum, although their secretion is apparently not possible. Using a combination of classical histology, immunohistology, and immunoelectron microscopy, it could be shown that the GGH (121, which are a histopathologic hallmark of chronic HBV carriers, are filled by LHBs. This conclusion has been previously drawn for transgenic mice that overexpressed LHBs under the control of the albumin promoter (25). These findings are also consistent with the observation in transfected cell systems that LBHs cannot be secreted when the smaller HBs are absent (6-11). Furthermore, the present data obtained from examinations in liver tissues exactly reflect the situation in the sera of viremic and nonviremic carriers of HBsAg as found in a recent study (28). It is reasonable to assume that expression of the HBs originate predominantly from episomal DNA in the viremic carriers, whereas in nonviremic carriers, integrated DNA may often be the only form. It is not clear why integrated HBV DNA should have an altered expression pattern, but our data suggest that LHBs expression is relatively increased in nonviremic carriers, whereas MHBs and SHBs expression is unaltered or even decreased. One possible model would be that the integrated HBV DNA may be located next to a highly active cellular promoter that would increase transcription and translation of the most proximal LBs gene. The activity of such a promoter might possibly be increased by the combined action of the viral enhancer (29) and the X-gene product of HBV (30). If it is assumed that such an integration event would occur frequently, clonal expansion of the original hepatocyte would be necessary to allow its detection in a random liver sample. Alternatively, the state of the cell may change expression. Marquardt et al. (10) observed that transfer of an expression active viral integrate from the hepatoma cell line PLC/PRF/5 to the HeLa cells
1022 DIENES ET AL.
GASTROENTEROLOGY
Vol. 98. No. 4
Figure 6. Liver biopsy sample of a viremic patient with chronic acute hepatitis. Staining for MHBs is positive in some areas of the ER of a hepatocyte (arrowhead), in the plasma membrane (arrow], and above that, around (*) a lymphocyte (L) abutting on the MHBs-positive hepatocyte (original magnification x 18,500).
increased expression of LHBs (10). Growth of the transfected HeLa cell as nude mouse tumor completely shut off further expression of the smaller proteins and lead to selective expression of LHBs, as currently observed in nonviremic carriers. The current findings in the liver tissue are in accordance with these data obtained from in vitro studies in that the production of LHBs is regulated independently from that of the smaller HBs proteins (3), different processing of the proteins may occur at the translational level (31). The membranous localization of LHBs as currently observed may be explained by the aminoterminal myristilation of LHBs (32). It is not known if the apparent overexpression of LHBs in nonviremic carriers has any significance for the development or maintenance of chronic liver disease. A direct oncogenic effect of LHBs is, however, unlikely, at least in humans, because most hepatocellular carcinomas are reported to express no viral envelope proteins at all. References 1. Heermann KH, Goldmann U, Schwartz W, Seyffarth T, Baumgarten H, Gerlich WH. Large surface proteins of hepatitis B virus containing the pre-S sequence. J Viral 1984:52:396-402. 2. Cattaneo R, Will J, Hernandez N, Schaller H. Signals regulating hepatitis B surface antigen transcription. Nature 1983;305:336338. 3. Ou JH, Rutter WJ. Hybrid hepatitis B virus-host transcripts in a human hepatoma cell. Proc Nat1 Acad Sci USA 1985;82:83-87.
