WO1989004964A1

WO1989004964A1 - Determination of anti-pol as an early marker or viral hepatitis infection

Info

Publication number: WO1989004964A1
Application number: PCT/US1988/004076
Authority: WO
Inventors: Mark Feitelson; Baruch S. Blumberg; Irving Millman
Original assignee: Fox Chase Cancer Center
Priority date: 1987-11-16
Filing date: 1988-11-15
Publication date: 1989-06-01
Also published as: AU2913889A

Abstract

An effective immunoassay for the determination of hepatitis virus polymerase antibodies (anti-pol) as an early marker of hepatitis infections.

Description

DETERM - EARLY MARKER OF VIRAL HEPATITIS INFECTION

BACKGROUND OF THE INVENTION The present invention relates to the field of diagnostic techniques for the determination of immunologically reactive substances/ and in particular to an effective procedure for the determination of antibody to hepatitis virus DNA polymerase (anti-pol)/ as the first or an early marker of viral hepatitis infection.

The discovery and characterization of both antigen and antibody markers of hepatitis B infection and related viral infections has been essential in defining the patterns of virus gene expression and host immune responses involved in pathogensis. Among the most common markers of infection with hepatitis B virus (HBV)/ ground squirrel hepatitis virus (GSHV), woodchuck hepatitis virus (WHV) and duck hepatitis B virus (DHBV) are the surface (or envelope) antigen particles (HBsAg/ GSHsAg/ WHsAg and DHBsAg ) . Surface antigenic determinants have been found on the 42 nm diameter virion envelopes during productive infection and on small spherical. 22 n diameter surface antigen particles/ usually present in high concentrations in the blood of infected individuals during and after viremia. Antibodies to the surface antigen of these viruses (anti-HBs/ -GSHs, -WHs and -DHBs) are protective and signal resolution of infection. However/ the finding of many infected patients and animals lacking surface antigen and antibody demonstrates both the diagnostic and prognostic limitations of relying on surface antigen and antibody as markers of hepatitis.

A second serological marker of infection with hepatitis viruses is antibody to the nucleocapsid or core antigen (anti-HBc, -GSHc , -WHc and -DHBc). Anti-core has been described as a marker for continued viral replication and progression to chronic liver disease in some cases, although it has most often been considered a past marker of HBV infection. 7 Developments in Molecular Virology/ pp. 1-273 (Becker et al. eds., 1985). The recent finding of anti-HBc as a paradoxical marker associated with non-A/ non-B hepatitis infection/ reported by Koziol et al . / Ann.Intern.Med. 488-95 (1986)/ raises the question as to whether anti-HBc is specific to HBV infection/ especially^" when anti-HBc is the only serological marker of infection. The other antigenic markers associated with the nucleocapsid of these viruses comprise the e and X antigens and corresponding antibodies. The e antigen seems to be a cleavage or degradation product of the nucleocapsid- associated polypetides and is detected in sera that usually have a large number of virions/ suggesting it is a marker of virus replication. Anti-e often follows the clearance of e antigen from sera and indicates recovery from acute and chronic hepatitis. See: Developments in Molecular Virology/ supra. This e antigen/antibody system has been very helpful in identifying the status of HBV replication in both acutely or chronically infected individuals. However/ e antigen is often a transient marker of infection/ and seroconversion to anti-e sometimes takes place long after e antigen disappears. X is a new antigenic specificity associated with nucleo-capsid polypeptides. Feitelson, 6 Hepatology 191-98 (1986). The distribution of X antigenic determinants in sera and the significance of anti-X in infection remain to be studied. Determinations of the foregoing antigens and antibodies as markers of hepatitis infection may be achieved using various immunoassays including the radioimmunoassay (RIA)/ described in U.S. Patent 4,012,494 and the enzyme immunoassay (EIA) or enzyme-linked immunosorbent assay (ELISA), described in U.S. Patent 4,016,043. These so-called "sandwich assay" techniques may be conveniently carried out using monoclonal antibodies, as decsribed in U.S. Patent 4,376,110. Nevertheless, even with the screening assays currently available, some cases of post-transfusion hepatitis are still due to HBV. The further characterization and early detection of known and putative markers of infection may provide better diagnostic and prognostic tools in both prevention and treatment of HBV infection. SUMMARY OF THE INVENTION

In accordance with the presnt invention, there is provided a novel method for the determination of antibody to hepatitis virus DNA polymerase (anti-pol) which, experiments have shown, may be the earliest marker of viral hepatitis infection in humans, as well as in animals susceptible to hepatitis infection. In certain cases, anti-pol has been found to be the only evidence of hepatitis infection.

In performing the method of the invention, a test reagent is provided which comprises an antigenic substance capable of binding immunospecifically to anti-pol, a test medium is formed by contacting a sample of biological material suspected of containing anti-pol with the test reagent, under conditions causing immunological binding of anti-pol to the antigenic substance on the test reagent, and a detectable indicator, such as enzyme-labeled antibody, is provided which interacts immunospecifically with anti-pol, for indicating the occurrence of immunological binding between anti-pol and the antigenic material on the test reagent. Thereafter, the test medium is separated into a portion containing the immunologically bound anti-pol and a portion substantially free thereof, and the presence or absence of the indicator in one of the separated portions of the test medium is detected, as a determinant of whether or not the biological material contains anti-pol.

This method is preferably carried out as a solid-phase, labeled-reagent immunoassay, typical of which is the so-called "sandwich assay" technique. In assays of 5 this type, the target immunological substance is bound to and detected on the solid phase. Of course, other assay procedures or formats may suggest themselves to skilled immunologists.

