CA2250076A1 - Method for predicting the outcome of hepatitis b infection - Google Patents

Abstract

The present invention is concerned with prognosis of the development of Hepatitis B infection. The method according to the invention may also be used to monitor antiviral therapy with, for example, antiviral compounds like alpha-interferon. The present invention is based on the finding that the presence of specific biomolecules able to form complexes with HBeAg is a prognostic marker for the further development of an <u>acute</u> Hepatitis B infection. It has been found that based on the presence or absence of said specific biomolecules or complexes of said biomolecules and HBeAg during the acute phase of infection it can be predicted whether patients will recover and clear the virus after acute infection or will develop a chronic infection. The present invention therefore involves the use of an assay for the detection of circulating complexes comprising specific biomolecules bound to HBeAg in a method for predicting the course of infection in patients with an acute HBV
infection. The present invention provides a new method for the prognosis of the development of Hepatitis B infection which can also be used to monitor therapy with antiviral agents and to indicate what the effect of such therapy is.

Description

CA 022~0076 1998-09-18 METHOD FOR PREDICTING THE OUTCOME OF HEPATITIS B INFECTION
The present invention is cGncerl,ed with a method for predicting and ri,or,iloring the course of an infection with the Hepatitis B virus (HBV). Furthermore the s invention is concerned with antibodies and amino acid sequences that can be used in such methods.
Hepatitis B virus (HBV) c~ ses acute and chronic infection in man. The infection~ can be subclinical, lead to minimal or severe acute hepatitis and can progress to chronic infection eventually leading to li~/er cirrhosis and primary liver carcinoma.
o Hepatitis B virus is a small DNA virus which consists of a lipid envelope and an inner core containing the viral genome and a DNA polymerase. The HBV core gene (nucleotides 1814-2449) is divided in a pre-core and a core region composed of 29 and 183 codons respectively. The gene codes for two functionally dirrere"t proteins: a 21.5 kDa protein that assembles to form 15 nucleocarsid particles (HBcAg) in the cytoplasm of infected cells and a pre-core protein, which is directed to the endoplasmic reticulum, subsequently N- and C-terminally processed and secreted as e-antigen (HBeAg).
HBeAg was firstly identified in 1972. It is present in a nonl~a, (icul~te state in the serum of HBV infected patients. The presence of HBeAg is ~e~ ~erally considered 20 to be an indication of active viral replication in the infected host. Experiments indicated that HBeAg may be a complex of antigens and up to three precipitin lines, designated e1, e2 and e3, have been detected by agar gel diffusion in sera from individual patients. Both IgG-bound HBeAg having a molecular size of approximately 300 kDa and a smaller HBeAg component with a molecular weight 25 of approximately 16 kDa have been detected in the sera of HBV-infected patients. HBcAg is related to HBeAg on basis of the fact that HBcAg can be converted into HBeAg upon denaturation and/or limited proteolysis.
Immunological distinction between HBcAg and HBeAg has become possible with the introduction of monoclonal antibodies specific for either antigen. HBcAg has30 been proposed to have one major discontinuous epitope but also the presence of linear HBcAg determinants and of an internal determinant have been suggested.
By the use of different murine monoclonal antibodies two distinct B-cell epitopes on HBeAg have been detected. The HBe/alpha, or HBe1 is a linear determinant 35 and resides in the amino acid sequence (76) L-E-D-P-A-S-R-D-L-V-V-S-Y (89).
The HBe1 determinant overlaps with the conformational HBc determinant and appears to be also exposed in an HBe conformation on the core particle surface (Salfeld et al., J.Virol., 63, 798-808, 1989).
HBe/beta or HBe2 is a discontinuous determinant and requires for its correct 40 conformation not only amino acid sequences ending around amino acid 138 but also the indirect or direct intramolecular participation of aminoterminal amino acid sequences. The HBe2 determinant is either absent or inaccessible on the core particle surface (Salfeld et al., 1989). According to Sallberg et al.
(Mol. Immunol., 28,719-726, 1991 ) murine monoclonal antibodies and human 45 sera may recognize the HBe2 epitope as a linear determinant residing around aa 130. Two anti-HBe2 monoclonal antibodies were found to recognize the CA 022~0076 1998-09-18 sequence T-P-P-A-Y-R at residues 128-133. By substitution synthesis the PP
was shown to be essential for the recognition by both monoclonal antibodies.
Monoclonal antibodies that recognized a peptide containing the linear HBe2 epitope as a single peak were those with the most efficient inhibiting activities in 5 the anti-HBe RIA. In patient sera reactivity to the HBe2 epitope containing peptide was mainly found in patient sera that gave strong inhibition in an anti-HBe RIA (95%).
The function of HBeAg is not clear but it may protect the virus inrected hepatocytes by blocking cytotoxic T cells during infection. In 1990 Milich et al.
lO (Proc.Natl.Acad.Sci., USA, 87, 6599-6603,1990) suggested that HBeAg may induce immunologic tolerance in utero. Expression of HBeAg may represent a viral sll aleyy to guarantee persistence after perinatal infection.
When patients come nearer to seroconversion to anti-HBe they are more likely to recover from the infection and to respond to antiviral therapy.
15 Little is known about the mechanisms and factors which determine the course of infection. Therapy of chronic HBV infected patients usually involves treatment with interferon. However, only a limited number of patients respond to interferon therapy with complete virus elimination. It is difficult to predict whether an infection will be self-limiting or chronic and which patients will eventually 20 respond to i, lte, rerol~ ll ,erapy. Therefore a need exists for reliable methods for the prognosis of Hepatitis B. lf such a method is available it can be established whether a patient should be treated with anti-viral agents and the effect of such treatment can be predicted or monitored.
25 The present invention is concerned with proy,)osis of the development of hepatitis B infection. The method according to the invention may also be used tomoi,ilor antiviral ~I,er~py with, for example, antiviral co~,pounds like alpha-interferon.
Some methods have already been described relating to prognosis of hepatitis B.
30 In WO 91/1478 a method is described wherein, to help predicting the outcome of hepatitis B infection a patient sample is tested for the presence/absence of a pre-core mutant. However, other studies (Guenther et al. Virology 187(1), 271 -279,1992) have shown that the appearance or lack of mutations in the pre-C
region was independent of virus elimination. A method for prognosis of Hepatitis35 B infection wherein the ratio of anti-HBS IgG and IgM (IgG/lgM) is used for the prediction of development of chronic hepatitis B is described in SU1751680.
Many workers have focused on the distinction between acute and chronic Hepatitis B infections. The distinction between acute and chronic HBV infection is important in terms of prognosis and possible treatment modalities. Antibody to 40 HBV X antigen has been reported to be preferentially produced in chronic HBV
infections (Moriarty et al., Science, 227, 429433, 1985).
It has also been described that higher levels of IgM anti-HBc are generally produced during the acute phase as compared with chronic infection and this quantitative difference has become the only serological means of differentiating45 an acute HBV infection of an acute elevation of a chronic infection (Gerlich et al., J.Clin.Microbiol.,2, 228-293, 1986). Additionally, it has been reported that CA 022~0076 1998-09-18 separalion of IgM anti-HBe by molecular weight is useful for distinguishing between an acute HBV and a chronic hepatitis B infection (Tsuda et al., Ga~-oe,)terology, 87,159-164, 1984). The usefulness of IgM anti-HBc assays in the differentiation of acute from chronic HBV infection has also been questioned.
All these methods however do not provide a possibility of predicting the course of the infection.
This is also the case in Maruyama et al. who investigated the presence of circulating immune complexes of HBeAg during the chronic phase of infection.
Maruyama et al. (J.lmmunol.Meth., 155,65-75, 1992) developed an assay to o detect circulating immune cG~ xes in sera from patients with a chronic HBV
infection. This assay consists of a solid phase coated with antibodies raised toan HBeAg/ayw derived peptide comprising amino acids 73-87 (exclusive of the procore sequence) designated as HBe73-87:GVNLEDPASRDLWSC Accor~ing to Maruyama et al. the anti-peptide antibodies do not compete with ar,libodies raised against native HBeAg. Antibody competition experiments were performed in both directions and yielded the same results. It was therefore concluded that, the lack of competition is not c~used by a difference in antibody avidity between the anti-peptide antibodies and antibodies raised against the native prot~
Acc~rding to Maruyama et al. the selected anti-peptide antibodies bind at uniquesites which do not significantly interfere with simultaneous binding by anlibod;es raised against the native protein. The assay for the detection of immunoco~ lcxes is also described in WO 94/08597. In further articles of the same authors (Maruyama et al., Gasl~ oe~ ~terology, 105, 1 141-1 151, 1993) the optimization of this assay is described; The anti-peptide antibodies were replaced by a l"or,oclonal antibody that efficiently binds native HBeAg that is simultaneously complexed to serum anti-HBe antibodies. This assay was used in detecting silent immune responses in patients with chronic HBV infections.
In 1994 Maruyama et al. (Gastroenterology, 106, 1006-1015,1994) published a study including patients with acute HBV who recovered completely from illness and patients with chronic HBV who were persistently positive for HBeAg and HBsAg for more than one year. It was observed that the serum levels of suspected anti-HBe antibody, HBeAg/anti-HBe complexes and HBsAg/anti-HBs complexes are elevated in patients with chronic HBV compared with patients with acute HBV.