4. Gerken G, Manns M, Gerlich WH, Hess G, Meyer zum Biischenfelde KH. Pre-S encoded surface proteins in relation to the major viral surface antigen in acute hepatitis B virus infection. Gastroenterology 1987;92:1864-1868. 5. Meyer zum Biischenfelde KH. Gerken G, Hess G, Manns M. The significance of the pre-S region of the hepatitis B virus. J Hepatol 1986;3:273-279. 6. Cheng KC, Smith GL, Moss B. Hepatitis-B-virus large surface protein is not secreted but is immunogenic when selectively expressed by recombinant Vaccinia virus. J Virol 1986;60:337344. 7. Chisari FV, Milich DR, McLachlan A. Expression of hepatitis B virus large envelope polypeptide inhibits hepatitis B surface antigen secretion in transgenic mice. J Virol1986;60:880-887. 8. Persing DH, Varmus HE, Ganem D. Inhibition of secretion of hepatitis B surface antigen by a related presurface polypeptide. Science 1986;234:1388-1391. 9. Standring DN, Ou JH, Rutter WJ. Assembly of viral particles in Xenopus oocytes: pre-S-surface antigens regulate secretion of the hepatitis B viral surface envelope particle. Proc Nat1 Acad Sci USA 1986;83:9338-9342. 10. Marquardt 0, Heermann KH, Seifer M, Gerlich WH. Cell type specific expression of pre-Sl antigen and secretion of hepatitis B virus surface antigen. Arch Virol1987;96:249-256. 11. Ou JH, Rutter WJ. Regulation of secretion of the hepatitis B virus major surface antigen by the pre-Sl protein. J Virol1987;61:782786. 12. Hadziyannis S, Gerber MA, Vissoulis C, Popper H. Cytoplasmic hepatitis B antigen in “ground-glass” hepatocytes of carriers. Arch Path01 1973;96:327-330. 13. Bianchi L, Gudat F. Immunopathology of hepatitis B. In: Popper H, Schaffner F, eds. Progress in liver diseases. Volume 6. New York: Grune & Stratton, 1979: pp. 371-392. 14. Thung SN. Gerber MA, Kasambalides EJ, Gilja BK, Keh W,
April 1990
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
EXPRESSION
OF PRE-S POLYPEPTIDES OF HEPATITIS B VIRUS IN THE LIVER
Gerlich WH. Demonstration of pre-S polypeptides of hepatitis B virus in infected livers. Hepatology 1986$:1315-1318, Hadziyannis S, Raimondo G, Papaioannou C. Anastassakos C. Wong D, Sninsky J, Shafritz D. Expression of pre-S geneencoded proteins in liver and serum during chronic hepatitis B virus infection in comparison to other markers of active virus replication. ] Hepatol 1987;5:253-259. Hess G, Gerlich WH, Gerken G. Manns M, Meyer zum Bttschenfelde KH. Relationship of pre-S encoded antigens in liver and clinical manifestations of chronic hepatitis B infection. Liver 1987;7:245-250. Fraiese A, Pontisso P, Cavaletto D, Fattovich G. Alberti A. Expression of preS1 in the liver of chronic hepatitis B virus carriers. ) Hepatol1988;7:157-163. International Group: Bianchi L, De Groote J, Desmet V, Gedigk P, Korb G, Popper H, Poulsen H, Scheuer PJ, Schmid M, Thaler H, Wepler W. Acute and chronic hepatitis revisited. Lancet 1977;2:914-919. Scotto J. Hadchouel M, Hery C, Yvart J, Tiollais P, Brechot C. Detection of hepatitis B virus DNA in serum by a simple spot hybridization technique: comparison with results for other viral markers. Hepatology 1983;3:279-285. Hess G, Gerlich GH. Gerken G, et al. Treatment of HBsaAg positive chronic active hepatitis with recombinant leucocytes (Y A interferon: The effect on synthesis of pre-S derived proteins. Hepatology 1987;7:704-708. Heermann KH. Waldeck F. Gerlich WH. Interaction between native human serum and the pre-S2 domain of the hepatitis B virus surface antigen. In: Zuckerman AJ, ed. Viral hepatitis and liver diseases. New York: Liss, 1987. Hsu SM. Raine L, Fanger H. A comparative study of the peroxidase-anti peroxidase method and an avidin-biotin complex method for studying polypeptide hormones with radioimmunoassay antibodies. Am J Clin Path01 1981;75:734-738. Dienes HP, Essling K, John HD. Application of immunoelectron microscopy for identification of T lymphocyte subsets and HLA DR antigen positive cells in tissue. J Histochem Cytochem 1984;32:547-551. Marquardt 0, Heermann KH. Seifer M, Gerlich WH. Cell-type
25.
26.
27.
28.
29. 30. 31.
32.
1023
dependent expression and secretion of hepatitis B virus pre-S1 surface antigen. Postgrad Med J 1987;63:41-50. Chisari FV, Filippi P, Buras Jon, McLachlan A, Popper H, Pinkert CA, Palmiter RD. Brinster RL. Structural and pathological effects of synthesis of hepatitis B virus large envelope polypeptide in transgenic mice. Proc Nat1 Acad Sci USA 1987;84:6909-6913. Theilmann L, Klinker M-Q, Gmelin K, Salfeld J, Schaller H. Pfaff E. Detection of pre-Sl proteins in serum and liver of HBsAg-positive patients: a new marker for hepatitis B virus infection. Hepatology 1986;6:186-190. Gerken G, Manns M, Gerlich WH, Hess G. Meyer zum Bttschenfelde KH. Immunoblot analysis of viral surface proteins in serum and liver of patients with chronic hepatitis B virus infection. J Med Viroll990 (in press]. Budkowska A, Dubreuil P, Gerlich WH, Lazizi Y. Pillot J. Occurrence of preS antigen in viremic and nonviremic carriers of hepatitis B surface antigen. J Med Virol1988;21:217-225. Shaul Y, Rutter WJ. Laub 0. A human hepatitis B viral enhancer element. EMBO J 1985;4:427-430. Spandau FD, Lee C-K. Transactivation of viral enhancers by the hepatitis B virus X protein. J Virol1988;62:427-434. Price PM, Mohamad A, Zelent A, Neurath AR, Acs G. Translational selection in the expression of the hepatitis B virus envelope proteins. DNA 1988;7:417-422. Persing DA, Varmus HA, Ganem D. The pre-Sl protein of hepatitis B virus is acylated at its amino terminus with myristic acid. J Virol1987;61:1672-1677.