The antigen component of the test reagent is a 0 polypetide encoded by the hepadavirus genome and responsible for hepatitis virus DNA polymerace activity. Synthetic peptides suitable for use in the method of the invention include

H2N-pro-val-arg-val-his-phe-ala-ser-pro-leu-his- val-ala-trp-arg-pro-pro-cys-COOH (pol 1),

15

H2N-pro-ala-asp-asp-pro-ser-arg-gly-arg-leu-gly-leu- ser-arg-pro-cys-COOH (pol-2), or

H2N-arg-leu-ala-asp-glu-gly-leu-asn-arg-arg-cys-COOH (pol-6).

20 Mixtures of these peptides may be used i-f desired. The frequency of binding of anti-pol has been shown to be particularly high with pol-1, making it the peptide of choice in practicing the invention.

Also within the scope of the invention is the _,_ diagnostic test reagent decribed above, which is easily and inexpensively produced and which enables the rapid determination of anti-pol in biological materials, such as blood or antibodies eluted from tissue. The test reagent of the invention preferably comprises a suitable solid

30 phase carrier to which is affixed one or more of the aforementioned peptides. This test reagent is especially convenient for determination of anti-pol using a sandwich assay, following the EIA technique.

The present invention further provides a diagnostic kit comprising all of the reagents and accessories required for successfully performing the assay. A preferred embodiment of a diagnostic kit according to this invention includes the aforesaid diagnostic test reagent, an enzyme-linked antibody that interacts immunospecifically with anti-pol and a suitable substrate for the enzyme. This enzyme-linked antibody (anti-Ig) has specifities. capable of detecting IgM, IgG, IgA and other major classes of immunoglobulin.

The specific binding of anti-pol in the sera of infected test subjects, both human and animal, to selected polypeptides, as described above, strongly indicates that one or more of the peptides products are made early in the course of infection. Thus, anti-pol appears to have both diagnostic and prognostic value, from the standpoint of its being the earliest detectable marker of viral replication. The presence of anti-pol as a sole marker of HBV infection or in combination with other antibody markers in HBsAg negative individuals may signal the presence of low level viral replication in the liver. The presence of these antibodies as a sole marker may also be of value in detecting infected individuals who are silent carriers and may unknowingly transmit HBV by blood donation. These carriers could be detected by direct testing of blood given to patients having post-transfusion hepatitis of unknown etiology. The presence of these antibodies in organ transplant donors, in individuals on renal dialysis, or in blood from which clotting concentrates are made, may act to significantly reduce the incidence of post-transfusion hepatitis due to HBV, if such sera contains low levels of virus. Further, the appearance of anti-pol in babies born to HBeAg positive mothers, or in individuals unresponsive to the hepatitis B vaccine, may detect underlying infection, which could significantly alter response to vaccination in both cases. It is also believed that determination of anti-pol in accordance with the method of the invention may be useful in screening sera for non-A, non-B hepatitis. Furthermore, the presence of anti-pol may help to identify other HBV-like viruses which have immunologically cross-reacting viral polymerases.

DETAILED DESCRIPTION OF THE INVENTION

The polypeptides used in the practice of this invention are readily prepared by solid phase peptide synthesis procedures well known in the art- Pol-1, pol-2 and pol-6, having the amino acid sequence shown above, were

10 supplied by Johnson & Johnson Biotechnology Center, Inc.,

La Jolla, California. The syntheses of these peptides were based upon polymerase gene sequences from the HBV ayw genome. The complete DNA nucleotide sequences of clones

. m. from HBV WHV, GSHV and DHBV have previously been reported.

Galibert et al., 281 Nature 646-50 (1979). Ono et al. , Nucleic Acids Research 1747-57 (1983). A large open region, spanning about 75% of the hepadnavirus genome, is believed to encode one or more polypeptides responsible for

20 the endogenous DNA po-lyrnerase activity found in intact virions and liver-derived core particles. Developments in Molecular Virology, supra. This activity is known to make the partially double-stranded viral genome fully double stranded. The polymerase product(s) may also be „₅ responsible for reverse transcriptase and/or RNaseH activites, also associated with HBV and related viruses. The size of the polypeptide product, if the entire open region is translated, is about 95,000 daltons, which is similar in size to other DNA polymerases.

30 Choosing the N-terminus of the full length peptide a-3 residue number 1, pol-1 spans residues 822-838, inclusive, pol-2 spans residues 781-795, inclusive, and pol-6 spans residue 29-38, inclusive, as schematically represented below, against the length of HBV DNA in kilobase pairs. Kk

CORE . X . ≤__φ _=} tlllttϊ "$

This representation is provided to show the open reading frames in HBV DNA in relation to one another. The relative positions of the synthetic peptides derived from the polymerase gene are also indicated. The amino acid residues (starting with the N-terminus of the polymerase polypeptide) present in each synthetic peptide are indicated in parantheses below each peptide. The occurrence of sera binding to polymerase peptides derived from both the amino and carboxy ends of the polymerase product(s), as will be described below, suggests that the * entire^' open reading frame of the polymerase gene is ' translated in productive infection.

The aforementioned diagnostic test reagent is prepared by affixing one or more of the above-described peptides to a suitable carrier material, such as polystyrene, nylon, cross-linked dextran or carboxy ethyl cellulose. The peptides may be adsorbed on the carrier material or chemically coupled to the carrier material, if desired, by procedures well known in the art. The carrier material may be in the form of microtiter plates, test tubes, beads, test strips, or any other form which may be utilized in performing an immunoassay.