In contrast to the above mentioned findings where immunocomplexes were detected during the chronic phase of infection, the present invention is based on the finding that the presence of specific biomolecules able to form complexes with HBeAg is a prognostic marker for the further development of an acute Hepatitis B infection. It has been found that based on the presence or absence of said specific biomolecules or complexes of said biomolecules and HBeAg during the acute phase of infection it can be predicted whether patients will recover and clear the virus after acute infection or will develop a chronic infection. The present invention therefore involves the use of an assay for the ~ 45 detection of circulating complexes comprising specific biomolecules bound to HBeAg in a method for predicting the course of infection in patients with an CA 022~0076 1998-09-18 4 PcTlEp97lol3 acute HBV infection. A group of patients, all infected with the same virus strain from the same source, and all HBeAg positive in co~ ercially available diagnostic tests (Abbott HBeAg EIA; Organon Teknika Hepanostika HBeAg/anti-HBe) was tested for the presence of said biomolecules, three Illonl~,s after s infection. When the course of infection of these patients was monitored it tumed out that patients positive for said molecules in the acute phase of the infection cleared the virus, while patients negative for said molecules developed a late seroconversion or chronic infection.
Based on these findings the present invention provides a new method for the o prognosis of the development of Hepatitis B infection which can also be used to monitor therapy with antiviral agents and to indicate what the effect of such therapy is.
The biomolecules which are detected with the method according to the present invention (the presence of which is indicative for the further course of infection) lS are presumably antibodies of the IgG type. Said biomolecules are capable of masking an epitope on the HBeAg that is located in the region comprising amino acids 85-109 of the complete HBeAg amino acid sequence (When the numbering started at the beginning of the HBeAg sequence). With masking is meant that, when the biomolecules have formed a complex with HBeAg, certain 20 antibodies that normally recognize the epitope are prevented from binding to the HBeAg when the HBeAg is present as a cori~l2x with the biomolecules.
The method according to the invention can be pe~ rormed in several ways.
The presence or absence of the biomolecules can be determined in several ways. For example by determining the presence of complexes (biomolecules 25 bound to HBeAg). An assay can be based on the capability of the biomolecules to mask an epitope on the HBeAg for binding of certain antibodies specific for said epitope. Said antibodies will only be capable of binding to the HBeAg in case the biomolecules are absent. In case biomolecules are bound to the HBeAg the antibodies will not be able to bind to their epitope. Thus, by detecting 30 whether any HBeAg will bind to said antibodies, the presence or absence of the biomolecules can be determined.
The antil,odies used are preferably monoclonal antibodies. The antibodies used with the method of the invention are antibodies preferably having a low affinitytowards the HBeAg present in the sample compared to the affinity of the 3S biomolecules optionally complexed to the HBeAg. As a result of their lower affinity towards the HBeAg, the antibodies only bind to HBeAg in the absence of biomolecules capable of masking their epitope. The antibodies can be used, for example, coated on a solid support. Immune complexes formed on the solid phase may, for example, be detected using another anti-HBe antibody 40 recognizing HBeAg in all forms or anti-human antibodies. This second antibody might be con,ug~ted to any label.
A negative reaction in such an assay will indicate that HBeAg is present as a complex in the sample and thus not capable of binding to the antibodies on the solid phase. (Provided of course that in the sample under investigation the 45 presence of HBeAg has been established by way of a screening procedure using an assay detecting HBeAg positivity in all cases. For example, the Abbott CA 022~0076 l998-09-l8 HBeAg EIA, OrganGn Teknika Hepanostika HBeAg/anti-HBe or the HBe MAB
assay (see example 2) or using alternative techniques such as SDS-page or immunoblotting.) A positive reaction in the assay will indicate that the sample is HBeAg positive and that the HBeAg epitope can be approached by antibodies.

- 5 Apparently, HBeAg is present in non-complexed form or as a complex in which the epitope recognized by the antibody is not masked.
Preferred antibodies used with the method according to the invention to detect the presence or absence of biomolecules bound to HBeAg are monoclonal antibodies having the same reactivity towards HBeAg as monoclonal antibodies produced by the cell line deposited at the ECACC under no. 95090611.
Monoclonal antibodies produced by the cell line deposited at the ECACC under no. 9509611 (HB.OT95A) recognize an epitope on HBeAg located around the sequence RDLWNWNTN. This epitope overlaps with the sequence defined as HBe1 by Salfeld et al. (J.Virol., 63(2), 798-808, 1989). Co~,pared to the HBe1 sequence the sequence of the recombinant HBeAg to which HB.OT95A
antibodies were raised has a substitution of an Asparagine (N) residue for the Serine (S) residue at position 87 in the HBe1 sequence as described by Salfeld et al. (The numbering adhered to by Salfeld et al. starts at the beginning of the HBcAg sequence. Different numberings are used throughout the literature.
When the numbering is started at the beginning of the complete HBeAg sequence 10 additional amino acids will be present at the N-terminal site and the amino acid with number 87 (according to Salfeld) will then have number 97. The numbering that starts at the beginning of the complete HBeAg sequence is the numbering adopted in the Examples and claims of this application.) Throughout this patent application HBeAg including a Serine (S) residue at position 97 willbe refened to as ~S-type" HBeAg and HBeAg including an Asparagine (N) residue at position 97 will be referled to as HBeAg of the UN-type".
Since the HB.OT95A antibodies have been raised against N-type HbeAg their affinity for HBeAg including the S-residue at position 97 (S-type HBeAg) is presumably lower than their affinity towards HBeAg including the N residue at that position. This might be the reason why the antibodies, raised against the N-type HBeAg, are not able to bind to the same region of HBeAg of the S-type, in the presence of the biomolecules. Presumably, because of their low affinity towards S-type HBeAg, antibodies raised to the N-type are unable to compete with biomolecules bound to HBeAg of the S-type.
It is clear that in the same way, other monoclonal antibodies can be raised having an affinity that, for example, is high for HBeAg of the "S-type" and low for the "N-type".
Whether a certain patient is infected with HBV including the N or the S residue at position 97 of HBeAg can be established in several ways. For example the reactivity of HBeAg present in the sample with antibodies having different affinities for the different "sub-types" (either S- or N-type HBeAg) can be compared. Furthermore a comparison of the reactivity of the sample with an antibody like HB.OT95A and the reactivity of the HBeAg in a sample with an anti-HBeAg antibody as present in a normal, commercially available, assay for the detection of HBeAg also gives an indication of what type of HBeAg is present CA 022~0076 1998-09-18 in the sample. A method to distinguish between HBeAg of the "S-type" and HBeAg of the UN-typeD is also part of the present invention. Said ",elh~
comprises the steps of:
(a) contacting part of a sample containing HBeAg with a solid phase coated with s antibodies having the same reactivity towards HBeAg as monoclonal antibodies derived from the cell line deposited at the ECACC under No. 95090611 and detecting any immune complexes fc.r",e~l on the solid phase and (b) contacting part of the sample with a solid phase coated with anti-HBeAg anlibo~'ics having the same reactivity towards HBeAg of the N-type and HBeAg l0 of the S-type and detecti,1g immune complexes formed on the solid phase, (c) determining from the ratio between signals obtained in step (a) and (b) whether the Itepatitis B e antigen present in the sample is of the N- or the S-type. Serum samples containing HBeAg or the "S-type~ will produce a lower signal in step (a) than serum samples containing HBeAg of the "N-typen. Thus 15 the ratio between signals obtained in step (a) and (b) will be higher in case of a sample co,)laining UN-type" HBeAg. The exact values of the ratio's characleristic for S- and N-type HBeAg will be dependent on the format of the assay systems used.
(Of course the nucleic acid sequence of the specific part of the HBeAg gene of 20 the particular virus can also be determined to est~hlish whether the HBeAg present in the sample is of the "N-type" or the US-typen.) Other ways of indicating the presence of the biomolecules i"~icali~/e of the further course of infection accordin~~ to the invention include the following:
Biomolecules can be determined by contacting sera with a solid support coated 25 with a peptide, a recoi"binallt protein or any other biomaterial mimicking at least the minimal sequence to which the biomolecules bind of the epitope. The biomolecules may be derived from dissociated HBeAg complexes, from anti-HBe positive sera or any other serum derived from HBV infected patients after clearance of the virus. Complexes formed between the peptides coated on the 30 solid phase and biomolecules present in the sera can be visualized in cJifrerent ways known to persons skilled in the art of diagnostic assays.
The biomolecules can also be detected by using an inhibition assay. In this caseHBeAg positive sera in which no biomolecules are present or aller"ali-/e forms of HBeAg such as rec~"~binant HBeAg can be used as positive sample. This 35 sample can be preincubated with samples unknown of including epitope masking biomolecules. Inhibition of the signal in HBeAg assays based on antibodies like,for example, HB.OT95A indicate the appearance of masking biomolecules in the sample.
Biomolecules forming complexes with HBeAg which can be related to the clinical 40 outcome of the HBV infection can be purified for example by affinity chromatography. Subsequently, these molecules can be used as reagent in HBeAg assays. The molecules will only react with non-complexed HBeAg and can thus also be used to detect the presence or absence of complexes in the sample of a patient with an acute HBV infection.
45 Solid supports which can be used with the method according to the invention are known in the art, for example, the inner wall of a microtest well or a cuvette, a CA 022~0076 1998-09-18 tube or capillary, a i"e"~bra,)e, filter, test strip or the surface of a particle such as, for example, a latex particle, an erythrocyte, a dye sol, a metal sol or mstal colnpound as sol particle, a carrier protein such as BSA or KLH.
Labeling substances which can be used are, inter alia, a radioactive isotope, a s fluorescent compound, an enzyme, a dye sol, metal sol or metal compound as sol particle. Depending on the nature and further characteristics of the assay method used the immunochemical reaction that takes place is a so called sandwich reaction, an agglutination reaction, a competition reaction or an inhibition reaction.
o Monoclonal antibodies that can be used with the method according to the invention are also part of the present invention. I\lonoclonal anlibGdies accordir,g to the invention are monoclonal antibodies having the same reactivitytowards HBeAg as monoclonal antibodies derived from the cell line deposited at the ECACC under No. 95090611.
l5 Immortalized cell lines capable of excreting monoclonal antibodies according to the invention are also part of the present invention.
The preparation of cell lines producing monoclonal antibodies may occur by, for example, by the Kohler and Milstein technique (Kohler and Milstein devised the techniques that resulted in the formation of monoclonal a,1libGdy-producing 20 hybridomas (G. Kohler and C. Milstein, 1975, Nature 256:495497; 1976, Eur. J. Immunol. 6:511-519)), I,~nsrol",a~ionwith Epstein-BarrVirus, ora direct l,ansfor"~ation technique of B-ly~phocytes with oncogenic DNA, or a direct fusion of human B-lymphocytes with a fusion partner being either a human or a mouse-human hybrid myeloma cell line, or a direct fusion of an EBV-transformed 25 B cell line with said myeloma cell lines.
A prefer,ed cell line accor-Jing to the invention is the cell line deposited at the European Collection of Animal Cell Cultures, Porton Down (~JK) under deposit No. 95090611. This hybridoma cell line was produced by the fusion of a mouse myeloma cells with a splenic Iymphocyte derived from a mouse previously 30 inoculated with reco,nbi"a"t HBeAg from E.coli.