Received September 8.1988. Accepted September 15,1989. Address requests for reprints to: Hans P. Dienes. M.D., Department of Pathology, Medical School, University of Mainz, Langenbeck&r. 1, D-6500 Mainz, Federal Republic of Germany. This work was supported by the Deutsche Forschungsgemeinschaft SFB 311/A5 and Ge 345/7-3. The authors gratefully acknowledge the secretarial assistance of Christina Biirkner.
1990;98:1017-1023
Hepatic Expression Patterns of the Large and Middle Hepatitis B Virus Surface Proteins in Viremic and Nonviremic Chronic Hepatitis B HANS P. DIENES, WOLFRAM H. GERLICH, MARITA WORSDORFER, GUIDO GERKEN, LEONARDO BIANCHI, GEORG HESS, and KARL-HERMANN MEYER ZUM BtiSCHENFELDE Department of Pathology and First Department of Internal Medicine, University of Mainz, Federal Republic of Germany: Department of Pathology, University of Easel. Switzerland: Department of Medical Microbiology, University of Giittingen, Federal Republic of Germany
The envelope of hepatitis B virus consists of large, middle, and small hepatitis B surface proteins. Recent data from in vitro studies suggest that intracellular expression and distribution of the three polypep tides may be variable. These observations in artificial expression systems prompted this analysis of the occurrence and distribution of the three hepatitis B surface proteins in the liver tissue of substantial viremic (hepatitis B virus DNA- and hepatitis B e antigen-positive) and low-viremic or nonviremic (hepatitis B virus DNA-negative, anti-hepatitis B e antigen-positive) carriers by specific monoclonal antibodies against large, middle, and small proteins. Patients with an active form of viral replication showed a prevalence of middle and small hepatitis B surface proteins in the liver. In nonviremic carriers, the large hepatitis B surface protein was the predominant intrahepatic antigen, a finding that was confirmed at the ultrastructural level by staining of the entire filaments of the viral envelope material in ground glass hepatocytes. The present data are thus consistent with observations in hepatitis B virustransgenic mice and in transfected cell systems, suggesting that the different patterns of the envelope proteins in the liver may be due to different processing at the translational level.
*The Nomenclature and approved by the Harbour Conference 1985 and was referred
LHBs, MHBs. and SHBs has been discussed participants of a meeting at the Cold Spring on Molecular Biology of Hepatitis B Virus in by P. Tiollais in Nature 1985;317:489-495.
T
he hepatitis B virus (HBV] codes for at least three coterminal envelope proteins that have different aminotermini. The large hepatitis B surface protein (LHBs*; P39 or GP 42 in its glycosylated form) contains the pre-S1 domain, a nonglycosylated pre-S2 domain, and a gene S domain (1). The middle hepatitis B surface protein (MHBs*; GP33 or GP36) has a glycosylated pre-S2 domain and the gene S domain, whereas the most abundant small hepatitis B surface protein (SHBs*; P24 or GP27) consists of the gene S domain only. Infected hepatocytes secrete large amounts of surplus hepatitis B surface proteins (HBs) to the blood of virus carriers, where they are found as 20-nm hepatitis B surface antigen (HBsAg) particles. Expression of MHBs and SHBs is controlled by the same transcriptional promoter (21, whereas expression of LHBs may be regulated independently by a second, usually less active, promoter (3). The intrahepatic occurrence and distribution of the three envelope proteins has not yet been completely studied. In hepatitis B, LHBs is mainly found in viremic sera (44. This would suggest that viremic carriers should have more of LHBs in the liver than Abbreviations used in this paper: ABC, avidin-biotin complex; DAB, diaminobenzindine; GGH, ground-glass hepatocytes; HBcAg, hepatitis B core antigen; HBeAg, hepatitis B e antigen; HBs, hepatitis B surface protein; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HRPO, horseradish peroxidase; LHBs, large hepatitis B surface protein; MHBs, middle hepatitis B surface protein; SHBs, small hepatitis B surface protein. 0 1990 by the American Gastroenterological Association 0016-5085/90/$3.00
