Journal of Medical Virology 3287-91 (1990)
Hepatitis B Pre-S1 and Pre-S2 Proteins: Clinical Significance and Relation to Hepatitis B Virus DNA T.J.M. van Ditzhuijsen, L.P.C. Kuijpers, M.J. Koens, P.J.M. Rijntjes, A.M. van Loon, and S.H. Yap Division of Gastrointestinal and Liver Diseases, Departments of Medicine (T.J.M.v.D.1 and of Medical Microbiology (A.M.v.L.1, St. Radboud University Hospital, Nijmegen and Diagnostics Research Laboratories, Organon, Oss (L.P.C.K., M.J.K., P.J.M.R.), The Netherlands; Division of Liver and Pancreatic Diseases, University Hospital Gasthuisberg, Leuven, Belgium (S.H.Y.) Pre-S proteins may have an important role in virus assembly and virus entry into the host cell. The presence of pre-S proteins in serum has also been thought to correlate with active viral replication. To investigate whether pre-S proteins in serum might have additional diagnostic and/or predictive value for liver sequelae in HBV infection, sera from six different serological groups of patients with HBV markers (total number 363) and different manifestations of liver histology were examined for the presence of pre-S1 and pre-S2 proteins using micro-ELISAs. Pre-S1 and pre-S2 proteins were detected significantly more often in HBV-DNA-positive than in HBV-DNAnegative sera from HBsAg carriers. However, pre-S1 and pre-S2 proteins were also found in HBV-DNA-negative HBsAg carriers irrespective of serum HBeAg/anti-HBe or liver histologic findings. These results suggest that the presence of the pre-S1 and or pre-S2 proteins in serum either does not seem to reflect the presence of active viral replication and active liver disease or pre-S proteins are more readily detectable than HBeAg and HB-DNA as measured by a dot-blot technique. Furthermore, the presence of pre-S proteins in serum is strongly correlated with that of HBsAg.
KEY WORDS: hepatitis B virus, DNA, pre-S proteins
INTRODUCTION The open reading frame of the HBsAg gene of hepatitis B virus (HBV) contains three ATG (adenine, thymidine, guanosine) initiator codons for protein synthesis. The most downstream situated initiation site generates the major HBsAg (226 amino acids). The second ATG site, 165 bases upstream, initiates the production of the major protein extended with a 55 amino acid sequence known as pre-S2, thus yielding the HBsAg middle protein. The third ATG codon on the S gene, 357 bases further upstream, codes for the HBsAg 0 1990 WILEY-LISS, INC.
large protein with another 119-128 amino acid extension known as the pre-S1 antigen. The large protein therefore contains the S-,pre-S2-, and pre-S1-antigenic determinant [Robinson et al., 1987; Takahashi et al., 19861. The pre-S peptides have the ability to bind to plasma membranes of human hepatocytes, directly or indirectly, and are presumed to have a function in virus entry into the host cell [Neurath et al., 1986, 19871. Both pre-S1 and pre-S2 antigens may also have an important role in virus assembly [Ou and Rutter, 1987; Wong et al., 19851. In chronic HBsAg carriers, pre-S1 can be found in virus particles [Heermann et al., 1984; Stibbe and Gerlich, 19841, whereas pre-S2 containing proteins are also present in the small spherical particles [Heermann et al., 19841. The presence of pre-S1 in serum has also been found to correlate well with active virus replication [Gerken et al., 1987; Theilmann et al., 1986a,b; Kuijpers et al., 19891,while the pre-S2 antigen has an immunodominant epitope and may have additional value in vaccines [Neurath et al., 19871. The aim of the present study was to investigate whether the detection of pre-S1 and or pre-S2 encoded proteins in serum has diagnostic and/or predictive value in staging of HBV infection and whether the pre-S1 and or pre-S2 proteins are the expression of circulating virions as determined by serum HBV DNA and HBeAg.
PATIENTS AND METHODS Sera obtained from 363 subjects with various serologic (see Tables I, 11) and histologic (see Figs. 1, 2) manifestations of HBV infection, and from ten patients with chronic liver disease not related to HBV were examined under code for the presence of pre-S1 and pre-S2 proteins. The diagnosis of acute hepatitis (AH) and chronic HBsAg positive carrier (CAR) with unknown liver histology was based on clinical, biochemAccepted for publication June 29, 1990. Address reprint requests to Dr. S.H. Yap, Div. of Liver and Pancreatic Diseases, Dept. of Medicine, University Hospital Gasthuisberg, B-3000 Leuven, Belgium.
van Ditzhuijsen et al.
88
TABLE I. Pre-S1-, Pre-S2 Proteins a n d HBV-DVA in Six Groups of Sera Obtained From 373 Subjects With Different HBV Serologic Markers Serologic HBVpre-S1 PRE-s~ Group marker(+) n DNA(+) % (+I % (+I 96 I
I1 111 IV V
VI
HBsAg HBeAg anti-HBc HBsAg anti-HBe anti-HBc
HBsAg anti-HBc anti-HBc (anti-HBe pos. or neg.) anti-HBc anti-HBs No HBV markers
109
77
(70.6)
102
(93.6)
100
(91.7)
98
13
(13.3)
79
(80.6)
87
(88.8)
25
1
(4.0)
3
(12.0)
9
(36.0)
49
1
(2.0)
1
(2.0)
1
(2.0)
82
0
(0.0)
4
(4.9)
5
(6.1)
10
0
(0.0)
0
(0.0)
0
(0.0)
TABLE 11. Serum Pre-S Proteins in Relation to Serum HBV-DNA in HBsAg-Positive Patients
HBsAg+ sera-
/HBV-DNAf HBeAGS 'HBV-DNA/HBV-DNA' anti-HBe'
i
'HBV-DNA/HBV-DNAf
73
(94.8)
74
(96.1)
32 13
29 13
(90.6) (100.0)
27 13
(100.0)
85 1
66
(77.6)
76 1
(89.4)
24
2
(8.3)
7
(29.2)
1
Pre-S2
(84.4)
liver histology unknown
liver histology
100-
%
Pre-S1
HBeAganti-HBe--
\HBV-DNA-
%
n
77
total
CAHIC
CPH
PHC
AH
CAR
n=109
n=l4
n=17
n=3
n=35
n=4O
75 -
50 -
25 -
HBV DNA pos. patients pre-Sl pos. patients pre-S2 pos. patients
Fig. 1. Distribution ofye-Sl, pre-S2, and HBV-DNA in sera obtained from HBsA ositiveIHBeAg-positivepatients ( roup I) with various manifestations o liver histology. CAHIC, chronic active hepatitiskirrhasis; ETH, chronic persistent hepatitis; P$C, primary hepatocellular carcinoma; AH, acute hepatitis; CAR, chronic HBsAg-positive carrier.