BRIEF DESCRIPTION OF THE FIGURES:
35 Fiqure 1:
A) Abbott HBeAg EIA. Reactivity of eight sera (nrs. 162, 216, 255, 336, 181, 343, 098, 228) from donors accidently infected with HBV in a clinical setting (see example 3). Samples were taken during the acute phase of infection. All samples were tested HBeAg positive. Sera obtained from patients who became 40 chronically infected could not be discriminated from sera obtained from patients who recovered between 6 (recovery) or 9 (late recovery) months after infection.
B) HBe MAB assay. Reactivity of sera mentioned in Figure 1A. All samples were tested HBeAg positive. Sera obtained from patients who became chronically 45 infected could not be discriminated from sera obtained from patients who recovered between 6 (recovery) or 9 (late recovery) months after infection.

CA 022~0076 1998-09-18 C) HBe selection assay. Reactivity of sera mentioned in Figure 1A and B. Sera from patients who cleared the virus within 6 months after infection (negative =
A450 nm ~0.2 or S/CO <0.9) could be discriminated from sera obtained from 5 patients who became chronically infected or showed late recovery (positive =
A450 nm ~0.3 or S/CO >2.3).
Fi~ure 2.
A) Sera from HBV infected donors tested in the Abbott HBeAg EIA and the HBe l0 selection assay. Normal human serum (NHS) was used as negative control.
Positive control (PS) was obtained from a patient during the acute phase of infection. This serum sample is positive in both assays. Serum 343b was obtained from a HBV infected donor four years after infection. Serum 162a was obtained during the acute phase of infection from a donor who cleared the virus within 6 months. Serum 162b is obtained from the same donor as serum 162a after viral clearance ten months after primary infection.
B) Reactivity of the positive control (PS) mentioned in Figure 2A after incubation with normal human serum (NHS), serum 343b,162a and 162b. With exception of 20 NHS the original signal declines after adding one of above menliGned sera Fi~ure 3. Determination of HBe-S and HBe-N types by using the HBe MAB
assay and HBe selection assay. The total amount of arbitrary units (one arbitrary unit = reciprocal value of dilution) ~ssoci ~te~ with an absorbance of 1000 was 25 determined for each serum (AU1000) in both assays. AU 1000 deter",ined in theHBe MAB assay is divided by AU 1000 resolved the HBe selection àssay (y-ax).
Ratios below 0.03 indicate the HBe-S type. Ratios above 0.13 indicate the HBe-N type. S and N types were checked by determination of the complete HBeAg DNA sequence.

EXAMPLES:
ExamPle 1:
This Example illustrates how antibodies according to the invention wereprod~ Iced and the corresponding epitope was mapped.
Murine monoclonal antibodY HB.OT95A
Murine monoclonal antibodies were produced by injecting Balb/c mice with E.coli derived recombinant HBeAg or HBcAg in Freund's complete adjuvans. After 2 months mice were boosted with the antigen mixed in Freund's incomplete adjuvans, which was repeated after two weeks. The best responding mouse 45 received an intravenous dose of the recombinant antigen dissolved in PBS.
Three days later the spleen was removed and splenic Iymphocytes were fused CA 022~0076 l998-09-l8 WO 97/35204 q PCT/EP97/01351 with P3x63Ag8.6.5.3 mouse myeloma cells using polyethylene glycol 1000 accordi"gtosl~ rdmethods. HyLriJ~ acellswereexaminedforthe production of HBeAg or HBcAg specific antibodies using human sera and/or reco,nbinant antigens. Reactive clones were recloned to 100% clonality by the s technique of single-cell cloning.
Human monoclonal HBe.OTHuO3 Peripheral blood was taken from a human chronic HBsAg carrier. To 100 ml blood samples 10.9 ml dextran (MW 200.000) 0.37 ml NaCI solution (0.27 g.ml~
1 ) and 1 ml heparin solution (510 IU ml~1 ) were added. The mixture was inc~ ~h~t~l for 1 h at room t~"~per~l~Jre to allow erythrocytes to settle. The upper layer contai"ing the leukocytes was isolated and spun down for 10 min at 2000 N kg~1. The cell pellet was resuspended in cor~plete culture medium (D~lll,ecco's modified Eagles medium (DMEM) 10% foetal calf serum) supplemented with 15 10% dimethylsulphoxide (DMSO) frozen and stored in liquid nilroge~,~
For transformation leucocytes were thawed washed once with complete culture medium and resuspended to 2x1 o6 cells ml~1. Epstein-Barr virus (EBV) containing supernatant from the Marmoset B95.8 cell line was added and the suspension distributed over a 24-well culture dish (Costar). After 16-24 h 20 incubation at 37~C under 5% C02 1X108 viable Iymphocytes were isol~te~l diluted to 1x104 cells ml~1 in culture medium and distributed over ten 96-well cluster dishes containing human fibroblasts as feeder cells. Cultures then were inc~ ~h~ted for 4~ weeks: medium was refreshed t~,vice a week. From wells containing microscopically visible transformed Iymphocytic clones supernatant 2~i was harvested to be tested for the presence of specific al~tiL.~lies to native HBeAg. Positive results were confirmed after 7 days. The cultures were continued and frozen in liquid nitrogen.
Svnthesis and purification of the ~ lated PePtides Abbreviations and symbols for amino acids and peptides are in accordance with the r~c~",n~enda~ions of the IUPAC-IUB joint Commision on Biochemical Nomenclature as given in: IUPAC-IUB joint Commision on Biochemical Nomenclature (JCBN) Eur. J. Biochem. (1984) 138: 9 - 37. Other abbreviations:
35 NMP N-methyl-pyrrolidone; Fmoc N-9-fluorenylmethyloxycarbonyl; PS
polystyrene.
Based on the HBeAg sequence published by Ono et al. in 1983 (adw) 25-mers were constructed covering the complete gene. The following peptides were obtained:

1)... : H-SKLCL GWLWG MDIDP YKEFG ATVEL-NH2 2)... : H-LWGMD IDPYK EFGAT VELLS FLPSD-NH2 3).... H-PYKEF GATVE LLSFL PSDFF PSVRD-NH2 4)... : H-TVELL SFLPS DFFPS VRDLL DTASA-NH2 45 ~;).. : H-LPSDF FPSVR DLLDT ASALY REALE-NH2 6)... : H-SVRDL LDTAS ALYRE ALESP EHCSP-NH2 CA 022~0076 1998-09-18 7)... ~ TASAL YREAL ESPEH CSPHH TALRQ-NH2 8)... : H-EALES PEHCS PHHTA LRQAI LCWGE-NH2 9)... : H-HCSPH HTALR QAILC WGELM TLATW-NH2 10).. : H-ALRQA ILCWG ELMTL ATWVG NNLED-NH2 11).. : H-CWGEL MTLAT WVGNN LEDPA SRDLV-NH2 12).. : H-LATWV GNNLE DPASR DLWN YVNTN-NH2 13).. : H-NLEDP ASRDL WNYV NTNMG LKIRQ-NH2 14).. : H-RDLW NY~NT NMGLK IRQLL WFHIS-NH2 15).. : H-VNTNM GLKIR QLLWF HISCL TFGRE-NH2 16).. : H-KIRQL LWFHI SCLTF GRETV LEYLV-NH2 17).. : H-FHISC LTFGR ETVLE YLVSF GVWIR-NH2 18).. : H-FGRET VLEYL VSFGV WIRTP PAYRP-N1~2 19).. : H-EYLVS FGVWI RTPPA YRPPN APILS-NH2 20).. : H-WIRTP PAYRP PNAPI LSTLP ETTW-NH2 The solid-phase synthesis of the peptides was carried out in a semi-a~,lo",ated manner. The Fmoc amino acid derivatives were obtained from Bachem (Bubendorf, Switzerland). The peptides were sy~ llhesi~ed on a TentaGel S i~AM
Fmoc resin (i~APP Polymere, Ti.bin$ien, Germany) via the FmocltBu chemistry.
The linker is of a Rink-amide type, which automatically yields a C-termina~ly amidated peptide. During solid phase peptide synthesis the amino acid side-chains were protected with acid-labile protecting groups: the Epsilon-aminogroup of Iysine with Boc, the delta-guanidino group of arginine with 2,2,5,7,8-pentamethylchrolnan~-sulphonyl (Pmc), the gamma-c~iJo~yl group of glutamic acid and the 13-carboxyl group of aspartic acid with OtBu, the ga",h,a-amide group of glutamine and the 13-amide group of asparagine with trityl (Trt),histidine and cysteine with Trt, the 13-hydroxyl group of serine and ~I)reonii ,e with tBu, and tyrosine with tBu. All reaclants were dissolved in NMP. The cleavage of the Fmoc groups was carried out with 25% (vol/vol) piperidine in NMP during 3 consecutive cycles of 6 min. Coupling of the first Fmoc amino acid derivative (Fmoc-Aaa-OH, 2.5 eq.) was performed by in situ activation with diisoprc pylcarbodiimide (DIC, 2.5 eq.) and 1 -hydroxybenzotriazole (HOBt, 2.6 eq.) The acylation reaction was monitored by bromophenolblue (the end-point of the acylation reaction is indicated by a yellowish-green color). The acylation reaction was followed by a capping-step with acetic anhydride in NMP. The fully protected peptides were cleaved from the resin during a 2-hr reaction at room temperature under nitrogen with 5% thioanisole (vol/vol), 3% ethanedithiol (vol/vol), 2.5% water (vol/vol), and 2% anisole (vol/vol) in trifluoroacetic acid (87.5% vol/vol) followed by precipitation in diethylether. The crude peptides were washed twice with diethylether, dried at the air, dissolved in water/acetonitrile (3:1) and Iyophilized.
The HPLC analysis and purification (when necessary) were carried out on a Beckman Gold HPLC system. HPLC analyses were performed on a RP-C2/C18 column (Superpack prepS, 4x2~0 mm, Pharmacia) at a fiow rate of 1 ml/min, using a 3 min isocratic elution with 0.1 % followed by a 30 min linear gradient CA 022~0076 1998-09-18 WO 97/35204 l ¦ PCT/EP97/01351 from in water (100%) to 75 % 0.1 % trifluoroacetic acid in acetonitrile. Peptides were detected by UV measurement at 206 nm.
Svnthesis and Purification of the amidated peptides OTP-145 and OTP-146.
5 Abbreviations and symbols for amino acids and peptides are in accordance with the rt:co,nmendations of the IUPAC-IUB joint Commision on Biochemical Nomenclature as given in Eur.J.Biochem. (1984) 138: 9-37. Other abbreviations: NMP N-methylpyrrolidone; Fmoc N-9-fluorenylmethyloxycarbonyl; PS polystyrene.
10 Sequence of OTP-145: H-NLEDPASRDLW N YVNTNMGLKIRG-NH2 Sequence of OTP-146: H-NLEDPASRDLW S Y~NTNMGLKIRG-NH2 The synthesis of the peptides was carried out on a Perkin Elmer/Applied Biosystems Inc. 433A peptide synthesizer using standard FastMoc 0.25 mmol procedures with UV-monitoring and feedback option. The Fmoc amino acid 15 derivatives were obtained from Bachem (Bubendorf Swikerland). The peptides were synthesized on a TentaGel S RAM Fmoc resin (RAPP Polymere Tubingen Germany) via the Fmoc/tBu chemistry. The linker is of a Rink-amide type which automatically yields a C-terminally amidated peptide. During solid phase peptidesynthesis the amino acid side-chains were protected with acid labile ~r~,tecting20 groups: the epsilon-aminogroup of Iysine with Boc the delta~uanidino group ofarginine with 2 2 5 7 8-pe"ta")et~,ylchroman~-sulphonyl (Pmc) the gamma-carboxyl group of glutamic acid and the 13-carboxyl group of aspartic acid with OtBu the ga"""a-amide group of glutamine and the 13-amide group of asparagine with trityl (Trt) histidine and cysteine with Trt the 13-hydroxyl group 25 of serine and theonine with tBu and tyrosine with tBu. All re~tants were dissolved in NMP. The cleavage of the Fmoc groups was carried out with 25%
(vol/vol) piperidine in NMP during at least 2 consecutive cycles of 1.5 min.
Coupling of the first Fmoc amino acid derivative (Fmoc-Aaa-OH 4 eq. 1 mmol) was pe, rormed by in situ activation with 2-(1 H-benzotriazol-1-yl)-1 1 3 3-30 tetramethyluronium hexafluorophospl~ale (HBTU) 1-hydroxy~e"~ot,iazole (HOBt 4 eq 1 mmol) and diiso~ropylethylamine (DIPEA). After coupling of each amino acid derivative (at least 20 min) no check for completion of the acylationreaction was carried out. The acylation reaction was followed by a capping-step with acetic anhydride in NMP. The fully protected peptides were cleaved from 35 the resin during a 2-hr reaction at room temperature under nitrogen with 5%
thioanisole (vol/vol) 3% ethanedithiol (vol/vol) 2.5% water (vol/vol) and 2%
anisole (vol/vol) in trifluoroacetic acid (87.5% vol/vol) followed by precipitation in diethylether. The crude peptides were washed twice with diethylether dried at the air dissolved in water/acetonitrile 3:1 ) and Iyophilized.
40 The HPLC analysis and purification (when necessary) were carried out on a Beckman Gold HPLC system. HPLC analysis were performed on a RP-C2/C18 column (Superpack prepS 4x250 mm Pharmacia) at a flow rate of 1 ml/min using a 3 min isocratic elution with 0.1% trifluoroacetic acid in acetonitrile followed by a 30 min linear gradient from in water (100%) to 75%. Peptides were 45 detected by UV measurement at 206 nm.