1018 DIENES ET AL.
GASTROENTEROLOGY Vol. 98, No. 4
nonviremic carriers. However, recent findings in transfected cell lines suggest a more complicated pattern. Particles are formed by LHBs in the lumen of the endoplasmic reticulum that are not secreted (6-11). It requires an excess of the small proteins to be secreted as a component of HBs particles (6-11). It is known that HBsAg is present in the so-called ground-glass hepatocytes (GGH) (12,13). First studies on the presence of the three HBs proteins in liver biopsy samples are inconsistent. They were based on immunofluorescence technique (14-177, which does not allow an exact localization of antigens in the liver cells regarding GGH. To achieve deeper insight into the hepatocellular distribution on the three HBs proteins at the light-microscopic and ultrastructural level, liver biopsy samples of viremic and nonviremic HBV carriers were studied by a sensitive avidin-biotin technique using monoclonal antibodies. Tissue sections were also examined by immunoelectron microscopy.
Materials and Methods Patients Forty patients with chronic hepatitis B were included in this study. All patients were positive for HBsAg and anti-HB,Ag (antibody to hepatitis B core antigen) in the serum. Patients were divided in one group with viremia (20 cases, positive for hepatitis B e antigen [HB,Ag] and HBV DNA in the serum) and in another group without massive viremia (sera being negative for HBeAg and HBV DNA). From all patients, a percutaneous liver biopsy sample was obtained after written consent. Eight otherwise healthy patients undergoing cholecystectomy whose biopsy samples were taken during surgery served as negative controls. All these patients were negative for HBV markers in the serum. Histopathological diagnosis of hepatitis was performed according to the recommendations of an international group (18). Serological
and Biochemical
Techniques
lmmunoelectron
Microscopy
Eight biopsy samples (4 from each group) were according to a protocol published previously (23). However, the fixative periodate-lysine-paraformaldehyde was replaced by 2% paraformaldehyde alone when it was learned that the staining for LHBs was reduced considerably by the periodate. Briefly, after incubation with the same antibodies as for light microscopy, an indirect technique with peroxidasecoupled secondary antisera was used. The sections were embedded in EPON on the slides, removed, and cut on an LKB ultramicrotome. The ultrathin sections were stained with uranyl acetate and viewed with a Philips EM 401. prepared
Tests
The sera were tested for activities of aminotransferases, alkaline phosphatase. and y-glutamyltranspeptidase. Bilirubin level was also determined. The HBV-associated antigens and antibodies HBsAg, anti-HBs, anti-HBc, HBeAg, and anti-HBeAg were assayed by radioimmunoassays (all purchased from Abbott Laboratories, North Chicago, Ill.). The presence of HBV DNA was detected by blotting technique according to Scotto (19). Tests for pre-S polypeptides were performed according to recently published methods (5,201.
Immunohistological Preparation
patts, F.R.G.; dilution 1:50); LHBs antigen, monoclonal mouse anti-pre-Sl, sequence 38-47 (MA 18/7; biotinylated, dilution 1:400 on cryostat and 1:lOO on paraffin sections) (1); and MHBs antigen, monoclonal mouse antibody against pre-S2 glycopolypeptide (QlWll; biotinylated, dilution 1:400 on cryostat and 1:lOO on paraffin sections) (21). All antisera were incubated on the tissue sections for 4 h. Secondary antiserum for SHBs was goat anti-mouse immunoglobulin G biotinylated at a dilution of 1:~ (Dakopatts), and for the HBcAg swine anti-rabbit biotinylated at a dilution of 1:50 (Dakopatts) was used. After application of the biotinylated antisera, the avidin-biotin complex (ABC) technique (22) was performed with streptavidin-horseradish peroxidase (HRPO) (Dakopatts) and DAB as disclosure solution (Sigma). Tissue preparation. From all patients, formaldehydefixed specimens were available, and routine paraffin sections were cut and dewaxed before incubation with the given antisera was performed. Additionally, frozen biopsy specimens were available from 13 of the 40 patients. Serial sections were used for the incubation with SHBsAg, LHBs, and MHBs antibodies to obtain an exact comparison of the staining patterns. After final incubations, the sections were counterstained with Mayer’s hemalum. The slides were interpreted by two of the authors (L.B. and H.P.D.) independently, each ignoring the viral replication status.
and Tissue
Antibodies. The antibodies used were: HBsAg (SHBs), monoclonal mouse anti-HBsAg (Dynatech, F.R.G.; dilution l:lOO), HBcAg, polyclonal rabbit anti-HBc (Dako-
Results The patients were grouped into 20 highly viremic and 20 nonviremic or lowviremic carriers on the basis of HBeAg and HBV DNA in the serum. One of the patients of the latter group, however, was HBVDNA positive without HBeAg. Serological findings of surface antigens in the two groups are given in Table 1. Histopathology
showed chronic active hepatitis in 10 (out of 20) viremic and in 6 [out of 20) nonviremic
carriers. The other viremic patients had chronic persistent hepatitis. In the nonviremic carrier group, 9 displayed chronic persistent hepatitis, and 5 of the patients had only minimal changes of liver tissue. Results of 15 biopsies of the viremic group were positive for HBcAg. In all cases it was present in the
EXPRESSION OF PRE-S POLYPEPTIDES OF HEPATITIS B VIRUS IN THE LIVER
April 1990
Table 1.