Hepatitis B Pre-S1 and Pre-S2 Proteins
89
% liver histology unknown
liver histology
total
CAHIC
CPH
n=98
n=l4
n=9
PHC
n=5
AH
CAR
n=20
n=50
HBV DNA pos. patients pre-Sl pos. patients p r e - s z pos patients
Fig. 2. Distribution of [re-Sl, pre-S2, and HBV-DNA in sera obtained from HBsAg-positiveianti-HBe-positivepatients (group 11) with various manifestations o liver histology. CAHIC, chronic active hepatitis or cirrhosis; CPH, chronic persistent hepatitis; PHC, primary hepatocellular carcinoma; AH, acute hepatitis; CAR, chronic HBsAg-positive carrier.
Coupling of the peptides with the C-terminal ATAgroup to the activated carrier proteins succinimidyl-4-(N-maleimidomethyl)-cyclohexane1-carboxylate was accomplished by the formation of stable disulfide bonds after deacetylation of the ATA group with hydroxylamine. Specific PCAs against the S domain of HBsAg were produced in sheep by injection of purified serum-derived HBsAg, in which the pre-S determinant was destroyed by pepsin digestion. Anti-S specificity was confirmed by testing HBsAg before and after reductive treatment with 2-mercaptoethanol, which affects the major S determinant [Petit et al., 19861. Murine monoclonal antibodies against HBsAg (antiHBs MCAs) were selected on the basis of reactivity with a panel of 12 subtypes of HBsAg [Courouce-Pauty et al., 19831 in a micro-ELISA for HBsAg. Anti-HBs MCAs were considered S-specific when no discrepancy was found in their reactivity with pepsin-digested and undigested HBsAg. Anti-HBs MCAs which were nonreactive with pepsin-digested HBsAg were considered pre-S specific. Anti-pre-S2-specificMCAs were selected by their propensity to become blocked by the pre-S Detection of Pre-S1 and Pre-S2 Proteins [126-1451 peptide. Micro-Elisa tests for pre-S1 and pre-S2 were prePre-S-specific micro-ELISAs were composed of antipared essentially as described earlier [Kuijpers et al., pre-S solid phase and HRP-labeled anti-S. For specific19893. Two reactive sequences of the pre-S regions of ity reasons, the respective antibodies were either PCA hepatitis B surface antigen were synthesized chemi- and MCA (pre-S1test) or MCA and PCA (pre-S2 test) as cally. described previously [Kuijpers et al., 19891.
ical (transaminases), and serologic findings. Serological tests for conventional HBV markers were carried out by radioimmunoassay or enzyme immunoassay (Abbott Laboratories, North Chicago, IL) Detection of HBV-DNA in serum was performed by molecular hybridization according to the method described by Shimizu et al. [19841with slight modifications. Briefly, 25 pl of serum was denaturated in 25 p1 of 2 M NaCl and 100 ~1 of 1M NaOH for at least 30 minutes at room temperature. Prior to application of the samples, the nitrocellulose filter was rinsed in 6 X SSC (1 x SSC [standard saline citrate]: 0.15 M NaC1, 0.015 M Na citrate). After application of the mixture, 100 pl of 3 M NaCl, 0.5 M Tris-HC1 (pH 7.4) was passed through a filtration apparatus under slightly negative pressure. The filter was removed from the apparatus, rinsed in 6 x SSC, dried at room temperature, and baked at 80°C for 2 hours. On each filter a standard series of HBVDNA was present in the samples. With this technique less than 2.5 pg of HBV-DNA could be detected. Mantel-Haenszel’s chi-square test was used for the statistical analysis of the data.
90
Cut-Off Determination and Control Sera Results were considered positive if the absorbance with the test sample was at least three standard deviations above the mean absorbance of the negative controls. Normal human serum pools, free from all hepatitis B markers, served as negative controls. HBsAg produced by the PLC/PRF/5 human hepatoma cell line, known to produce pre-S2 sequences [Rutter et al., 19841, was used as a positive control in the pre-S2 test, whereas HBsAg/HBeAg-positive sera (reactive with anti-pre-S PCA and non-reactive when this PCA was blocked with the corresponding peptide) were used as positive controls in the pre-S1 test.