Another reagent required in carrying out the method of the invention, in addition to the above-described test reagent, is a labelled form of im unoreactive substance which interacts specifically with anti-pol. Antibody which is immunospecific to anti-pol may be linked to any suitable label, including radioactive, fluorescent, electron dense, or enzymatic materials in various ways known in the art. As previously noted, enzymes are the labels of choice in the practice of the invention. It has been found that the interaction between anti-pol and a suitable antibody is readily detected by linking an enzyme label to the antibody. The procedure selected for linking the enzyme to the immunoreactive substance should be one that does not significantly alter the immunological properties of the immunoreactive substance or the activity of the label.

There are numerous bivalent or polyvalent chemical agents available that may be used to couple enzyme to a immunoreactive substance, including glutaraldehyde, difluorodinitrodiphenylsulphone, toluene diisocyanate, and di- and ti-chloro-s-triazine. Specific procedures for coupling immunoreactive substances to enzymes are provided in L. Steinberger, Immunochemistry, (1974). Enzyme may also be linked to the immunoreactive substance by means of the well known biotin-avidin interaction. This is achievable through biotinylation of the immunoreactive substance, which will then bind specifically to avidin, which may be covalently coupled to enzyme using procedures well known in the art.

The selection of a suitable enzyme label for use in the present invention is determined by such properties as specific binding activity and simplicity of determination of the enzyme. The smaller the amount of enzyme-labeled immunoreactive substance needed to produce measurable enzyme activity, the greater the sensitivity of the assay. With respect to simplicity of determination, those enzymes are preferred that can be determined colorimetrically, spectrophotometrically, or fluorimetrically. Represent¬ ative examples of enzymes that may be used in the practice of this invention are horseradish peroxidase glucose oxidase or alkaline phosphatase. Good results have been obtained using horseradish peroxidase as the enzyme label, as will appear in the examples below.

The method of the invention may be performed either qualitatively or quantitatively. In the qualitative tests, using an enzyme label, enzyme activity is preferrably determined by a color reaction, e.g. the production of a colored end product, resulting from incubation of the test reagent with a suitable substrate for the particular enzyme label used. As other possibilities, the enzyme label may generate the substrate for a second enzyme which gives a colored end product, or the enzyme label may convert a pro-enzyme into an enzyme which undergoes a reaction involving a colored compound? or the enzyme label may undergo a reaction in which substrate or end product may be easily stained. Horse radish peroxidase used in conjunction with o-phenylenediamine as the color producing substrate have given satisfactory results in assays on human sera.

For quantitative assays, a standard assay curve must be prepared, in which increasing amounts of anti-pol are plotted against measured enzyme activity. By means of such a standard curve, the quantity of anti-pol in any given test sample may be determined from measured enzyme activity. The means for measuring enzyme activity are well known in the art. See, for example, H. Bergmeyer, Methods of Enzymatic Analysis (1965).

In carrying out the method of the invention, a diagnostic kit may conveniently be used which comprises the above-described test reagent and a labeled form of antibody which interacts immunospecifically with anti-pol. When an enzyme label is employed, the kit will additionally contain a substrate capable of producing a detectable color under the influence of the enzyme label. The diagnostic kit may also contain any accessories, if necessary or desired, for making a dilution series of the test sample for a quantitative determination, such as test tubes, pipettes and flasks containing diluent.

Standard serological tests have been conducted using the method of the invention to determine the presence of anti-pol in the sera of renal dialysis patients and woodchucks acquiring HBV or WHV infection, respectively, in relation to other known markers of hepadnavirus infection.

Hepatitis B virus, which afflicts humans, has been shown to be very similar to ^"the hepatitis virus which infects woodchucks (WHV). Woodchucks infected with WHV may show acute and chronic hepatitis and nearly all woodchucks with chronic hepatitis develop primary hepatocellular carcinoma. This pattern is not unlike what is found in HBV-infected humans in which nodules of primary hepatocellular carcinoma often appear in areas of liver cirrhosis after many years. The molecular biology of HBV and WHV are also very similar. See, for example, Hantz et al., 25(2) Antimicrobial Agents and Chemotherapy, 242-46 (1984). Millman et al. , 4(5) Hepatology, 817-24 (1984)', and references cited therein. Thus, the woodchuck model is very useful for evaluating the efficacy of diagnostic tests for hepatitis in humans.

It is noted that the peptides used for these serological tests were chosen from hydrophilic regions of the HBV polymerase gene product which share more than 50% primary sequence homology with the analogous regions of the WHV polymerase product. This was done to increase the chances that cross-reactive anti-pol would be detected in woodchucks using HBV-derived peptides. The human sera used for the screening tests were obtained from renal dialysis patients while they were undergoing clinical treatment. The woodchuck sera was collected from 55 woodchucks trapped in the wild, of which 41 were naturally infected with WHV, 8 WHV negative animals were experimentally infected oy intrahepatic (I.H.) injection of WHV and 13 WHV negative animals were experimentally infected by I.H. injection of covalently closed circular (ccc) WHV DNA.

The WHV particles used for experimental infection were prepared from infectious sera by determining the endpoint titer of surface antigen in the Ausria II assay (Abbott Laboratories, North Chicago, IL) and then injecting 1 ml of 5 times the concentration of this dilution. Prior to titering, virus particles were pelleted by ultra- centrifugation and resuspended in Hanks balanced salt solution. Virus samples were millipore filtered (0.22 m Millipore) to remove bacteria and injected I.H. into WHV negative woodchucks anesthesized with a mixture of ketamine (1.84 ml/animal, Vetalar) and Rompun (0.16 ml/animal. Miles Labs).