CA 022~0076 1998-09-18 WO 97/35204 12. PCT/EP97101351 EPitoPe macpinq of HB.OT95A
Peptides were dissolved in DMSO (minimal concer,l,alion 0.5 mg/ml) and diluted in 0.05 M Na2C03 (pH 9.6) to a final concenl,ation of 5 ~ugr/ml. Diluted peptides 5 were coated overnight at room temperature in microliler,.,lates (96-wells), Each well was post-coated during 30 minutes using 0.05 M TRIS~HCI including 0,2%
casein. MAB HB.OT95A was diluted in 20% normal goat serum and 1% Triton in PBS pH 7.4. Dilution series between 1:500 to 1:16.000 were tested by adding 100 ,ul of each diluted sample per well. After incubation for 1 hour at 37~C thel0 sample was repl~oe~l by rabbit anti-mouse conjugate (DAKO P0260) after four extensive wash-steps. The conjugate was diluted 1:1000 as described above.
Organon Teknika UniForm ready-for-use TMB su~sl(ate was used to ViS! l~l j7l3 the reacivity of the peptide. After 20 minutes the reaction was terminated using1 M H2SO4 and absorbance was measured at 450 nm.
lS Results indicated that HB.OT95A showed high affinity for peptides 12 and 13 (S/CO ratio 30 in both cases). For peptide 14 S/CO ratio 16 was obtained. All other peptides showed S/CO ratio's below 1. From these findings was concluded that the epitope of HB.OT95A is located around sequence LATWV GNNLE
DPASR DLWN WNTN TNMGL KIRQR. This sequence includes the HBe1 20 epitope desc;, ibed by Salfeld et al.
In an identical experi,nent peptides OTP-145, OTP-146 and pepti~le 13 were tested simultaneously. Peptide OTP-145 (amino terminal aminoacid = glycine) and peptide 13 (amino terminal aminoacid = glutamine) showed no dirrere,1ces in reactivity with HB.OT95A. Hence must be concluded that the amino terminal 25 aminoacid of the peptide is not essential for its reactivity with HB.OT95A and may be replaced by Q or G. On the cont~ ai y serious clirrerences in reactivity were observed after using peptides OTP-145 and OTP-146 (HB,OT95A
concentralion 1:4000; OTP-145 S/CO 14; OTP-146 S/CO 5.5). It was concluded that aminoacid 97 (S or N) is considerabely affecting the reactivity of the 30 peptide. HB.OT95A shows low affinity for peptide OTP-146 comprising S at position 97.
ExamPle 2:
35 This example illustrates how, based on antibodies produced as described in Example 1 an immunoassay was constructed which will be referred to as the "
HBe selection assay". A second assay was constructed to ensure HBeAg positivity of a sample (referred to as HBe MAB assay).
40 HBe selection assav.
The assay is based on two monoclonal antibodies. MAB HBe.OTHuO3-HRP
(HRP = horse-radish peroxidase) is used as conjugate in a 1:8000 dilution. HRP
is conjugated to IgG according to the SPDP conjugation procedure (Pharmacia) Briefly, HBe.OTHuO3 was desalted using PD-10 chromatography (Pharmacia).
4S MAB-PDP was prepared by adding û.040 M SPDP (diluted in ethanol) to desalted MAB HBe.OTHuO3 (molar ratio MAB-lgG:SPDP = 1:13). After 30 CA 022~0076 1998-09-18 minutes the mix was desalted as mentioned above. HRP-PDP was prepared by adding 0.040 M SPDP to HRP (molar ratio SPDP:HRP = 1.6:1) and inc~h~ting for 30 minutes before desalting. HRP-SH was prepared by adding HRP-PDP,1 M HAC pH 4.3 and 1.6 M DTT in a ratio of 25:2.5:1. After incuh~tion for 15 5 minutes the mix was desalted as mentioned above. lgG-PDP and HRP-SH were coupled by mixing both components in a ratio of 13.4:1 mollmol and inC~h~ti~g during 30 minutes at AT followed by overnight incubation at 2-8~C. The conjugate was stored at -20~C until needed.
Murine MAB HB.OT9SA (selector) is coated on the solid phase. Briefly, l0 HB.OT95A is diluted to a final ~nc6i It,ation of 40 l~gr/ml in coathuffer (0.05M
NaHCO3 pH 9.6). The mix is inc~h~ted overnight with the solid phase (microtitre plates Greiner 96-wells) at AT (135 ,ul/well). After incubation the MAB solutionwas replaced by the same volume buffer (0.2 M TrislHCI, 0.2 M NaCI, 0.05%
Tween 20 pH 7.4) and inu ~h~ted for 10 minutes. The buffer was then replaced 15 subsequently by above mentioned buffer excluding Tween 20, a 0.2 M TRISIHCI
buffer including 0.2 M NaCI pH 7.4 and a 0.05 M TRISIHCI buffer pH 7.4. Finally,the plates are dried using nitrogen gas. After 24 hrs the plates were packed in sachets and stored at 4~C.
EIA was performed to standard procedures. Briefly, 100 ,ul sample is incub~ted 20 for 1 hour at 37~C followed by four wash-steps using PBS-Tween pH 7.4. After washing 100 ,ul conjugate (diluent 20% go~tserum and 1 % Triton X100 in PBS
pH 7.4) is added, incubated and removed as above mentioned. Hereafter 100 lul TMB ready-for-use substrate (OrganGn Teknika UniForm ll substrate) is added and incubated during 30 minutes at room temperature. The reaction is 25 terminated using 1 M sulfuric acid (H2S04). Absorbance is measured at 450 nm.
HBe MAB assav This assay is based on two monoclonal antibodies. As in the HBe selection 30 assay MAB HBeOT.HuO3-HRP is used as conjugate in a 1:8000 dilution. Sorin MAB 364C8 is coated on the solid phase (5 ,ug/ml). Monoclonal 364C8 was directly obtained from the Sorin Biomedica co,npany (Italy) and was used for practical purposes. The assay was merely constructed, since an assay was needed including the HBeOT.HuO3-HRP conjugate to ensure that specificity of 35 the selection assay was due to the HBOT95A monoclonal antibodies and not due to the monoclonal used in the conjugate. The choose of the Sorin monoclonal is not essential for the performance of the assay for this particularpurpose. This monoclonal can be replaced by any other anti-HBeAg monoclonal antibody recognizing HBeAg complexes.
40 EIA is performed according to standard procedures as mentioned above.
ExamPle 3: The example describes how using the HBe selection assay and the HBe MAB assay HBeAg positive samples were tested. From the results can be concluded that using the HBe selection assay the clinical outcome of the 45 infection could be predicted in the acute phase of infection.