RESULTS Pre-S Proteins The results of determinations of pre-S1 and pre-S2 proteins and HBV-DNA in serum samples obtained from 373 subjects are listed in Table I. Pre-S1 and pre-S2 proteins could be detected in nearly all HBsAg-positive/HBeAg-positive sera (group I). As shown in Figure 1,pre-S antigens were present in sera of patients with acute hepatitis as well as in sera obtained from patients with long-standing HBs antigenemia, regardless of the various forms of histologic changes. HBV-DNA was found in 70.6% of these serum samples (80-90% of sera from patients with chronic hepatitis). In group I1 (HBsAg positive/anti-HBepositive patients) pre-S1 and pre-S2 antigens were also found frequently despite the absence of HBV-DNA in the majority of sera and regardless of the type of liver disease as determined by liver histology (Fig. 2). When group I and I1 were considered together, 87.4% of sera were positive for pre-S1, and 90.3% were positive for pre-S2 proteins. In contrast, in HBeAg-negative, anti-HBe-negative HBsAg carriers (Table I, group 111),pre-S1 and pre-S2 proteins were found in only a minority of patients. Pre-S1 was found less frequently than pre-S2 in this group. In sera obtained from patients with anti-HBc (HBsAg-negative with or without anti-HBe (Table I, group IV), both pre-S1 and pre-S2 were found in serum of only one case. This particular patient, whose serum was also positive for HBV-DNA, suffered from chronic persistent hepatitis, reflected by moderate elevation of serum aminotransferase values. In this regard it is interesting to note that in all (n = 8) sera obtained from patients during the window phase after acute hepatitis, neither HBV-DNA nor pre-S proteins were detectable (data not shown). In anti-HBc/anti-HBs-positive sera (Table I, group V), pre-S1 and pre-S2 proteins were found in two patients with chronic liver disease and in two other patients during the reconvalescent phase after acute hepatitis. As shown in Table 11,both pre-S1 and pre-S2 proteins were present in most HBeAg- or anti-HBe-positive specimens, whereas in sera obtained from HBeAg-
van Ditzhuijsen et al.
negative/anti-HBe-negative patients pre-S1 and pre-S2 were not frequently detectable. In addition, pre-S1 protein was found in 87/91 of HBsAg positive/HBVDNA-positive subjects (95.6%) vs. 971141 of HBsAgpositive/HBV-DNA-negative cases (68.8%)(P< .0001), and pre-S2 protein was detected in 88/91 (96.7%) of HBsAg positive/HBV-DNA-positive patients compared with 110/141 (78.6%) of HBsAg-positive/HBV-DNAnegative cases (P < .0001). Sera from patients without HBV serologic markers were all negative for both pre-S1 and pre-S2 proteins.
DISCUSSION Most of the studies on the presence of pre-S antigens and antibodies against these proteins focused on patients with acute type B hepatitis [Brahm et al., 1988; Gerken et al., 1987; Theilmann et al., 1986bl. The very early phase of acute hepatitis is serologically characterized by the presence of pre-S2 antigen and HBsAg [Hu et al., 1987; Robinson et al., 1987; Rutter et al., 19841, when HBV-DNA is still undetectable [Theilmann et al., 1986133. Pre-S2 antigen titers increase rapidly when HBV-DNA becomes detectable in serum and become negative shortly after disappearance of HBV-DNA from serum [Gerken et al., 19871. Persistence of pre-S2 antigen for a period longer than 4 months correlated with chronic hepatitis [Budkowska et al., 1988a,bl. Pre-S1 antigen can also be detected during the early course of acute hepatitis, 2 weeks after the appearance of HBsAg in serum [Gerken et al., 19871. This antigen is, however, no longer detectable when HBsAg production is maximal, i.e., usually soon after the disappearance of HBV-DNA and HBeAg from serum [Theilmann et al., 1986al. Pre-S1 antigen was also found to correlate with the presence of HBV-DNA and was no longer detectable on cessation of viral replication, after natural recovery and after successful treatment with alpha-interferon [Kuijpers et al., 19891. Pre-S2 proteins were also lost after treatment with alpha-interferon [Kuijpers et al., 19891. In this study the presence of pre-S1 and pre-S2 proteins in serum correlated well with the presence of HBsAg in serum (Tables I, II), irrespective of liver histologic or biochemical changes (data not shown) indicative of liver cell injury (Figs. 1,2).Pre-S1 and or pre-S2 proteins were present significantly more often in serum of patients with evidence of active viral replication (HBeAg and HBV-DNA-positive). Our results are however at variance with those of Hadziyannis et al. [19871, who reported a strong correlation between the presence of pre-S proteins in serum and active chronic liver disease. As in our study, pre-S proteins were detectable in 85% of HBV-DNA-negative chronic HBsAg carriers irrespective of the liver histology, the presence of this protein in serum either does not seem to reflect the presence of active viral replication and active liver disease or pre-S proteins are more readily detectable than HBV-DNA as determined by a dot blot technique
91
Hepatitis B Pre-S1 and Pre-S2 Proteins
7
Hu PS, Carithers RL, Fiorenza V, Peterson DL (1987): uantitative studies of the hepatitis B viral pre-S proteins: lack o correlation with the HBeAg status. Journal of Virological Methods 16:97-114. Kuijpers L, Koens M, Jacobs A, Boon P, Matthyssen L (1988):Pre-S1 and ~-S2-specific HBsAg and anti-HBs sandwich micro ELISAs, deve oped with synthetic peptides. In Zuckerman AJ. “Viral Hepatitis and Liver Disease.’ New York: Alan R. Liss, Inc. pp 280-283. Kui’ ers L, Koens M, Murra Lyon I, Coleman JC, Karayiannus P, &omas HC, Zanetti A, zirgnel A, Harrison TJ, Zuckerman AJ (1989):Pre-S proteins in hepatitis B. Journal of Medical Virology 28:47-51. Marion PL,, Salazar FH, Alexander JJ, Robinson WS (1980): State of he atitis B viral DNA in a human hepatocellular cell line. Journal of birology 33:795-806. Neurath AR, Kent SBH, Strick N, Parker K (1986):Identification and chemical synthesis of a host cell receptor binding site on hepatitis B virus. Cell 46:429-436. Neurath AR, Kent SBH, Strick N, Parker K (1987):The hepatitis B virus s nthesis a peptide that specifically binds to hepatocytes: antiboies to this peptide are important for host defence. Gastroenterology 92:262, 263. Neurath AR, Kent SBH, Strick N (1984): Location and chemical synthesis of a pre-S gene coded immunodominant epitope of hepatitis B virus. Science 224:392-394. Neurath AR, Kent SBH, Strick N, Taylor P, Stevens ,CE (1985): Hepatitis B virus contains pre S gene-encoded domains. Nature 315:154-156. Ou J-H. Rutter WJ (1987): Reeulation of secretion of the heuatitis B virus major surface antigen by the pre-S1 protein. Jdurnal of Virology 61:782-786. Petit MAT Maillard P, Cape1 F, Pillot J (1986): Immunochemical structure of the hepatitis B surface antigen vaccine-11. Analysis of antibody responses in human sera against the envelope proteins. Molecular Immunology 23511-523. Robinson WS, Miller RH, Masion PL (1987): Hepadnaviruses share genome homology and features of replication. Hepatology 7:64S73s. Rutter WJ, Ziemer M, Ou J, S h a d Y, Lamb 0, Garcia P, Standrin DN (1984): Transcription units of he atitis B virus genes an8 structure and expression of integratef viral sequences. In Vyas GN, Dienstag JL, Hoofna le J H (eds): “Viral Hepatitis and Liver Disease. Orlando: Grune Stratton, pp 67-86. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, REFERENCES Mullis KB, Erlich HA (1988):Primer directed enzymatic amplifiBrahm J , Vento S, Rondanelli EG, Ranieri S, Fagan EA, Williams R, cation of DNA with a thermostable DNA polymerase. Science Eddleston AL (1988): Sequential studies of pre-S2 antigenemia 239:487-491. and anti-pre S2 antibodies in relation to viral replication in acute Shimizu Y, Ida S, Matsukara T, Yuasa T (1984): Determination of hepatitis B followed from the early incubation phase. Journal of hepatitis B virus DNA in serum by molecular hybridization. Medical Virology 24:205-209. Microbiology and Immunology 28:1117-1123. Budkowska A, Dubreuil P, Gerlich WH, Lazizi Y, Pillot J (1988b): Stibbe W, Gerlich WH (1984): Variable protein composition of hepaOccurrence of pre-S1 antigen in viremic and nonviremic carriers of titis B virus containing the pre-S sequence. Journal of Virology hepatitis B surface antigen. Journal of Medical Virology 26:21752:396-401. 225. Takahashi K, Kishimoto S, Ohnuma H, Machida A, Takai E, Tsuda F, Budkowska A, Dubreuil P, Pillot J (1988a):Prognostic value of pre-S2 Miyamoto H, Tanaka T, Matsushita K, Oda K, Miyakawa Y, epitopes of hepatitis B virus and anti-pre S2 res onse evaluated by Mayumi M (1986):Polypeptides coded for by the region re S and monoclonal assays. In Zuckerman A J (ed): “&a1 Hepatitis and ene S of hepatitis B virus DNA with the receptor for pokmerized Liver Disease.” New York: Alan R. Liss, Inc., pp 287-289. guman serum albumin: expression on hepatitis B particles proCourouce-Pauty AM, Plancon A, Soulier J P (1983): Distribution of duced in the HBeA or anti-HBe phase of hepatitis B virus HBsAg subtypes in the world. Vox Sanguinis 44:197-211. infection. Journal of fmmunology 136:3467-3472. Gerken G, Manns M, Gerlich WH, Hess G, Meyer zum Buschenfelde Theilmann L, Klinkert MQ, Gmelin,K, Salfeld I, Schaller H, Pfaff E KH (1987):Pre-S encoded surface proteins in relation to the major (1986a):Detection of pre-S1 proteins in serum and liver of HBsAgviral surface antigen in acute hepatitis B virus infection. Gastroositive patients: A new marker for hepatitis B virus infection. enterology 92:1864-1868. kepatology 6:186-190. Hadziyannis S, Raimondo G, Papaioannou C, Anastassakos C, Won Theilmann L, Klinkert MQ, Gmelin K, Salfeld J , Schaller H, Pfaff E D, Sninsky J, Shafritz D (1987):Expression of pre-S gene-encode8 (1986b):Detection of pre-S proteins in serum and liver of HBsAgproteins in liver and serum during chronic hepatitis B virus ositive patients: a new marker for hepatitis B virus infection. infection in comparison to other markers of active virus replicaRepatology 6:186-190. tion. Journal of Hepatology 5253-259. Wong DT, Nath N, Sninsky JJ (1985): Identification of hepatitis B virus poly e tides encoded by the entire pre-S open reading frame. Heermann K-H, Goldmann U,Schwartz W, Seyffarth T, Baumgarten Journal o!&ology 55:223-231. H, Gerlich WH (1984):Large surface proteins of hepatitis B virus containing the pre-S sequence. Journal of Virology 52:39&402.
in such a replicative phase. A recently developed sensitive method for detection of HBV-DNA by gene amplification [Saiki et al., 19881 may clarify this issue. Despite this controversy, pre-S1 and pre-S2 were found in the present study on 79.3% and 84.5%,respectively, of HBsAg-positive specimens vs. 3.8% and 4.6% of HBsAg-negative sera, suggesting that the presence of pre-S proteins in serum correlates strongly with that of HBsAg. Our findings are, therefore in agreement with those reported by Budkowska et al. [1988b1, who showed that 65%of sera from anti-HBe-positive chronic HBsAg carriers reacted positively to the pre-S determinant, and with those of Hu et al. [19871who demonstrated that the presence of pre-S proteins in serum is only a function of the plasma HBsAg titer. Finding of pre-S in the absence of viral replication is also supported by an in vitro study which demonstrated that pre-S proteins are present in the supernatant of a PLC/PRF/5 cell line producing HBsAg but no viral particles [Hu et al., 1987; Marion et al., 19801. It is concluded on the basis of the present study, that despite the strong correlation between the presence of pre-S proteins and markers indicating active viral replication, tests for pre-S proteins cannot simply be regarded equivalent to HBeAg and conventional HBVDNA tests for the detection of viral replication and infectivity in HBsAg-positive patients, since pre-S1 and/or pre-S2 proteins persist in serum of the vast majority of patients after disappearance of HBeAg and HBV-DNA from serum. Furthermore, the presence of pre-S does not reflect active chronic liver disease. However, the presence of pre-S1 and or pre-S2 proteins in serum is strongly correlated with that of HBsAg.