In preparing ccc DNA for experimental infection, Escherichia coli (HB101) carrying a single full-length .Eco RI insert of WHV DNA in pBR322 were grown and the recombinant plasmid recovered exactly as described by Mukhopadhyay et al . , 133 Analytic Biochemistry, 265-70 (1983). The plasmid was checked for size and purity by agarose gel electrophoresis according to the procedure of Feitelson et al., 83 Proc. Natl. Acad, Sci. USA, 2233-37 (1986), and the presence of the WHV insert checked for by Eco RI digestion prior to electrophoresis on the same gel. Preparations of Eco RI digested plasmid were phenol extracted, ethanol precipitated and subjected to ligation by addition of T4 ligase (New England Biolabs, Beverly, MA) according to the procedure of Hirschman et al., 77 Proc. Natl. Acad. Sci. USA, 5507-11 (1980). The formation of ccc WHV DNA was monitored by gel electrophoresis. Approx¬ imately 100 g of DNA was injected I.H. into each animal.

Screening of human sera was also performed on samples obtained from 41 test subjects from the Senegalese village complex of Tip, 34 test subjects from several Korean villages, 150 test subjects from scattered locations in Japan, 132 test subjects from Great Britain having documented cases of biliary cirrh'osis, 156 test subjects from several isolated Icelandic communities (a gift from Dr. Olafur Jensson, Icelandic Red Cross) and 14 members of the staff at Fox Chase Cancer Center. The Senegalese and Korean test subjects included individuals who were HBeAg positive and others who were anti-HBe positive in groups having the diagnosis of the PHC chronic hepatitis and silent carriers (without pathology). A number of the

Senegalese and Korean test subjects without any markers of HBV infection were also used in these tests. A substantial majority of the Japanese test subjects were anti-HBc positive. All had histologically confirmed hepatitis and were negative for all other HBV markers. The biliary cirrhosis patients were included so as to provide a population having a liver disease not due to HBV infection, in order to determine if anti-pol comes about as. a consequence, of liver disease, and is not specific to HBV infection. The patients tested had a low incidence of HBV markers. The sera from the Fox Chase Cancer Center and Icelandic test subjects were chosen because they represented groups with a low frequency of HBV markers. The serological tests were performed using HBV reagents obtained from Abbott Labs, North Chicago, IL, as generally described by Millman et al., 35 Infec. Immun. 752-57 (1982). see also, Millman et al. Hepatology, supra. Human sera were tested for HBsAg, HBeAg, anti-HBs, anti-HBe and anti-HBe. In screening woodchuck sera, crossreactive anti-WHc and anti-WHe were determined by the appropriate HBV assays, while WHsAg and anti-WHs were detected with specific reagents by both immonodiffusion and EIA assays. All sera were then assayed for anti-pol, as described in detail in the examples below, which are intended to illustrate and not to limit the invention. The general procedure used to detect anti-pol in human and woodchuck sera entailed coating microtiter wells with the synthetic peptides derived from the polymerase gene, followed by the addition of human or woodchuck sera to be tested, and finally by addition of an enzyme-linked anti-Ig and substrate for color .development. The sera were maintained at -30°C prior to assay.

Alternative screening procedures were used, depending on whether serial sera samples, as in the assays of sera from renal dialysis patients, or single sera samples were being tested. The procedure used on the single serum samples would be the one most often used in clinical applications.

Details of the screening tests are described below. Earlier experiments aimed at determining the binding specificity of serial sera from dialysis patients or woodchucks showed that all of the sera from some test subjects resulted in consistently higher optical density (O.D.) values than all of the sera from other test subjects. In some of these individuals there was little difference in the O.D. values between the first (or earliest) serum and successive sera tested. Further experiments showed that blocking the binding of sera to assay wells by preincubation with synthetic peptides was only successful if the O.D. values of successive sera were higher than in the corresponding first serum tested in each case. Successful blocking reduced the level of binding in serial sera to that of the first serum tested for each individual. In these experiments the O.D. value of the first serum from each individual was not reduced upon preincubation with the appropriate synthetic peptides. When serial sera were tested in wells coated with a phosphate buffered saline fetal calf serum (PBS/FCS) solution in the absence of peptides, the O.D. value of the ^"first serum collected from each individual was similar to those values in successive sera from the same individual. These results suggested that subtraction of the O.D. value of the first serum tested from the values of the successive sera in a single individual would be the best way to eliminate the contribution of nonspecific binding in each case. By setting the cutoff O.D. value for a positive score as two standard deviations above the mean value of several negative controls, which included everything except test sera, atr least 95% of the test sera which was determined to be positive could not be accounted for by random fluctuations in the assay.

Recognizing that normal human or woodchuck sera demonstrate some measure of non-specific binding compared to wells without such sera, suggested that in assays where serial sera were not available, the cutoff O.D. value for scoring positive samples should be based upon normal sera as negative controls. Single sera scored positive by this assay were successfully blocked by preincubation with synthetic peptides, confirming that the' positive sera specifically bound the peptides. Blocking was unsuccessful with anti-pol negative sera. When selected individual sera from dialysis patients scoring positive or negative in assays containing serial serum samples were retested in the above assay without the first serum available, the same results were obtained, demonstrating that each of the alternative procedures described above produce substantially equivalent results.