1019
Serological Titers of Surface Proteins LHBs
Nonviremic group (n = 201 Viremic group (n = 20)
MHBs
SHBs
Negative
1:lO
1:lOO
Negative
1:lO
1:lOO
Negative
1:lO
1:lOO
14 3
4 1
2 16
18 6
1 1
1 12
0 0
6 2
14 18
nuclei of hepatocytes, and in 5 biopsies an additional staining in the cytoplasm of few hepatocytes was noted. In 4 biopsies from the nonviremic group, only nuclear HBcAg was detected, with a faint but distinct staining in a few hepatocytes. The staining patterns of the HBV surface polypeptides were identical with those of the cryostat and paraffin sections when the same biopsy results were compared (13 cases tested]. However, there was a difference in titers of antibodies that were applied: the “working” dilutions of the antisera were higher on cryostat sections (anti-SHBs, 1:300; anti-LHBs and anti-MHBs, 1:500) than on paraffin sections (antiSHBs, 1:lOO;anti-LHBs and anti-MHBs, 1:200). There was no significant difference in the staining behavior of SHBs when viremic and nonviremic cases were compared. Results were positive in 17 cases of the viremic group and in 15 cases of the nonviremic group. Positive staining was more frequently found in groups of cells than in single cells. Some hepatocytes showed a membranous pattern of SHBs which, however, could not be correlated with the presence of HBcAg and the degree of inflammatory activity. Ground-glass hepatocytes were not always positive for SHBs staining [Figure 1). The predominant antigen in the liver biopsy results was LHBS (pre-Sl] for the nonviremic patients. Eightypercent were positive for this antigen, whereas only 62% of the patients showed positive staining for
MHBs in this group (Table 2). The quantitative evaluation of hepatocytes containing LHBs and MHBs by counting positive cells in serial sections revealed considerable differences: in the nonviremic group, LHBs-positive hepatocytes were approximately 5 times more frequent than MHBs-positive cells, and in 4 biopsies only LHBs could be detected (Figure 2). In the viremic patients, the differences were not as conspicuous. Some of the LHBs-negative specimens showed only very few (approximately 10-20) MHBspositive hepatocytes. Ground-glass hepatocytes were always positive for LHBs, but not for SHBs or MHBs. Non-GGH positive for the HBs proteins showed a variable staining pattern: membranous decoration was observed as well as patchy cytoplasmic positive reactions. In HBeAg-positive carriers, MHBs antigen was always detectable in the hepatocytes, ranging from few isolated cells to large confluent clusters of liver cells. Ground-glass hepatocyte-containing MHBs were also found in areas of inflammatory activity (Figure 3). The monoclonal antibody Q19/10 against MHBs seemed to be the most reactive reagent for intrahepatic HBs proteins. The antibody against SHBs did not react in 22% of the MHBs positive samples. The presence or absence of the HBs proteins did not correlate with the disease activity. Half of the patients (10 of 20) had a chronic active hepatitis. There was no staining for HBV-associated antigens (HBc, HBs) by our technique in the negative control group. By immunoelectron microscopy, the staining of LHBs was almost ubiquitous [Figure 4) in biopsy samples from patients with inactive viral replication. It correlated well with the appearance of filaments within the endoplasmic reticulum. Most of the filaments also reacted with anti-SHBs, but some areas Table 2. Presence
of Hepatitis in the Liver
B Virus-Associated
Viremic patients
Figure 1. Ground-glass hepatocytes in the liver of a nonviremic chronic carrier with chronic persistent hepatitis: some GGHs (arrows] are spared from staining with an antiserum directed at SHBs (original magnification x 475).
Antigens
Nonviremic patients
Antigen
(n = 201
(n = ZO]
LHBs MHBs (glycosylated pre-S2) SHBs HBc
12 (60%) 20 (95%) 17 (85%) 15 (75%)
16 (82%) 12 (62%) 15 (75%) 4 (20%)
“Statistically significant (p < 0.01).
1020
GASTROENTEROLOGY
DIENES ET AL.