Hepatitis B Pre-S1 and Pre-S2 Proteins: Clinical Significance and Relation to Hepatitis B Virus DNA T.J.M. van Ditzhuijsen, L.P.C. Kuijpers, M.J. Koens, P.J.M. Rijntjes, A.M. van Loon, and S.H. Yap Division of Gastrointestinal and Liver Diseases, Departments of Medicine (T.J.M.v.D.1 and of Medical Microbiology (A.M.v.L.1, St. Radboud University Hospital, Nijmegen and Diagnostics Research Laboratories, Organon, Oss (L.P.C.K., M.J.K., P.J.M.R.), The Netherlands; Division of Liver and Pancreatic Diseases, University Hospital Gasthuisberg, Leuven, Belgium (S.H.Y.) Pre-S proteins may have an important role in virus assembly and virus entry into the host cell. The presence of pre-S proteins in serum has also been thought to correlate with active viral replication. To investigate whether pre-S proteins in serum might have additional diagnostic and/or predictive value for liver sequelae in HBV infection, sera from six different serological groups of patients with HBV markers (total number 363) and different manifestations of liver histology were examined for the presence of pre-S1 and pre-S2 proteins using micro-ELISAs. Pre-S1 and pre-S2 proteins were detected significantly more often in HBV-DNA-positive than in HBV-DNAnegative sera from HBsAg carriers. However, pre-S1 and pre-S2 proteins were also found in HBV-DNA-negative HBsAg carriers irrespective of serum HBeAg/anti-HBe or liver histologic findings. These results suggest that the presence of the pre-S1 and or pre-S2 proteins in serum either does not seem to reflect the presence of active viral replication and active liver disease or pre-S proteins are more readily detectable than HBeAg and HB-DNA as measured by a dot-blot technique. Furthermore, the presence of pre-S proteins in serum is strongly correlated with that of HBsAg.
KEY WORDS: hepatitis B virus, DNA, pre-S proteins
INTRODUCTION The open reading frame of the HBsAg gene of hepatitis B virus (HBV) contains three ATG (adenine, thymidine, guanosine) initiator codons for protein synthesis. The most downstream situated initiation site generates the major HBsAg (226 amino acids). The second ATG site, 165 bases upstream, initiates the production of the major protein extended with a 55 amino acid sequence known as pre-S2, thus yielding the HBsAg middle protein. The third ATG codon on the S gene, 357 bases further upstream, codes for the HBsAg 0 1990 WILEY-LISS, INC.
large protein with another 119-128 amino acid extension known as the pre-S1 antigen. The large protein therefore contains the S-,pre-S2-, and pre-S1-antigenic determinant [Robinson et al., 1987; Takahashi et al., 19861. The pre-S peptides have the ability to bind to plasma membranes of human hepatocytes, directly or indirectly, and are presumed to have a function in virus entry into the host cell [Neurath et al., 1986, 19871. Both pre-S1 and pre-S2 antigens may also have an important role in virus assembly [Ou and Rutter, 1987; Wong et al., 19851. In chronic HBsAg carriers, pre-S1 can be found in virus particles [Heermann et al., 1984; Stibbe and Gerlich, 19841, whereas pre-S2 containing proteins are also present in the small spherical particles [Heermann et al., 19841. The presence of pre-S1 in serum has also been found to correlate well with active virus replication [Gerken et al., 1987; Theilmann et al., 1986a,b; Kuijpers et al., 19891,while the pre-S2 antigen has an immunodominant epitope and may have additional value in vaccines [Neurath et al., 19871. The aim of the present study was to investigate whether the detection of pre-S1 and or pre-S2 encoded proteins in serum has diagnostic and/or predictive value in staging of HBV infection and whether the pre-S1 and or pre-S2 proteins are the expression of circulating virions as determined by serum HBV DNA and HBeAg.
PATIENTS AND METHODS Sera obtained from 363 subjects with various serologic (see Tables I, 11) and histologic (see Figs. 1, 2) manifestations of HBV infection, and from ten patients with chronic liver disease not related to HBV were examined under code for the presence of pre-S1 and pre-S2 proteins. The diagnosis of acute hepatitis (AH) and chronic HBsAg positive carrier (CAR) with unknown liver histology was based on clinical, biochemAccepted for publication June 29, 1990. Address reprint requests to Dr. S.H. Yap, Div. of Liver and Pancreatic Diseases, Dept. of Medicine, University Hospital Gasthuisberg, B-3000 Leuven, Belgium.
van Ditzhuijsen et al.
88
TABLE I. Pre-S1-, Pre-S2 Proteins a n d HBV-DVA in Six Groups of Sera Obtained From 373 Subjects With Different HBV Serologic Markers Serologic HBVpre-S1 PRE-s~ Group marker(+) n DNA(+) % (+I % (+I 96 I
I1 111 IV V
VI
HBsAg HBeAg anti-HBc HBsAg anti-HBe anti-HBc
HBsAg anti-HBc anti-HBc (anti-HBe pos. or neg.) anti-HBc anti-HBs No HBV markers
109
77
(70.6)
102
(93.6)
100
(91.7)
98
13
(13.3)
79
(80.6)
87
(88.8)
25
1
(4.0)
3
(12.0)
9
(36.0)
49
1
(2.0)
1
(2.0)
1
(2.0)
82
0
(0.0)
4
(4.9)
5
(6.1)
10
0
(0.0)
0
(0.0)
0
(0.0)
TABLE 11. Serum Pre-S Proteins in Relation to Serum HBV-DNA in HBsAg-Positive Patients
HBsAg+ sera-
/HBV-DNAf HBeAGS 'HBV-DNA/HBV-DNA' anti-HBe'
i
'HBV-DNA/HBV-DNAf
73
(94.8)
74
(96.1)
32 13
29 13
(90.6) (100.0)
27 13
(100.0)
85 1
66
(77.6)
76 1
(89.4)
24
2
(8.3)
7
(29.2)
1
Pre-S2
(84.4)
liver histology unknown
liver histology
100-
%
Pre-S1
HBeAganti-HBe--
\HBV-DNA-
%
n
77
total
CAHIC
CPH
PHC
AH
CAR
n=109
n=l4
n=17
n=3
n=35
n=4O
75 -
50 -
25 -
HBV DNA pos. patients pre-Sl pos. patients pre-S2 pos. patients
Fig. 1. Distribution ofye-Sl, pre-S2, and HBV-DNA in sera obtained from HBsA ositiveIHBeAg-positivepatients ( roup I) with various manifestations o liver histology. CAHIC, chronic active hepatitiskirrhasis; ETH, chronic persistent hepatitis; P$C, primary hepatocellular carcinoma; AH, acute hepatitis; CAR, chronic HBsAg-positive carrier.