The above-described peptides were prepared for generation of antisera by coupling through their free cysteine sulfhydryl (either^'in the peptide sequence or added to the carboxy terminus) to keyhole limpet hemocyanin (Sigma, St. Louis, MO) using the coupling agent m-maleimidobenzyol-N-hydroxysuccinimide ester (Pierce Chemical Co., Rockford, IL) according to the procedure described by Liu et al . , 18 Biochemistry 690-97 (1979). For antibody production, 5- to 10-week-old female New Zealand White rabbits (2 animals/peptide Hazelton) were bled and then injected with peptide conjugate as described by Bittle et al., 298 Nature 30-33 (1982). Dilutions of immune sera were assayed in parallel with preimmune sera in solid phase assays in wells (Immunolon 2 Removawell Strips, Dynatech Labs, Alexandria, VA) coated with the appropriate (unconjugated) synthetic peptide. Rabbit sera containing peptide antibodies were used for positive controls in the serological assays on human and woodchuck sera described herein.

Example 1 - Assay of Serial Sera for Anti-Pol a) Human Sera

To specifically assay serial sera, a mixture of the three synthetic peptides was used to coat microtiter wells (Dynatech, Immunolon 2) overnight at 4°C. Each well contained 1 M. g of each synthetic peptide. diluted in 50 M.1 of phosphate buffered saline (Fisher B-82, King of Prussia, PA) contaning 0.9% sodium chloride (Baker) and 10% fetal calf serum (FCS) (GIBCO, Grand Island, NY). Following overnight incubation, the wells were washed at least six times with PBS using a Nunc Immuno Wash 120. The serial dilutions of test sera (50M.1/well at 1:10 dilution in PBS/FCS ) were then added to the wells, and the plates incubated overnight at 4°C. The next day the wells were washed six times with PBS. Affinity purified horseradish peroxidase (HRP) conjugated goat anti-human Ig (50 1/well at a dilution of 1:100 in PBS/FCS; Cooper Biomedical, Malvern, PA) was added to each well. The plates were incubated 1 hour at 37°C, washed six times with PBS, and binding determined colorimetrically with o-phenylene- diamine(OPD) (Abbott Labs) at 492 nm. b) Woodchuck Sera

For woodchuck sera, the assay used for the detection of anti-pol was similar to that described above for assaying human sera, except that two different antibodies were sequentially added following the binding of woodchuck antibodies to the -synthetic peptides. The first antibody was rabbit anti-woodchuck sera preabsorbed for 1 hour at 37°C with normal woodchuck serum minus gamma globulins. An equal volume of saturated ammonium sulfate ((NH 2 SO4) solution, pH adjusted to 6.5 with 0.IN NaOH, was added to normal (without WHV serological markers) woodchuck serum. The precipatate was removed by centrifugation (5000 rpm) for 1 hour at 5°C. The crystal clear supermatant was dialyzed against PBS for 3 days with several charges of PBS. The clear dialyzed solution was concentrated to the original volume of woodchuck serum. Several techniques have been used for concentration. The "Centriprep" 30 made by Amicon Corp., Lexington, Mass is preferred. Care must be taken that the method of concentration does not increase the salt concentration (PBS) to greater than 0.15M. Lyophilization will tend to increase salt concentration and may necessitate further dialysis. The second antibody was affinity purified HRP conjugated goat anti-rabbit Ig, detecting IgM, IgG, IgA and other major Ig classes, (Cooper Biomedical) preabsorbed 1 hour with normal woodchuck serum.

In all of the assays of the serial sera from renal dialysis patients and woodchucks, the optical density

(O.D.) value of the first serum collected from a given test subject (which contained no markers of HBV or WHV infection) was subtracted from the successive serial bleedings from the same test subjecst. The remaining O.D. readings (that is, the difference in O.D.) were scored as positive if their values were greater than two standard deviations above the mean of the negative controls. Negative controls were defined as wells receiving all the reagents in the assay except the test samples of woodchuck or human sera. The results of the serological assays described in this example are set forth in Table I. The presence of anti-pol was reported as positive if it was reproducibly present in one or more serial sera collected from the infected test subject.

Table I

Presence and first appearance of polymerase antibodies (anti-pol) in serial sera of woodchucks and renal dialysis patients

a: As determined by occurrence of one or more markers of infection prior to assay for anti b: Reported values represent percentage of those anti-pol positive test subject that exhib anti-pol as the first marker.

As the data in Table I indicate, 21 of the 26 naturally infected woodchucks (81%) developed anti-pol in ^'one or more serial bleedings. Simliar high frequencies of antibody were observed in renal dialysis patients becoming naturally infected with HBV and in woodchucks injected I.H. with ccc WHV DNA. These results demonstrate that anti-pol is a common marker of hepadnavirus infection.

Data was also collected from the assays of the serial sera regarding the relationship of anti-pol to other markers of hepatitis infection. Two-by-two comparisons were made from the sum of test subjects with each virus marker in at least one serial serum sample obtained during the course_.of infection. Here again, a positive result was recorded for a given marker when that marker was found in one or more serial sera. In assaying the sera of the experimentally infected woodchucks, the appearance of markers of infection, including anti-pol, were determined on the basis of their time of appearance after infection. These data, which appear in Table II, are further summarized below.