Vol. 98. No. 4
Figure 2. Nonviremic carrier with minimal hepatitis. A conspicuous membranous pattern is visible for staining with LHBs (A), whereas the same area (serial sections) is negative for MHBs (B) (original magnification x250). The predominant staining for LHBs (C) is obvious around a portal tract compared with MHBs (D) (original magnification x 250)
filled with filaments did not react. In many MHBspositive cells anti-MHBs stained the filaments only segmentally (Figure 5). Sometimes the plasma membrane of the hepatocytes was positive as well. Moreover, LHBs antigen was found in the membranes of the endoplasmic reticulum and in the plasma membrane. In biopsy samples from viremic patients, MHBs
was the predominant antigen also at the ultrastructural level; it could be shown in the cytoplasm, and, in some hepatocytes, also in the plasma membrane (Figure 6). Fibrillar envelope material was seen rather infrequently, and the staining for MHBs was localized on segments of the membranes of the endoplasmic reticulum. However, HBcAg was never found in the plasma membrane. It was detected in some cells as cytoplasmic foci, but usually it was present only in the nucleus.
Discussion
Figure 3. Viremic carrier with CAH: MHBs-positive ground glass cells (k) in the area of necroinflammatory activity (original magnification x425).
Guided by recent findings in artificial production systems of HBV suggesting a different processing of the three HBs proteins of HBV, the expression patterns of the three HBs antigens were examined in liver biopsy samples of HBV carriers with and without serological markers of viral replication. We could show that their staining patterns were not always identical in the same biopsy, thus confirming previous findings from other laboratories (171, with pre-S antigen being the most prominent protein in nonviremic carriers (15). So far, all studies on the expression of the different HBV surface antigens in the tissue were
April
1990
EXPRESSION
OF PRE-S
Figure 4. Immunoelectron microscopy of a GGH in the liver biopsy sample of a nonviremic carrier. Heavy staining for LHBs in the endoplasmic reticulum covering up all fllamentous structures is regarded as ultrastructural equivalents of surface components (original magnification ~36,500)
performed by immunofluorescence technique, which has the disadvantage that exact localization and association with histopathology of the liver tissue are not possible. Using a sensitive immunohistologic technique (22) on formalin-fixed, paraffin-embedded liver biopsy samples, it could be shown that the presence of LHBs protein, especially in GGHs, indicate a storage of HBs proteins when viral replication is absent. Large hepatitis B surface protein is a necessary component of filaments (1,241 and probably also of virions, but secretion of these particles requires an excess of small proteins in vitro or in transgenic mice (10,25). In agreement with these previous findings, most liver samples from viremic carriers contained
Figure 5. Same biopsy sample as in Figure 4. Only a few segments of the abundant tllamentous material (arrows) are positive for MHBs (original magnification x 36,500).
POLYPEPTIDES
OF HEPATITIS
B VIRUS
IN THE LIVER
1021
some LHBs antigen, but its prevalence and its staining intensity was usually weaker than that of smaller proteins. The relative intensities of the three proteins in viremic carriers are consistent with the finding that the viral HBs messenger RNAs will encode predominantly the two smaller proteins, and that the pre-Sl promoter may be weak. The situation is strikingly different in nonviremic carriers. The usually lower serum titer of HBsAg in these carriers is in agreement with the lower prevalence of the intrahepatic MHBs-antigen. The high prevalence of intrahepatic LHBs found by the authors and other investigators (14,15) contrasts with the low titer or even absence of LHBs antigen in the serum (26) (Table 1). Western blots of the liver biopsy material confirmed the prevalence of LHBs antigen in the tissue of nonviremic carriers (27). The current ultrastrucural observations suggest that LHBs still forms filaments within the endoplasmic reticulum, although their secretion is apparently not possible. Using a combination of classical histology, immunohistology, and immunoelectron microscopy, it could be shown that the GGH (121, which are a histopathologic hallmark of chronic HBV carriers, are filled by LHBs. This conclusion has been previously drawn for transgenic mice that overexpressed LHBs under the control of the albumin promoter (25). These findings are also consistent with the observation in transfected cell systems that LBHs cannot be secreted when the smaller HBs are absent (6-11). Furthermore, the present data obtained from examinations in liver tissues exactly reflect the situation in the sera of viremic and nonviremic carriers of HBsAg as found in a recent study (28). It is reasonable to assume that expression of the HBs originate predominantly from episomal DNA in the viremic carriers, whereas in nonviremic carriers, integrated DNA may often be the only form. It is not clear why integrated HBV DNA should have an altered expression pattern, but our data suggest that LHBs expression is relatively increased in nonviremic carriers, whereas MHBs and SHBs expression is unaltered or even decreased. One possible model would be that the integrated HBV DNA may be located next to a highly active cellular promoter that would increase transcription and translation of the most proximal LBs gene. The activity of such a promoter might possibly be increased by the combined action of the viral enhancer (29) and the X-gene product of HBV (30). If it is assumed that such an integration event would occur frequently, clonal expansion of the original hepatocyte would be necessary to allow its detection in a random liver sample. Alternatively, the state of the cell may change expression. Marquardt et al. (10) observed that transfer of an expression active viral integrate from the hepatoma cell line PLC/PRF/5 to the HeLa cells