CA 022~0076 l998-09-l8 HBeAq Positive sera.
In November 1987 a group of women became accidentally infected with the HBV
virus in a clinical setting. About four months later the Tnfection became clinically evident and first blood samples were taken. In total eight women were s implemented in the study. Sequence analysis of the HBe region revealed that HBeAg was identical in all cases. The virus coded for HBeAg with a Serine (S) residue at position 97 of the HBeAg. For definition of the HBsAg subtype (ayw2) also a selected genomic region of ~IBsAg was amplified and sequenced according to Norder et al. 1992 (J.Gen.Virol. 73: 3141-3145). The sequence o analysis for HBeAg was performed as described below.
Sequencin~ of HBe PCR frac",)~"ls.
From all blood samples the virus strain and/or HBe heteroge"eity was resolved by deter",i,)ation of the complete HBe sequence. DNA was isolnte~ accor~ g to S the Boom protocol (Boom etal. J.Clin.Microbiol. 28 495-503 1990). Briefly to nine volumes of Iysis buffer one volume of serum was added (225 ~l and 25 ~l respectively). To this mixture 70 ~JI of silica suspension was added. The solution was incubated for 10 minutes at room te",per~t,~re and regularly vortexed. Afterincubation the silica was spun down by centrifugation for at least 15 sec. at 20 highest speed. The supernatant was removed. The pellet was dissolved in 1 ml of L2 wash buffer (0.05 M Tris/HCI pH 6.4 5M GuSCH). The silica was spun down again and the pellet washed once more with 1 ml of L2 wash buffer. After the removal of the supernatant the pellet was washed twice with 70% ethanol and one time with acetone (in the same way as the previous steps). The pellet 25 was dried in a heating block at 56~C for 10 minutes. The nucleic acid was eluted by adding 100 ul water and incubated at 56CC for 10 minutes (vortex once after 5minutes). After centrifugation for 2 minutes at 10.000 9 +/- 80 ,ul nucleic acidsolution was transferred into a new eppendorf tube. After another centrifugationstep in order to remove all the silica +/- 70 ~JI nucleic acid solution was 30 transferl~l to a new tube and stored at -20~C.
PCR was performed accor~ing to normal PCR protocol (total volume 50 ~I). The input of the DNA was 2 ,ul of the isolated fraction. If no product afler PCR wasobtained another PCR was performed with a higher DNA input. The primers used for the PCR were primers existing of specific HBe sequence 35 elongated with M13 forward or M13 reverse sequence.
4) Primer sequentie:
HBe M13 forward 5 CGA CGT TGT AAA ACG GCC AGT AGG AGG CTG TAG

40 HBe M13 reverse 5 CAG GM ACA GCT ATG ACC TTC TGC GAC GCG GCG

After the PCR was ready 2 ul of the product was checked qua,llilativeiy on a 2%
TAE agarose gel. The total product (50 ul) was put on a 2% agarose gel. The 45 right fragments (698bp) were cut out and purified according to the Qiaex protocol from Qiagen. After purification 1/10 of the purified fragment was put on CA 022~0076 1998-09-18 WO 97/35204 l ~ PCT/EP97/01351 a 2% agarose gel to check purification level. When the purification was pe, rorrned ~r, ecll~, the sequence was determined.
The sequence analysis was carried out according to the Sequitherm long read cycle sequencing kit (Epicentre Technologies, Madison USA). About 300 to ~00 ng of purified PCR fragment was used as DNA input. The primers used for the sequence analysis where M13 forward and M13 reverse. The reactions where run on an A.L.F. DNA sequencer (Pharmacia Biotech) according to the standard protocol.
Results Results indic~ted that during the acute phase of infeclion all samples were positive in the Abbott HBeAg EIA, Organon Teknika Hepanostika HBeAglanti-HBe and the newly developed MAB HBe assay suitable for the detection of all forms of HBeAg. Results are given in Figure 1A and B. Using the HBeAg lS selection assay different results were obtained. Sera from patients 162. 216, 255 and 336, who cleared the virus within 6 months after infection, remained negative while the sera from patients 343, 098 and 228 who became chronically infected were positive. Serum 181 from a patient who showed late seroconversion to anti-HBe (after more than nine months) was also positive in the selection assay (see Figure 1 C). The results show that in the acute phase of a HBV infection the selection assay can discriminate patients who become chronically infected or develop late seroconversion to anti-HBe from patients showing anti-HBe seroconversion within six months after infection. The discrimination is based on variation in immunoreactivity of HBeAg which is appearant in all sera according to leference assays. The differences in clinicaloutcome can not be related to variation of the virus because it was shown that the HBV HBeAg had the same sequence in all cases.
ExamPle 4:
This exampie shows how, by using an inhibition assay, it was confirmed that biomolecules able to mask the epitope of HB.OT9~A are present in samples tested negative in the Hbe selection assay and positive in another HBeAg assay.
Inhibition assay.
After sequencing the HBV HBe gene from all women involved in the study it became evident that they had been infected with the same virus strain. The HBe sequence showed no alterations during the course of infection. Variation in HBeAg immunoreactivity could therefore not have been caused by variation in HBeAg itself. It was suggested that one or more host derived factors present in the serum must play an important role. This was investigated by an inhibition experiment Serum from patient no. 343 was obtained during the acute phase of the infection (see Figure 2A). This sample was used as positive control (PS). First the PS wasdiluted 213 times using normal human serum. Subsequently, different sera, 343b, 162a/b and the positive control were diluted 1:1 with NHS and measured in the CA 022~0076 1998-09-18 WO 97135204 ~ ,~, PCT/EP97/01351 Abbott HBe EIA and the HBe selection assay. Results are shown in Figure 2A.
Sera 343b 162a and the PS were screened HBeAg positive in the Abbott HBeAg assay but only the PS was also positive in the selection assay. The samples 162b and normal human serum were negative in both assays.
5 Sample 343b and the PS were obtained from the same patient. The PS was taken during the acute phase of a HBV infection while serum 343b was obtained during the chronic phase four years after infection. Serum 162a was obtained from a patient during the acute phase of a HBV infection but patient 162 clearedthe virus within 6 months. Serum 162b was obtained after recovery ten morltl,s o after primary infection.
As a second step to 60 ~I positive control (2/3 dilution) 60 ~JI of various sera(100%) was added. The mixture was incub~ted for 8 hours at room temperature and subsequently the HBeAg was detected in the selection assay. The results (Figure 2B) indicated that the positive control showed a high absGrl,ance value 15 after incubation with normal human serum but signals declined if sera negative in the selection assay (indicating that biomolecules had bound to the HBeAg present) or anti-HBe positive in the Abbott anti-HBe assay were used instead.
These results clearly indicated that sera positive in the Abbott HBeAg EIA or the HBe MAB assay but negative in the HBe selection assay contain biomaterials 20 able to block the epitope of HB.OT95A. The same must be concluded for anti-HBe positive sera such as 162b.
Example 5:
25 This example shows that sera HBeAg positive in the Abbott HBeAg EIA but negative in the selection assay contain complexes of HBeAg and biomolecules and that these biomolecules presumably are anti-HBeAg antibodies of the IgG
type.
30 IgG and IgM molecules from HBeAg positive sera were purified using protein A
cl,ro",atography according to conventional procedures. HBeAg in the elution fraction was determined by using the Organon Teknika Hepanostika HBeAg/anti-HBe assay.
Subsequently IgG and IgM molecules bound to HBeAg were detected by using 35 an adapted EIA assay. In this assay the monoclonal 364C8 obtained from Sorin was coated on the solid phase. The HBeAg-lgG complexed HBeAg positive fraction was diluted 1/30 or 1/100 times (Dilution medium 20% normal goat serum in PBS (7.2) and 1% Triton-X100) and 100 ~l of the sample was contacted with the solid phase. Mouse anti-Human IgG or IgM HRP conjugate 40 was used as second antibody. In a second experiment various undiluted sera from different HBV infected donors were tested individually in the EIA without foregoing purification steps.
The results of both experiments indicated that sera positive in the Abbott or Organon Teknika HBeAg EIA but negative in the selection assay always 45 contained HBeAg bound to anti-HBe. It was observed that the anti-HBe bound to HBeAg was merely of the IgG subtype.

.

CA 022~0076 l998-09-l8 WO 97/35204 l7 PCT/EP97/01351 Example 6 This example illustrates that biomolecl ~les (anti-HBe an~ibod;es) obtained fromHBeAg complexes can inhibit HBeAg reactivity in the selection assay.

Purification of anti-HBe antibodies HBeAg complexes were purified by affinity chro,l,a~ography. MAB Sorin 364C8 (5 mg/ml Sepharose) was coupled to CNBr-Sepharose 4B (Pharmacia) accorcJing the method described by Pharmacia (In: Affinity Chromalograpl)y o Principles and Methods: Methods for coupling ligands to CNBr-activated Sepharose 4B). HBeAg was purified by adding 20 ml HBeAg positive serum (assays Organon Te~<nika Hepanostika HBeAg/anti-HBe (positive) and selection assay (negative)) to 1 ml Sepharose including MAB Sorin 364C8. The mixture was incubated overnight at 4~C under rotation. After incubation the unbound 15 fraction was separated from the gelmaterial and the gelmaterial was washed using PBS/Tween pH 7.2 washbuffer (Organon Teknika PBS washbuffer). The bound fraction was eluted using 5M NaSCN (batchwise 5 min AT). The elution fraction was further purified by gelfiltration (Sephadex G75 0.5x14 cm). In total 0.5 ml elution fraction was purified using TNE buffer (pH 7.4 0.1 M TRIS/HCI
20 0.1 M NaCI 0.1 M EDTA). Flow 1 ml/min. Fractions were collected at 0.5 ml intervals. The third 0.5 ml fraction (fraction 3) comprising the first proteins eluted was used for further experiments.
Inhibition experiment 25 Fraction 3 was diluted (1:10) in normal human serum or mixed directly 1:1 with a human serum HBeAg positive in the HBe selection assay (for example Serum nr 343 see example 3) or normal human serum (positive control). The mixture was incub~ted overnight at ambient temperature and tested in the selection assay as described before.