Among the 43 WHsAg positive woodchucks, 38 tested anti-pol positive. Of these, 23, or more than 60%, first demonstrated such anti-pol prior to the appearance of WHsAg. Similarly, among the 18 HBsAg positive renal dialysis patients, 14 tested anti-pol positive. Of these, 11, or approximately 79% first exhibited anti-pol prior to the appearance of HBsAg. Forty-eight percent of the renal dialysis patients had anti-pol as the first marker. In the remaining WHsAg and anti-pol positive woodchucks, anti-pol appeared at the same time as WHsAb in 6 cases, and after the appearance of surface antigen in 9 cases. Three renal dialysis patients also exhibited the appearance of anti-pol after HBsAg, but none showed the simultaneous appearance of these two markers. Among surface antigen negative test subjects demonstrating antibody markers of infection (anti-core, anti-e, anti-S), anti-pol was present in 10 of 12 woodchucks (83%) and 9 of 14 dialysis patients (64%), but in each case the appearance of anti-pol followed that of one or more other antibody markers.

These results indicate that anti-pol often constitutes an early marker, if not the first marker, of infection when surface antigen is present, and a late marker in the absence of surface antigen. These assays also revealed anti-pol as the sole marker of infection in 7 of 8 woodchucks and 4 of 8 renal dialysis patients, negative for all other markers of hepadnavirus infection, suggesting that such individuals were infected and that the anti-pol assay detected cases of HBV infection which could not be identified with other available serological assays of hepatitis B.

Table II

Relationship of anti-pol to other markers of HBV or WHV infections in assays on serial sera

HBsAg anti-HBs anti-HBc HBeAg anti-HBe. SGPT peak

+ - + - + - + - + - + -_ anti- +. 15 9 6 18 20 3 14 10 11 13 16 8 pol - 3 5 1 7 4 4 -3 5 2 6 2 6 p=0.2051(d) p=0.4227 p=0.0528 p=0.2699 p=0.2707 p-0.0354

WHsAg anti-WHs anti-WHc anti-WHe pathology(e) + - + - + - + - + - anti- + 38 5 7 34 38 10 33 15 14 4 pol - 10 2 3 3 4 1 5 2 6 2 p=-0.4807 p=0.1007 p=0.7255 p=0.6299 p=0.6216

(c) SGPT= Serum gluta ic pyruvic transaminase

(d) p values were calculated using the Fishers' exact test for independence in a 2 x 2 table. The relationship of two characteristics was considered significant when P .0.05.

(e) Woodchucks were evaluated for the presence of hepatitis by periodic liver biopsies, and at autopsy at the completion of the test.

As reflected in Table II, sera from renal dialysis patients were also tested for SGPT. Patients testing positive for "SGPT peak" demonstrated significantly elevanted SGPT values in one or more of their serial sera. The SGPT peaks usually occurred within several weeks after the appearance of HBsAg in the serum.

It is important to note that in HBV infection, there was a significant correlation between the appearance of anti-pol and SGPT, even though anti-pol did not correlate well with other serological markers of HBV and WHV infection, nor did anti-pol correlate with the appearance of chronic hepatitis in woodchucks. Since acute hepatitis is often characterized by the appearance of HBsAg followed by a peak of SGPT activity, and elevated SGPT is a reflection of recent liver damage, the appearance of anti-pol as a first marker of infection may be important in predicting liver damage, at least in acute HBV infection.

Example 2 - Determination of Frequency of

Binding of Anti-pol to Different Combinations of Polypeptides

Sera which scored positive for anti-pol with a mixture of the three synthetic peptides were retested with individual peptides to determine the amount of binding to each peptide. These assays differed from those described above only in that a single peptide was used for coating assay wells (1 < g/well ) instead of a mixture.

Table III shows the number of test subjects and their frequency of binding to the individual polypeptides.

Table I I I

Frequency of binding to different combinations of polymerase peptides

woodchucks patients anti-pol (n = 48) (n = 23) -1 -2 -6

31 (65%) 8 (35%) +

6 (12%) 2 (9%) +

6 (12%) 3 (13%) +

3 (6%) 5 (22%)

2 (4%) 2 (9%)

0 (0%) 3 (13%) +

The most frequent anti-pol specificity detected was to the peptide pol-1. This peptide was generated from the carboxy terminus of the polymerase product schematically represented above. In 65% of woodchucks and 35% of renal dialysis patients, their anti-pol response was exclusively directed toward pol-1 in these assays. Although antibodies were generated in natural and experimental infections to the other polymerase peptides, the frequency of their appearance in different hosts was relatively low compared to pol-1, suggesting pol-2 and pol-6 are less immunogenic. Similarly, in rabbits immunized with synthetic peptide conjugates, pol-1 also elicited the highest antibody titers. Considering the sum total of infected test subjects responding to each polypeptide, 96% (46/48) of the woodchucks and 78% (18/33) of the dialysis patients had antibody to pol-1, indicating that pol-1 is immunodominant. Example 3 - Assays of Single Sera for Anti-Pol Single serum samples from test subjects in different populations were also tested for anti-pol using the sequence of reagents and steps outlined above. Instead of subtracting the optical density values of the appropriate first serum sample in each case, however, positive values were scored as being greater than two standard deviations above the mean of the negative controls. Negative controls were defined as wells containing different human sera negative for serological markers of HBV infection. To determine the specificity of binding, selected anti-pol positive human or w^'oodchuck sera were preincubated for 1 hour at 37°C with a mixture containing 25 H g of each synthetic peptide, and then assayed. Successful blocking was demonstrated by a minimum reduction in signal of 50%. The assays of anti-pol in sera collected as single samples from the different populations- of human test subjects were conducted after testing for other markers of HBV infection. Screening of the single serum samples for the various other markers was conducted in the manner previously described. The results are shown in Table IV.