1022 DIENES ET AL.
GASTROENTEROLOGY
Vol. 98. No. 4
Figure 6. Liver biopsy sample of a viremic patient with chronic acute hepatitis. Staining for MHBs is positive in some areas of the ER of a hepatocyte (arrowhead), in the plasma membrane (arrow], and above that, around (*) a lymphocyte (L) abutting on the MHBs-positive hepatocyte (original magnification x 18,500).
increased expression of LHBs (10). Growth of the transfected HeLa cell as nude mouse tumor completely shut off further expression of the smaller proteins and lead to selective expression of LHBs, as currently observed in nonviremic carriers. The current findings in the liver tissue are in accordance with these data obtained from in vitro studies in that the production of LHBs is regulated independently from that of the smaller HBs proteins (3), different processing of the proteins may occur at the translational level (31). The membranous localization of LHBs as currently observed may be explained by the aminoterminal myristilation of LHBs (32). It is not known if the apparent overexpression of LHBs in nonviremic carriers has any significance for the development or maintenance of chronic liver disease. A direct oncogenic effect of LHBs is, however, unlikely, at least in humans, because most hepatocellular carcinomas are reported to express no viral envelope proteins at all. References 1. Heermann KH, Goldmann U, Schwartz W, Seyffarth T, Baumgarten H, Gerlich WH. Large surface proteins of hepatitis B virus containing the pre-S sequence. J Viral 1984:52:396-402. 2. Cattaneo R, Will J, Hernandez N, Schaller H. Signals regulating hepatitis B surface antigen transcription. Nature 1983;305:336338. 3. Ou JH, Rutter WJ. Hybrid hepatitis B virus-host transcripts in a human hepatoma cell. Proc Nat1 Acad Sci USA 1985;82:83-87.
4. Gerken G, Manns M, Gerlich WH, Hess G, Meyer zum Biischenfelde KH. Pre-S encoded surface proteins in relation to the major viral surface antigen in acute hepatitis B virus infection. Gastroenterology 1987;92:1864-1868. 5. Meyer zum Biischenfelde KH. Gerken G, Hess G, Manns M. The significance of the pre-S region of the hepatitis B virus. J Hepatol 1986;3:273-279. 6. Cheng KC, Smith GL, Moss B. Hepatitis-B-virus large surface protein is not secreted but is immunogenic when selectively expressed by recombinant Vaccinia virus. J Virol 1986;60:337344. 7. Chisari FV, Milich DR, McLachlan A. Expression of hepatitis B virus large envelope polypeptide inhibits hepatitis B surface antigen secretion in transgenic mice. J Virol1986;60:880-887. 8. Persing DH, Varmus HE, Ganem D. Inhibition of secretion of hepatitis B surface antigen by a related presurface polypeptide. Science 1986;234:1388-1391. 9. Standring DN, Ou JH, Rutter WJ. Assembly of viral particles in Xenopus oocytes: pre-S-surface antigens regulate secretion of the hepatitis B viral surface envelope particle. Proc Nat1 Acad Sci USA 1986;83:9338-9342. 10. Marquardt 0, Heermann KH, Seifer M, Gerlich WH. Cell type specific expression of pre-Sl antigen and secretion of hepatitis B virus surface antigen. Arch Virol1987;96:249-256. 11. Ou JH, Rutter WJ. Regulation of secretion of the hepatitis B virus major surface antigen by the pre-Sl protein. J Virol1987;61:782786. 12. Hadziyannis S, Gerber MA, Vissoulis C, Popper H. Cytoplasmic hepatitis B antigen in “ground-glass” hepatocytes of carriers. Arch Path01 1973;96:327-330. 13. Bianchi L, Gudat F. Immunopathology of hepatitis B. In: Popper H, Schaffner F, eds. Progress in liver diseases. Volume 6. New York: Grune & Stratton, 1979: pp. 371-392. 14. Thung SN. Gerber MA, Kasambalides EJ, Gilja BK, Keh W,
April 1990
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
EXPRESSION
OF PRE-S POLYPEPTIDES OF HEPATITIS B VIRUS IN THE LIVER