Results Results showed that signals of HBeAg positive sera declined in the selection assay after incubation with fraction 3. This in contrast to the positive control in which normal human serum was used. In total 15.3 % inhibition was obtained if 35 fraction 3 was used without previous dilution steps. Higher inhibition can beexpected if fraction 3 is concentrated before incubation with the HBeAg positiveserum.
The results indicated that fraction 3 contained biomolecules which are able to reduce the HBeAg signal in the selection assay. These biomolecules are derived 40 from HBeAg complexes which naturally appear in human sera HBeAg positive in various HBeAg assays but negative in the HBe selection assay. Apparently the biomolecules remain competent of epitope masking after partial purification.

CA 022~0076 l998-09-l8 WO 97/35204 PCT/EP97/013~1 ExamPle 7:
This example illustrates how HBeAg complexes can be dissociated using NaSCN.
s Method Various HBeAg positive sera (HBe MAB assay) were obtained from HBV
infected patients. The sera were diluted with 5M NaSCN or PBS pH 7.2. Normal human serum was used as negative control. Sera were mixed and incubated on o a shaker for 1 h at ambient temperature or overnight at 4~C. After incubation the sera were diluted using normal human serum (1:5, 1:10, 1:20, 1:30 and 1:100).
Using the HBe selection assay sera were tested on HBe reactivity.
Similar expe~ .en~s were pe,rorl"ed handling purified HBeAg complexes.
HBeAg complexes were purified as described in example 5. Non-purified sera 15 were used as positive control. After purification the elution fractions were diluted with NaSCN or PBS pH 7.2 (1:1 ) and tested in the HBe selection assay (1:~0).
Normal human serum was used as diluent. Starting material and unbound fractions was treated the same way.
20 Results Results indicated that in sera which were medium to high positive in the HBe selection assay (signals from 1000 A450 nm at 1:20 dilution) the addition of NaSCN slightly decreased the original signal at dilutions of 1:5. At dilutions of 1 :10 or more, however, there was no difference observed between PBS or 25 NaSCN treatment. In sera originally low positive in the HBe selection assay the HBe signal increased strongly if the serum was incubated with NaSCN. This effect was also observed at dilutions of 1:100. Sera which were completely negative in the selection assay and low positive in the HBe MAB assay remained negative after treatment with NaSCN. This effect was probably CAI ~sed by the 30 sensitivity of the assay.
After purification it was observed that HBeAg complexes could be identified in the elution fraction using the HBe MAB assay. It was also found that these complexes remained negative in the HBe selection assay after dilution with PBS
but became positive after treatment with NaSCN. In conclusion, dissociation of 35 HBeAg complexes with NaSCN results in the recognition of HBeAg epitopes in the HBe selection assay.

Example 8:
This example illustrates a new method which can be used to distinguish HBe N-type from HBe S-type.
Biolo~ical material 45 In total 22 sera were obtained from HBV infected patients of which 19 were collected during the chronic phase and 3 during the acute phase of the infection.

CA 022~0076 1998-09-18 WO 97/35204 Iq PCT/EP97/01351 All sera were tested HBeAg positive in the Abbott HBeAg IMX assay and the Organon Teknika HBe MAB assay.
Sequencinq 5 The complete HBeAg sequence was determined as described in example 3.
Sera were grouped as a HBe S-type or N-type on basis of aminoacid 97 (serine (S) or asparagine (N).
Method for alternative subtypinq 10 Each HBeAg positive serum was tested in an extensive dilution series (1/3, 1/5, 1110 till 1/1000) in both the Organon Teknika HBe MAB assay and the Organon Teknika HBe selection assay. Normal human serum (HBeAg negative in both assays) was utilised as diluent. Both assays were performed under the same conditions.
15 Using various models such as cumulative, doseresponse or sigmoidal the best fit of each dilution curve was determined and used for further c~lcul~tions. The total amount of arbitrary units (one arbitrary unit = reciprocal value of dilution) associated with an absorbance of 1000 was determined for each serum (AU
1000). Subsequenly, for each serum the AU 1000 determined in the Organon 20 Teknika HBe MAB assay was divided by the AU 1000 determined in the Organon Teknika HBe selection assay.
Results.
From our experiments could be concluded that the method is suitable to 25 distinguish the HBe S-type from the HBe N-type without determination of the HBeAg sequence (Figure 3). Our results indicated that ratio's below 0.03 impliedthe HBe S-type while ratio's above 0.13 are associated with the HBe N-type.

WO 97/35204 2~ PCT/EP97tO1351 SEQUENCE LISTING

(I) GENERAL rNFORMATION:

(i) APPLICANT:
(A) NAME: Alczo Nobel N.V.
(B) STREET: Velperweg 76 (C) CITY: Arnhem (E) COUNTRY: the ~
(F) POSTAL CODE (ZIP): 6824BM
(ii) TITLE OF INVENTION: Method for p,~ ;ling the outcome of ~p-'itic B infP~tir~-~
(iii) NUMBER OF SEQUENCES: 24 (iv) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC . . ' '~
(C) OPERATING SYSTEM: PC-DOStMS-DOS
(D) SOFI~tARE: Patentln Release #1.0, Version #1.25 (EPO) (2) INFORMATION FOR SEQ ID NO: 1:
(i) SEQUENCE CHARA~ RISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: l ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRI~IION: SEQ ID NO: 1:
Asn Leu Glu Asp Pro Ala Ser Arg Asp Leu Val Val Asn Tyr Val Asn Thr Asn Met Gly Leu Lys Ile Arg Gly (2) INFORMATION FOR SEQ lD NO: 2:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide WO 97/35204 2,1 PCT/EP97/01351 (xi) SEQUENCE DESCRI}~llON: SEQ ID NO: 2:
Asn Leu Glu Asp Pro Ala Ser Arg Asp Leu Val Val Ser Tyr Val Asn Thr Asn Met Gly Leu Lys lle Arg Gly (2) INFORMATION FOR SEQ ID NO: 3:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 24 arnino acids (B) TYPE: arnino acid (C) STRANDEDNESS: single (D) TOPOLOGY: ~ ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:
Lys Leu Cys Leu Gly Trp Leu Trp Gly Met Asp Ile Asp Pro Tyr Lys Glu Phe Gly Ala Thr Val Glu Leu (2) INFORMATION FOR SEQ ID NO: 4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 23 an~ino acids (B) TYPE: arnino acid (C) STRANDEDNESS: single (D) TOPOLOGY: l t .", (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:
Gly Met Asp Ile Asp Pro Tyr Lys Glu Phe Gly Ala Thr Val Glu Leu Leu Ser Phe Leu Pro Ser Asp (2) INFORMATION FOR SEQ ID NO: 5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids WO 97/35204 ~ , PCT/EP97/01351 (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: l ' ..n S (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:
Pro Tyr Lys Glu Phe Gly Ala Thr Val Glu Leu Leu Ser Phe Leu Pro Ser Asp Phe Phe Pro Ser Val Arg Asp (2)rNFORMATION FOR SEQ ID NO: 6:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: l ' (ii) MOLEC~LE TYPE: peptide (xi) SEQUENCE DESCRlPl'lON: SEQ ID NO: 6:
Thr Val Glu Leu Leu Ser Phe Leu Pro Ser Asp Phe Phe Pro Ser Val Arg Asp Leu Leu Asp Thr Ala Ser Ala (2) INFORMATION FOR SEQ ID NO: 7:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: arnino acid (C) STRANDEDNESS: single (D) TOPOLOGY:, ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7:
Leu Pro Ser Asp Phe Phe Pro Ser Val Arg Asp Leu Leu Asp Thr Ala Ser Ala Leu Tyr Arg Glu Ala Leu Glu WO 97/35204 ;13 PCT/EP97/01351 (2) INFORMATION FOR SEQ ID NO: 8:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: arnino acid (C) STRANDEDNESS: single (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide 15 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8:
Ser Val Arg Asp Leu Leu Asp Thr Ala Ser Ala Leu Tyr Arg Glu Ala Leu Glu Ser Pro Glu His Cys Ser Pro (2) INFORMATION FOR SEQ lD NO: 9:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: arnino acid (C) STRANDEDNESS: single (D) TOPOLOGY: l ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9:
Thr Ala Ser Ala Leu Tyr Arg Glu Ala Leu Glu Ser Pro Glu His Cys Ser Pro His His Thr Ala Leu Arg Gln (2) INFORMATION FOR SEQ ID NO: 10:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide ~ 4 (xi) SEQUENCE DESCRlErrlON: SEQ ID NO: 10:
Glu Ala Lcu Glu Ser Pro Glu His Cys Ser Pro His His Thr Ala Leu s Arg Gln Ala Ile Lcu Cys Trp Gly Glu (2) INFOl~MATION FOR SEQ ID NO: 11:
(i) SEQUENCE CHARA( lkRlSTlCS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: l ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPI ION: SEQ ID NO: 11:
His Cys Ser Pro His His Thr Ala Leu Arg Gln Ala lle Leu Cys Trp Gly Glu Leu Met Thr Leu Ala Thr Trp (2) rNFORMATION FOR SEQ ID NO: 12:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: ~
(D) TOPOLOGY: I ' ~."
(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCR~PTION: SEQ ID NO: 12:
Ala Leu Arg Gln Ala Ile Leu Cys Trp Gly Glu Leu Met Thr Leu Ala Thr Trp Val Gly Asn Asn Leu Glu Asp (2) INFORMATION FOR SEQ ID NO: 13:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 arnino acids (B) TYPE: amino acid (C) STRANDEDNESS: unknown WO 97/35204 2 ~ PCT/EP97/01351 (D) TOPOLOGY: I ' ....
(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 13:
Cys Trp Gly Glu Leu Met Thr Leu Ala Thr Trp Val Gly Asn Asn Leu Glu Asp Pro Ala Ser Arg Asp Leu Val (2) INFORMATION FOR SEQ ID NO: 14:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: .,--1 ."
(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 14:
Leu Ala Thr Trp Val Gly Asn Asn Leu Glu Asp Pro Ala Ser Arg Asp Leu Val Val Asn Tyr Val Asn Thr Asn 35 (2) INFORMATION FOR SEQ ID NO: 15:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 a nino acids (B) TYPE: amino acid (C) STRANDEDNESS: l ' ~...... -(D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 15:
Asn Leu Glu Asp Pro Ala Ser Arg Asp Leu Val Val Asn Tyr Val Asn Thr Asn Met Gly Leu Lys lle Arg Gln .