TABLE IV

Frequency of anti-pol in selected populations no. with anti-pol no. HBV other positive anti-pol Population tested markers no. % only

With high incidence of HBV infection

With low incidence of HBV infection

Icelandic Biliary

.cirrhosis (U.S. ) FCCC staff

(U.S. ) NIH NANB panel

normal controls

With NANB hepatitis

FCCC NANB Japanese NANB NIH NANB

panel II

a: All of these patients had only anti-HBc prior to testing.

As the data in Table IV indicate, populations with a high incidence of HBV infection, as indicated by a high frequency of HBV markers in their sera, also have a high frequency of anti-pol. Anti-pol was present at high frequencies in both HBeAg and anti-HBe positive patients and varied little in incidence among groups with primary hepatocellular carcinoma, chronic hepatitis and asymptomatic carrier state. For comparison, the frequency of antibodies in the renal dialysis patients, having a high frequency of other markers, is presented. The frequency of anti-pol among the renal dialysis patients is very high, in part, because serial sera were available from each of the 32 patients. In the other populations of this group the incidence varied from 10 to 51%. The factors responsible for this variation have not been elucidated. The data, however, are consistent with the conclusion that anti-pol is a sensitive marker of HBV infection. Populations with low incidence of HBV infection also have a correspondingly low incidence of anti-pol. Tnis is true in "normal" individuals without evidence of HBV infection or liver disease (i.e. in the Icelandic population, FCCC staff, and normal controls in the NIH NANB panel), as well as in patients with chronic liver disease unrelated to HBV infection (i.e., in biliary cirrhosis patients). The data are consistent with the conclusion • that anti-pol is not found in different populations which are not infected with HBV. Further, anti-pol is not a response to chronic liver disease, since the frequency of anti-pol in biliary cirrhosis patients is not different ^'from that in populations without liver disease.

A few individuals from different populations nave anti-pol in the absence of other markers of HBV infection. This finding is consistent with the conclusion that either these patients are infected with HBV and/or another virus related to HBV. At lea.st one non-A, non-B (NANB) hepatitis agent has characteristics which overlap with HBV. Among these characteristics is the finding that a large proportion of the sera from patients presumably infected with an HBV-like NANB agent have anti-HBc. Since anti-HBc is directed to the nucleocapsid antigen of HBV, the presence of anti-HBc in some NANB sera may indicate that the NANB agent shares one or more nucleocapsid determinants with HBV. The presence of the HBV polymerase within the nucleocapsid is consistent with the conclusion that a similar cross-reactive polymerase is present within the nucleocapsid of an HBV-like NANB agent. When sera from documented cases of NANB patients were tested for anti-pol, the antibody frequencies were higher than those obtained from populations having low incidence of HBV infection, but lower than those obtained from populations having a high incidence of HBV infection. These date tend to suggest, therefore, that anti-pol may be useful in screening sera for one or more NANB agents.

In several cases anti-pol was found to be present in the sera of "normal" test subjects as the only marker of infection, suggesting that the anti-pol assay could identify infected individuals negative for all other markers.

In summary, antibodies against synthetic peptides derived from the polymerase gene of the hepatitis B virus (HBV) were present in 80% of renal dialysis patients infected with HBV and in woodchucks infected with woodchuck hepatitis virus. The polymerase antibody or anti-pol appeared as the earliest marker of both HBV and WHV infections in approximately half of the subjects tested, suggesting that these antibodies were generated following early viral replication in the liver during the incubation period prior to the appearance of virus in the blood. In several cases, anti-pol was the sole marker of infection, indicating that underlying infection and low levels of virus replication were present.

From the foregoing detailed description, those skilled in the art will appreciate that the method of the invention provides an effective way of determining hepatitis infection at an early stage and may serve to materially reduce the incidence of post-transfusion hepatitis due to HBV in cases of sera containing low level of virus.

While certain preferred embodiments of the present invention have been described above, the invention may be practiced other than as specifically set forth in the foregoing description. For example, the peptide specificities described above may also be made by expressing the polypeptide sequences using recombinant DNA technology, encompassing the polymerase gene of HBV, or other related virus. K. Streimer, et al . , 58 J. Virology, (No. 1) 9-16 (1986). Accordingly, it is not intended to limit the invention to the preferred embodiments' described herein, but various modifications may be made therein and thereto, without departing from the scope and spirit of the present invention, as set forth in the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method for determining antibody to hepatitis virus DNA polymerase (anti-pol) in a biological material, which comprises: providing a test reagent comprising an antigenic substance capable of binding immuno- specifically to anti-pol; forming a test medium by contacting a sample of said biological material suspected of containing anti-pol with said test reagent under conditions causing immunological binding of said anti-pol to the antigenic substance on said reagent; providing a detectable indicator in said test medium for indicating the occurrence of said immunological binding; separating said test medium into a portion containing the immunologically bound anti-pol, and a portion substantially free of said immunologically bound anti-pol; and detecting the presence or absence of said indicator in one of said separated portions of said test medium, as a determinant of whether or not said biological material contains anti-pol.

2. A method as claimed in claim 1, wherein a test reagent is provided which comprises, as the antigenic substance, a polypeptide. encoded by the hepadnavirus genome and responsible for hepatitis virus DNA polymerase activity.

3. A method as claimed in claim 2, wherein a test reagent is provided which comprises, as the antigenic material, a synthetic peptide from the group of H2N-pro-val-arg-val-his-phe-ala-se -pro-leu— his- val-ala-trp-arg-pro-pro-cys-COOH ,

H 2 -pro -a la -asp-asp-pro -ser-arg -gly-arg -leu-gly-leu - ser-arg-pro-cys-COOH , or

H2N-arg-leu-ala-asp-glu-gly-leu-asn-arg-arg-cys-COOHj or a combination of two or more of said peptides.