Gerlich WH. Demonstration of pre-S polypeptides of hepatitis B virus in infected livers. Hepatology 1986$:1315-1318, Hadziyannis S, Raimondo G, Papaioannou C. Anastassakos C. Wong D, Sninsky J, Shafritz D. Expression of pre-S geneencoded proteins in liver and serum during chronic hepatitis B virus infection in comparison to other markers of active virus replication. ] Hepatol 1987;5:253-259. Hess G, Gerlich WH, Gerken G. Manns M, Meyer zum Bttschenfelde KH. Relationship of pre-S encoded antigens in liver and clinical manifestations of chronic hepatitis B infection. Liver 1987;7:245-250. Fraiese A, Pontisso P, Cavaletto D, Fattovich G. Alberti A. Expression of preS1 in the liver of chronic hepatitis B virus carriers. ) Hepatol1988;7:157-163. International Group: Bianchi L, De Groote J, Desmet V, Gedigk P, Korb G, Popper H, Poulsen H, Scheuer PJ, Schmid M, Thaler H, Wepler W. Acute and chronic hepatitis revisited. Lancet 1977;2:914-919. Scotto J. Hadchouel M, Hery C, Yvart J, Tiollais P, Brechot C. Detection of hepatitis B virus DNA in serum by a simple spot hybridization technique: comparison with results for other viral markers. Hepatology 1983;3:279-285. Hess G, Gerlich GH. Gerken G, et al. Treatment of HBsaAg positive chronic active hepatitis with recombinant leucocytes (Y A interferon: The effect on synthesis of pre-S derived proteins. Hepatology 1987;7:704-708. Heermann KH. Waldeck F. Gerlich WH. Interaction between native human serum and the pre-S2 domain of the hepatitis B virus surface antigen. In: Zuckerman AJ, ed. Viral hepatitis and liver diseases. New York: Liss, 1987. Hsu SM. Raine L, Fanger H. A comparative study of the peroxidase-anti peroxidase method and an avidin-biotin complex method for studying polypeptide hormones with radioimmunoassay antibodies. Am J Clin Path01 1981;75:734-738. Dienes HP, Essling K, John HD. Application of immunoelectron microscopy for identification of T lymphocyte subsets and HLA DR antigen positive cells in tissue. J Histochem Cytochem 1984;32:547-551. Marquardt 0, Heermann KH. Seifer M, Gerlich WH. Cell-type
25.
26.
27.
28.
29. 30. 31.
32.
1023
dependent expression and secretion of hepatitis B virus pre-S1 surface antigen. Postgrad Med J 1987;63:41-50. Chisari FV, Filippi P, Buras Jon, McLachlan A, Popper H, Pinkert CA, Palmiter RD. Brinster RL. Structural and pathological effects of synthesis of hepatitis B virus large envelope polypeptide in transgenic mice. Proc Nat1 Acad Sci USA 1987;84:6909-6913. Theilmann L, Klinker M-Q, Gmelin K, Salfeld J, Schaller H. Pfaff E. Detection of pre-Sl proteins in serum and liver of HBsAg-positive patients: a new marker for hepatitis B virus infection. Hepatology 1986;6:186-190. Gerken G, Manns M, Gerlich WH, Hess G. Meyer zum Bttschenfelde KH. Immunoblot analysis of viral surface proteins in serum and liver of patients with chronic hepatitis B virus infection. J Med Viroll990 (in press]. Budkowska A, Dubreuil P, Gerlich WH, Lazizi Y. Pillot J. Occurrence of preS antigen in viremic and nonviremic carriers of hepatitis B surface antigen. J Med Virol1988;21:217-225. Shaul Y, Rutter WJ. Laub 0. A human hepatitis B viral enhancer element. EMBO J 1985;4:427-430. Spandau FD, Lee C-K. Transactivation of viral enhancers by the hepatitis B virus X protein. J Virol1988;62:427-434. Price PM, Mohamad A, Zelent A, Neurath AR, Acs G. Translational selection in the expression of the hepatitis B virus envelope proteins. DNA 1988;7:417-422. Persing DA, Varmus HA, Ganem D. The pre-Sl protein of hepatitis B virus is acylated at its amino terminus with myristic acid. J Virol1987;61:1672-1677.
Received September 8.1988. Accepted September 15,1989. Address requests for reprints to: Hans P. Dienes. M.D., Department of Pathology, Medical School, University of Mainz, Langenbeck&r. 1, D-6500 Mainz, Federal Republic of Germany. This work was supported by the Deutsche Forschungsgemeinschaft SFB 311/A5 and Ge 345/7-3. The authors gratefully acknowledge the secretarial assistance of Christina Biirkner.