Hepatitis B Pre-S1 and Pre-S2 Proteins
89
% liver histology unknown
liver histology
total
CAHIC
CPH
n=98
n=l4
n=9
PHC
n=5
AH
CAR
n=20
n=50
HBV DNA pos. patients pre-Sl pos. patients p r e - s z pos patients
Fig. 2. Distribution of [re-Sl, pre-S2, and HBV-DNA in sera obtained from HBsAg-positiveianti-HBe-positivepatients (group 11) with various manifestations o liver histology. CAHIC, chronic active hepatitis or cirrhosis; CPH, chronic persistent hepatitis; PHC, primary hepatocellular carcinoma; AH, acute hepatitis; CAR, chronic HBsAg-positive carrier.
Coupling of the peptides with the C-terminal ATAgroup to the activated carrier proteins succinimidyl-4-(N-maleimidomethyl)-cyclohexane1-carboxylate was accomplished by the formation of stable disulfide bonds after deacetylation of the ATA group with hydroxylamine. Specific PCAs against the S domain of HBsAg were produced in sheep by injection of purified serum-derived HBsAg, in which the pre-S determinant was destroyed by pepsin digestion. Anti-S specificity was confirmed by testing HBsAg before and after reductive treatment with 2-mercaptoethanol, which affects the major S determinant [Petit et al., 19861. Murine monoclonal antibodies against HBsAg (antiHBs MCAs) were selected on the basis of reactivity with a panel of 12 subtypes of HBsAg [Courouce-Pauty et al., 19831 in a micro-ELISA for HBsAg. Anti-HBs MCAs were considered S-specific when no discrepancy was found in their reactivity with pepsin-digested and undigested HBsAg. Anti-HBs MCAs which were nonreactive with pepsin-digested HBsAg were considered pre-S specific. Anti-pre-S2-specificMCAs were selected by their propensity to become blocked by the pre-S Detection of Pre-S1 and Pre-S2 Proteins [126-1451 peptide. Micro-Elisa tests for pre-S1 and pre-S2 were prePre-S-specific micro-ELISAs were composed of antipared essentially as described earlier [Kuijpers et al., pre-S solid phase and HRP-labeled anti-S. For specific19893. Two reactive sequences of the pre-S regions of ity reasons, the respective antibodies were either PCA hepatitis B surface antigen were synthesized chemi- and MCA (pre-S1test) or MCA and PCA (pre-S2 test) as cally. described previously [Kuijpers et al., 19891.
ical (transaminases), and serologic findings. Serological tests for conventional HBV markers were carried out by radioimmunoassay or enzyme immunoassay (Abbott Laboratories, North Chicago, IL) Detection of HBV-DNA in serum was performed by molecular hybridization according to the method described by Shimizu et al. [19841with slight modifications. Briefly, 25 pl of serum was denaturated in 25 p1 of 2 M NaCl and 100 ~1 of 1M NaOH for at least 30 minutes at room temperature. Prior to application of the samples, the nitrocellulose filter was rinsed in 6 X SSC (1 x SSC [standard saline citrate]: 0.15 M NaC1, 0.015 M Na citrate). After application of the mixture, 100 pl of 3 M NaCl, 0.5 M Tris-HC1 (pH 7.4) was passed through a filtration apparatus under slightly negative pressure. The filter was removed from the apparatus, rinsed in 6 x SSC, dried at room temperature, and baked at 80°C for 2 hours. On each filter a standard series of HBVDNA was present in the samples. With this technique less than 2.5 pg of HBV-DNA could be detected. Mantel-Haenszel’s chi-square test was used for the statistical analysis of the data.
90
Cut-Off Determination and Control Sera Results were considered positive if the absorbance with the test sample was at least three standard deviations above the mean absorbance of the negative controls. Normal human serum pools, free from all hepatitis B markers, served as negative controls. HBsAg produced by the PLC/PRF/5 human hepatoma cell line, known to produce pre-S2 sequences [Rutter et al., 19841, was used as a positive control in the pre-S2 test, whereas HBsAg/HBeAg-positive sera (reactive with anti-pre-S PCA and non-reactive when this PCA was blocked with the corresponding peptide) were used as positive controls in the pre-S1 test.