WO 97/35204 ~6 PCT/EP97101351 (2) INFORMATION FOR SEQ ID NO: 16:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: u (D) TOPOLOGY: I ' .~ .~
(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIP'IlON: SEQ ID NO: 16:
IS Arg Asp Leu Val Val Asn Tyr Val Asn Thr Asn Met Gly Leu Lys lle Arg Gln Leu Leu Trp Phe His Ile Ser (2) INFORMATION FOR SEQ ID NO: 17:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: u ' L..--(D) TOPOLOGY: ~ ' ....
(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ II) NO: 17:
Val Asn Thr Asn Met Gly Leu Lys lle Arg Gln Leu Leu Trp Phe His lle Ser Cys Leu Thr Phe Gly Arg Glu (2) INFORMATION FOR SEQ ID NO: 18:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: 1- ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 18:

WO 97/3S204 2,7 PCT/EP97/01351 Lys Ile Arg Gln Leu Leu Trp Phe His Ile Ser Cys Leu Thr Phe Gly s l0 15 Arg Glu Thr Val Leu Glu Tyr Leu Val (2~ INFORMATION FOR SEQ ID NO: 19:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: I ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 19:
Phe His Ile Ser Cys Leu Thr Phe Gly Arg Glu Thr Val Leu Glu Tyr Leu Val Ser Phe Gly Val Trp Ile Arg (2) INFORMATION FOR SEQ ID NO: 20:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: ~ ..O....
(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPI ION: SEQ ID NO: 20:
Phe Gly Arg Glu Thr Val Leu Glu Tyr Leu VaJ Ser Phe Gly Val Trp Ile Arg Thr Pro Pro Ala Tyr Arg Pro (2) INFORMATION FOR SEQ ID NO: 21:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: unknown WO 97t35204 PCT/EP97/01351 (ii) MOLECULE TYPE: peptide S (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 21:
Glu Tyr Leu Val Ser Phe Gly Val Trp lle Arg Thr Pro Pro Ala Tyr Arg Pro Pro Asn Ala Pro Ile Leu Ser (2) INFORMATION FOR SEQ ID NO: 22:
(i) SEQUENCE CHARA(-l ~RISTICS:
(A) LENGTH: 25 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: single (D) TOPOLOGY: I ' (ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 22:
Trp lle Arg Thr Pro Pro Ala Tyr Arg Pro Pro Asn Ala Pro lle Leu Ser Thr Leu Pro Glu Thr Thr Val Val (2) INFORMATION FOR SEQ ID NO: 23:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 41 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY:, ' (ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCR~TION: SEQ ID NO: 23:

(2) INFORMATION FOR SEQ ID NO: 24:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 38 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single WO 97/35204 ~ c~ PCT/EP97/01351 (D) TOPOLOGY: ~ ' .. ., (ii) MOLECULE TYPE: cDNA
s (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 24:

Claims (15)

CLAIMS:

1. Method for predicting the course of infection in patients with an acute Hepatitis B infection by detecting in an acute phase sample the presence of circulating complexes containing HBeAg and biomolecules capable of masking the epitope on HBeAg located in the region comprising amino acids 85-109 of the complete HBeAg sequence.

2. Method according to claim 1, wherein the sample is contacted with antibodies having the same immune-reactivity towards HBeAg as antibodies produced by the cell line deposited by the ECACC under no. 95090611.

3. Method according to claim 1 or 2 wherein the antibodies are monoclonal antibodies produced by the cell line deposited by the ECACC under no. 95090611.

4. Method according to any of claims 1-3, wherein the antibodies are coated on a solid phase.

5. Method according to claim 4, wherein immune complexes formed on the solid phase are detected by contacting the solid phase with labeled antibodies to HBeAg that do not recognize the epitope recognized by the antibodies on the solid phase.

6. Method according to claim 5, wherein said labeled antibodies are human monoclonal antibodies.

7. Method according to claim 1, wherein HBeAg containing complexes present in the sample, are dissociated, thus releasing any biomolecules present in the sample, where after the sample is contacted with a peptide or recombinant protein or any other material capable of specific binding to the biomolecules, and any bound biomolecules are detected.

8. Method according to claim 7, wherein a peptide is used including at least part of the amino acid sequence as depicted in SEQ ID No. 1 or SEQ ID No. 2.

9. Use of an assay for the detection of circulating complexes comprising specific biomolecules bound to HBeAg in a method for predicting the course of infection in patients with an acute HBV infection.

10. Test-kit, which can be used in a method for predicting the course of an acute Hepatitis B infection, comprising, - a solid phase coated with first antibodies having the same reactivity towards HBeAg as monoclonal antibodies derived from the cell line deposited at the ECACC under No. 95090611, - a labeled reagent, said reagent comprising second antibodies capable of reacting with immune-complexes formed on the solid phase between the first antibody and any HBeAg present in the sample.

11. Monoclonal antibodies having the same reactivity towards HBeAg as monoclonal antibodies derived from the cell line deposited at the ECACC under No. 95090611.

12. Use of monoclonal antibodies having the same reactivity towards HBeAg as monoclonal antibodies derived from the cell line deposited at the ECACC under No. 95090611 in a method for predicting the outcome of an hepatitis B infection.

13. Cell line capable of producing monoclonal antibodies according to claim 11.

14. Cell line deposited under no 95090611 at the ECACC.

15. Method for distinguishing Hepatitis B e antigen (HBeAg) of the N-type from HBeAg of the S-type comprising the following steps:
(a) contacting part of a sample containing HBeAg with a solid phase coated with antibodies having the same reactivity towards HBeAg as monoclonal antibodies derived from the cell line deposited at the ECACC under No. 95090611 and detecting any immune complexes formed on the solid phase and (b) contacting part of the sample with a solid phase coated with anti-HBeAg antibodies having the same reactivity towards HBeAg of the S-type and HBeAg of the N-type, (c) determining from the ratio between reactivities obtained in step (a) and (b) whether the Hepatitis B e antigen present in the sample is of the N- or the S-type.