4. A method as claimed in claim 2, wherein a test reagent is provided which comprises, as the antigenic material,

H2N—pro-val-arg-val-his-phe-ala-ser-pro-leu-his- val-ala-trp-arg-pro-pro-cys-COOH.

5. A method as claimed in claim 2, wherein a test reagent is provided which comprises said antigenic substance affixed to a solid phase, and the step of separating comprises removing said solid phase from the remainder of said test medium.

6. A method as claimed in claim 5 wherein a test reagent is provided comprising, as the antigenic material, a synthetic peptide selected from the group of

H2N-pro-val-arg-val-his-phe-ala-ser-pro-leu-his- val-ala-trp-arg-pro-pro-cys-COOH,

H2N-pro-ala-asp-asp-pro-ser-arg-gly-arg-leu-gly-leu- ser-arg-pro-cys-COOH, or

H2N-arg-leu-ala-asp-glu-gly-leu-asn-arg-arg-cys-C00H. or a combination of two or more of said peptides.

7. A method as claimed in claim 5 wherein a test reagent is provided comprising, as the antigenic material,

H2N-pro-val-arg-val-his-phe-ala-ser-pro-leu-his- val-ala-trp-arg-pro-pro-cys-COOH.

8. A method as claimed in claim 6 wherein a detectable indicator is provided which comprises a substance capable of immunospecific interaction with said anti-pol on said test reagent, whereby said detectable indicator substance is bound to said test reagent.

9. A method as claimed in claim 8 wherein said detectable indicator is provided comprising an antibody to anti-pol linked with a detectable label.

10. A method as claimed in claim 8, wherein said detectable indicator is provided comprising an antibody linked with an enzyme capable of producing a detectable color in the presence of a suitable substrate.

11. A method as claimed in claim 10, wherein the step of detecting the presence or absence of said indicator comprises detecting the activity of said enzyme by contacting the test reagent with a substrate producing a detectable color under the influence of said enzyme, under conditions causing any enzyme present on said test reagent to produce said color and determining the appearance or relative intensity of said color.

12. A method for determining anti-pol in a biological fluid suspected of containing same, utilizing direct enzyme immunoassay, said method comprising:

(a) contacting a test sample of said biological fluid with a test reagent having the peptide ,

H2N-pro-val-arg-val-his-phe-ala-ser-pro-leu-his- val-ala-trp-arg-pro-pro-cys-COOH (pol-1), affixed thereto

(b) incubating said test sample while in contact with said test reagent to bind any anti-pol present in said test sample to said pol-1 on said test reagent;

(c) washing said incubated test reagent to remove any interfering substances, including unbound anti-pol;

(d) contacting said washed test reagent with an enzyme-linked antibody which interacts immunospecifically with anti-pol;

(e) incubating said test reagent while in contact with said enzyme-linked antibody so as to bind said enzyme-linked antibody to any anti-pol present on said test reagent;

(f) washing said test reagent to remove any unbound enzyme-linked antibody; and

(g) determining the activity of enzyme bound to said test reagent, as a measure of the presence or quantit ^"of anti-pol in said biological fluid.

13. A method as claimed in claim 12 wherein the step of determining enzyme activity comprises contacting the test reagent with a substrate producing a detectable color under the influence of said enzyme, under conditions causing any enzyme present on said test reagent to produce said color, and determining the appearance or relative intensity of said color.

14. A method as claimed in claim 12 wherein the antibody to which the enzyme is linked is a purified form of anti-human Ig.

15. A method as claimed in claim 13 wherein said enzyme is horseradish peroxidase and said substrate is o-phenylenediamine .

16. A test reagent for the determination of anti-pol, said reagent comprising a solid phase to 5 which is affixed a peptide derivable from a nucleotide sequence of the hepadavirus genome which is responsible for the hepatitis virus DNA polymerase activity of said hepadavius.

10 17. A test reagent as claimed in claim 16, wherein said peptide is selected from the group of

H2N-pro-ala-asp-asp-pro-ser-arg-gly-arg-leu-gly-leu- - - ser-arg-pro-cys-COOH, or

H2N-arg-leu-ala-asp-glu-gly-leu-asn-arg-arg-cys-COOH or a combination of two or more of said polypeptides.

20 18. A diagnostic kit for the determination of anti-pol in a biological material suspected of containing anti-pol, said kit including:

(a) a test reagent according to claim 13 and

(b) antibody which interacts 5 immunospecifically with anti-pol, said antibody being linked with a detectable label.

19. A diagnostic kit according to claim 18, wherein the labeled antibody component of said kit 0 comprises an enzyme capable of producing a detectable color in the presence of a suitable substrate, and said kit further includes a substrate suitable for producing a dectable color under the influence of said enzyme.

20. A diagnostic kit for the determination of anti-pol in a biological material suspected of containing anti-pol, said kit including:

(a) a test reagent comprising the peptide

H2N-pro-val-arg-val-his-phe-ala-ser-pro-leu-his- val-ala-trp-arg-pro-pro-cys-COOH affixed to a solid phase;

(b) horseradish peroxidase-linked antibody which interacts immunospecifically with anti-pol that is immunologically bound to said test reagent; and

(c) o-phenylenediamine in an amount sufficient for color detection under the influence of said horseradish peroxidase.