RESULTS Pre-S Proteins The results of determinations of pre-S1 and pre-S2 proteins and HBV-DNA in serum samples obtained from 373 subjects are listed in Table I. Pre-S1 and pre-S2 proteins could be detected in nearly all HBsAg-positive/HBeAg-positive sera (group I). As shown in Figure 1,pre-S antigens were present in sera of patients with acute hepatitis as well as in sera obtained from patients with long-standing HBs antigenemia, regardless of the various forms of histologic changes. HBV-DNA was found in 70.6% of these serum samples (80-90% of sera from patients with chronic hepatitis). In group I1 (HBsAg positive/anti-HBepositive patients) pre-S1 and pre-S2 antigens were also found frequently despite the absence of HBV-DNA in the majority of sera and regardless of the type of liver disease as determined by liver histology (Fig. 2). When group I and I1 were considered together, 87.4% of sera were positive for pre-S1, and 90.3% were positive for pre-S2 proteins. In contrast, in HBeAg-negative, anti-HBe-negative HBsAg carriers (Table I, group 111),pre-S1 and pre-S2 proteins were found in only a minority of patients. Pre-S1 was found less frequently than pre-S2 in this group. In sera obtained from patients with anti-HBc (HBsAg-negative with or without anti-HBe (Table I, group IV), both pre-S1 and pre-S2 were found in serum of only one case. This particular patient, whose serum was also positive for HBV-DNA, suffered from chronic persistent hepatitis, reflected by moderate elevation of serum aminotransferase values. In this regard it is interesting to note that in all (n = 8) sera obtained from patients during the window phase after acute hepatitis, neither HBV-DNA nor pre-S proteins were detectable (data not shown). In anti-HBc/anti-HBs-positive sera (Table I, group V), pre-S1 and pre-S2 proteins were found in two patients with chronic liver disease and in two other patients during the reconvalescent phase after acute hepatitis. As shown in Table 11,both pre-S1 and pre-S2 proteins were present in most HBeAg- or anti-HBe-positive specimens, whereas in sera obtained from HBeAg-
van Ditzhuijsen et al.
negative/anti-HBe-negative patients pre-S1 and pre-S2 were not frequently detectable. In addition, pre-S1 protein was found in 87/91 of HBsAg positive/HBVDNA-positive subjects (95.6%) vs. 971141 of HBsAgpositive/HBV-DNA-negative cases (68.8%)(P< .0001), and pre-S2 protein was detected in 88/91 (96.7%) of HBsAg positive/HBV-DNA-positive patients compared with 110/141 (78.6%) of HBsAg-positive/HBV-DNAnegative cases (P < .0001). Sera from patients without HBV serologic markers were all negative for both pre-S1 and pre-S2 proteins.
DISCUSSION Most of the studies on the presence of pre-S antigens and antibodies against these proteins focused on patients with acute type B hepatitis [Brahm et al., 1988; Gerken et al., 1987; Theilmann et al., 1986bl. The very early phase of acute hepatitis is serologically characterized by the presence of pre-S2 antigen and HBsAg [Hu et al., 1987; Robinson et al., 1987; Rutter et al., 19841, when HBV-DNA is still undetectable [Theilmann et al., 1986133. Pre-S2 antigen titers increase rapidly when HBV-DNA becomes detectable in serum and become negative shortly after disappearance of HBV-DNA from serum [Gerken et al., 19871. Persistence of pre-S2 antigen for a period longer than 4 months correlated with chronic hepatitis [Budkowska et al., 1988a,bl. Pre-S1 antigen can also be detected during the early course of acute hepatitis, 2 weeks after the appearance of HBsAg in serum [Gerken et al., 19871. This antigen is, however, no longer detectable when HBsAg production is maximal, i.e., usually soon after the disappearance of HBV-DNA and HBeAg from serum [Theilmann et al., 1986al. Pre-S1 antigen was also found to correlate with the presence of HBV-DNA and was no longer detectable on cessation of viral replication, after natural recovery and after successful treatment with alpha-interferon [Kuijpers et al., 19891. Pre-S2 proteins were also lost after treatment with alpha-interferon [Kuijpers et al., 19891. In this study the presence of pre-S1 and pre-S2 proteins in serum correlated well with the presence of HBsAg in serum (Tables I, II), irrespective of liver histologic or biochemical changes (data not shown) indicative of liver cell injury (Figs. 1,2).Pre-S1 and or pre-S2 proteins were present significantly more often in serum of patients with evidence of active viral replication (HBeAg and HBV-DNA-positive). Our results are however at variance with those of Hadziyannis et al. [19871, who reported a strong correlation between the presence of pre-S proteins in serum and active chronic liver disease. As in our study, pre-S proteins were detectable in 85% of HBV-DNA-negative chronic HBsAg carriers irrespective of the liver histology, the presence of this protein in serum either does not seem to reflect the presence of active viral replication and active liver disease or pre-S proteins are more readily detectable than HBV-DNA as determined by a dot blot technique
91
Hepatitis B Pre-S1 and Pre-S2 Proteins
7
Hu PS, Carithers RL, Fiorenza V, Peterson DL (1987): uantitative studies of the hepatitis B viral pre-S proteins: lack o correlation with the HBeAg status. Journal of Virological Methods 16:97-114. Kuijpers L, Koens M, Jacobs A, Boon P, Matthyssen L (1988):Pre-S1 and ~-S2-specific HBsAg and anti-HBs sandwich micro ELISAs, deve oped with synthetic peptides. In Zuckerman AJ. “Viral Hepatitis and Liver Disease.’ New York: Alan R. Liss, Inc. pp 280-283. Kui’ ers L, Koens M, Murra Lyon I, Coleman JC, Karayiannus P, &omas HC, Zanetti A, zirgnel A, Harrison TJ, Zuckerman AJ (1989):Pre-S proteins in hepatitis B. Journal of Medical Virology 28:47-51. Marion PL,, Salazar FH, Alexander JJ, Robinson WS (1980): State of he atitis B viral DNA in a human hepatocellular cell line. Journal of birology 33:795-806. Neurath AR, Kent SBH, Strick N, Parker K (1986):Identification and chemical synthesis of a host cell receptor binding site on hepatitis B virus. Cell 46:429-436. Neurath AR, Kent SBH, Strick N, Parker K (1987):The hepatitis B virus s nthesis a peptide that specifically binds to hepatocytes: antiboies to this peptide are important for host defence. Gastroenterology 92:262, 263. Neurath AR, Kent SBH, Strick N (1984): Location and chemical synthesis of a pre-S gene coded immunodominant epitope of hepatitis B virus. Science 224:392-394. Neurath AR, Kent SBH, Strick N, Taylor P, Stevens ,CE (1985): Hepatitis B virus contains pre S gene-encoded domains. Nature 315:154-156. Ou J-H. Rutter WJ (1987): Reeulation of secretion of the heuatitis B virus major surface antigen by the pre-S1 protein. Jdurnal of Virology 61:782-786. Petit MAT Maillard P, Cape1 F, Pillot J (1986): Immunochemical structure of the hepatitis B surface antigen vaccine-11. Analysis of antibody responses in human sera against the envelope proteins. Molecular Immunology 23511-523. Robinson WS, Miller RH, Masion PL (1987): Hepadnaviruses share genome homology and features of replication. Hepatology 7:64S73s. Rutter WJ, Ziemer M, Ou J, S h a d Y, Lamb 0, Garcia P, Standrin DN (1984): Transcription units of he atitis B virus genes an8 structure and expression of integratef viral sequences. In Vyas GN, Dienstag JL, Hoofna le J H (eds): “Viral Hepatitis and Liver Disease. Orlando: Grune Stratton, pp 67-86. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, REFERENCES Mullis KB, Erlich HA (1988):Primer directed enzymatic amplifiBrahm J , Vento S, Rondanelli EG, Ranieri S, Fagan EA, Williams R, cation of DNA with a thermostable DNA polymerase. Science Eddleston AL (1988): Sequential studies of pre-S2 antigenemia 239:487-491. and anti-pre S2 antibodies in relation to viral replication in acute Shimizu Y, Ida S, Matsukara T, Yuasa T (1984): Determination of hepatitis B followed from the early incubation phase. Journal of hepatitis B virus DNA in serum by molecular hybridization. Medical Virology 24:205-209. Microbiology and Immunology 28:1117-1123. Budkowska A, Dubreuil P, Gerlich WH, Lazizi Y, Pillot J (1988b): Stibbe W, Gerlich WH (1984): Variable protein composition of hepaOccurrence of pre-S1 antigen in viremic and nonviremic carriers of titis B virus containing the pre-S sequence. Journal of Virology hepatitis B surface antigen. Journal of Medical Virology 26:21752:396-401. 225. Takahashi K, Kishimoto S, Ohnuma H, Machida A, Takai E, Tsuda F, Budkowska A, Dubreuil P, Pillot J (1988a):Prognostic value of pre-S2 Miyamoto H, Tanaka T, Matsushita K, Oda K, Miyakawa Y, epitopes of hepatitis B virus and anti-pre S2 res onse evaluated by Mayumi M (1986):Polypeptides coded for by the region re S and monoclonal assays. In Zuckerman A J (ed): “&a1 Hepatitis and ene S of hepatitis B virus DNA with the receptor for pokmerized Liver Disease.” New York: Alan R. Liss, Inc., pp 287-289. guman serum albumin: expression on hepatitis B particles proCourouce-Pauty AM, Plancon A, Soulier J P (1983): Distribution of duced in the HBeA or anti-HBe phase of hepatitis B virus HBsAg subtypes in the world. Vox Sanguinis 44:197-211. infection. Journal of fmmunology 136:3467-3472. Gerken G, Manns M, Gerlich WH, Hess G, Meyer zum Buschenfelde Theilmann L, Klinkert MQ, Gmelin,K, Salfeld I, Schaller H, Pfaff E KH (1987):Pre-S encoded surface proteins in relation to the major (1986a):Detection of pre-S1 proteins in serum and liver of HBsAgviral surface antigen in acute hepatitis B virus infection. Gastroositive patients: A new marker for hepatitis B virus infection. enterology 92:1864-1868. kepatology 6:186-190. Hadziyannis S, Raimondo G, Papaioannou C, Anastassakos C, Won Theilmann L, Klinkert MQ, Gmelin K, Salfeld J , Schaller H, Pfaff E D, Sninsky J, Shafritz D (1987):Expression of pre-S gene-encode8 (1986b):Detection of pre-S proteins in serum and liver of HBsAgproteins in liver and serum during chronic hepatitis B virus ositive patients: a new marker for hepatitis B virus infection. infection in comparison to other markers of active virus replicaRepatology 6:186-190. tion. Journal of Hepatology 5253-259. Wong DT, Nath N, Sninsky JJ (1985): Identification of hepatitis B virus poly e tides encoded by the entire pre-S open reading frame. Heermann K-H, Goldmann U,Schwartz W, Seyffarth T, Baumgarten Journal o!&ology 55:223-231. H, Gerlich WH (1984):Large surface proteins of hepatitis B virus containing the pre-S sequence. Journal of Virology 52:39&402.
in such a replicative phase. A recently developed sensitive method for detection of HBV-DNA by gene amplification [Saiki et al., 19881 may clarify this issue. Despite this controversy, pre-S1 and pre-S2 were found in the present study on 79.3% and 84.5%,respectively, of HBsAg-positive specimens vs. 3.8% and 4.6% of HBsAg-negative sera, suggesting that the presence of pre-S proteins in serum correlates strongly with that of HBsAg. Our findings are, therefore in agreement with those reported by Budkowska et al. [1988b1, who showed that 65%of sera from anti-HBe-positive chronic HBsAg carriers reacted positively to the pre-S determinant, and with those of Hu et al. [19871who demonstrated that the presence of pre-S proteins in serum is only a function of the plasma HBsAg titer. Finding of pre-S in the absence of viral replication is also supported by an in vitro study which demonstrated that pre-S proteins are present in the supernatant of a PLC/PRF/5 cell line producing HBsAg but no viral particles [Hu et al., 1987; Marion et al., 19801. It is concluded on the basis of the present study, that despite the strong correlation between the presence of pre-S proteins and markers indicating active viral replication, tests for pre-S proteins cannot simply be regarded equivalent to HBeAg and conventional HBVDNA tests for the detection of viral replication and infectivity in HBsAg-positive patients, since pre-S1 and/or pre-S2 proteins persist in serum of the vast majority of patients after disappearance of HBeAg and HBV-DNA from serum. Furthermore, the presence of pre-S does not reflect active chronic liver disease. However, the presence of pre-S1 and or pre-S2 proteins in serum is strongly correlated with that of HBsAg.
