Enzyme Immunoassay for Anti-Hepatitis B Core (HBc) Immunoglobulin Gi and Significance of Low-Level Results in Competitive Assays for Anti-HBc

Transcription

1 JOURNAL OF CLINICAL MICROBIOLOGY, May 1989, p /89/ $02.00/0 Copyright (O 1989, American Society for Microbiology Vol. 27, No. 5 Enzyme Immunoassay for Anti-Hepatitis B Core (HBc) Immunoglobulin Gi and Significance of Low-Level Results in Competitive Assays for Anti-HBc MATTI SALLBERG AND LARS O. MAGNIUS* Department of Virology, The National Bacteriological Laboratory, S Stockholm, Swveden Received 23 August 1988/Accepted 11 January 1989 An enzyme immunoassay (EIA) for anti-hepatitis B core (HBc) immunoglobulin Gi (IgGl) was compared with a commercial radioimmunoassay (RIA) for anti-hbc antibody (Corab; Abbott Laboratories, North Chicago, 111.). In parallel tests of 445 consecutive samples, discrepant results were obtained with 2 samples, 1 of which was positive only by the RIA and the other of which was positive only by the EIA for anti-hbc IgG1. In tests of another 192 samples with low blocking activity in the RIA (inhibition range, 90 to 30%), 10 samples gave discrepant results, 5 of which were positive only by the RIA and the other 5 of which were positive only by the EIA for anti-hbc IgGl. Of 12 samples with discrepant results, 11 samples were tested further for anti-hbc IgG3, IgM, and IgAl by the EIA. Of these, seven samples were positive for anti-hbc IgGl, anti-hbc IgG3, or both. All seven samples were also positive for anti-hepatitis B surface (HBs) antigen. Three samples were negative for anti-hbc IgGl, anti-hbc IgG3, or both but were positive for anti-hbc IgM, anti-hbc IgAl, or both; and one sample was reactive only in the RIA. These four samples were all negative for anti-hbs. Thus, low-level results in the RIA caused by anti-hbc IgM, anti-hbc IgA, or both reflect the unspecific activation of immature B lymphocytes that is not related to previous exposure to hepatitis B virus (HBV). In contrast, the presence of anti-hbc IgGl, anti-hbc IgG3, or both indicates differentiated anti-hbc IgG-producing plasma cells and previous exposure to HBV, as was also shown by the presence of anti-hbs. On class and subclass determination for confirmation of positivity for anti-hbc in 19 serum samples, which was identified by screening of blood from 1,343 donors by a competitive EIA (Hepanostika; Organon), 9 samples with positive results, all low level, did not indicate previous exposure to HBV. It was concluded that determination of classes and subclasses of anti-hbc provides a tool for discriminating positive anti-hbc results not caused by HBV exposure. The specificity of low-level results obtained by commercially available competitive assays for anti-hepatitis B core (HBc) antigen is a problem recognized by those engaged in the laboratory diagnosis of hepatitis B. However, this matter has been the subject of only a few reports. Studies approaching this problem have either been epidemiological (5, 25) or have aimed at demonstrating previous exposure to hepatitis B virus (HBV) by assessing a booster anti-hepatitis B surface (HBs) antigen response after vaccination against HBV (4; F. Bonino, G. Poli, A. Ponzetto, G. C. Acitis, M. Rizetto, and G. Verme, Hepatology 5:1041, 1984 [abstract]). Both approaches have yielded results that render the specificity of some low-level results doubtful. There have been no previous attempts to determine to which immunoglobulin class and subclass low-level anti-hbc belongs. Immunoglobulin Gi (IgGl) is the dominating subclass of IgG in viral infections (12, 13, 16, 20, 21, 23). Since this also holds true for anti-hbc in chronic HBV infections (18) we compared here an enzyme immunoassay (EIA) for anti-hbc IgG1 with a competitive assay for anti-hbc. We also investigated the possibility of using immunoglobulin class- and subclassspecific EIA procedures to confirm results obtained by competitive assays for anti-hbc. MATERIALS AND METHODS Specimens. The following five groups of sera were investigated. (i) Seven serum samples that were negative for HBs antigen (Ausria; Abbott Laboratories, North Chicago, 111.), * Corresponding author. positive for anti-hbs antigen (Ausab; Abbott), and positive for anti-hbc antigen by the radioimmunoassay (RIA; Corab; Abbott) in various titers were used for comparing the endpoint titers obtained by the two assays. (ii) A total of 445 serum samples sent consecutively to the Department of Virology of the National Bacteriological Laboratory (Stockholm, Sweden) during a period in September 1987 were tested in parallel for anti-hbc by the RIA (Corab; Abbott) and for anti-hbc IgGl by the EIA. (iii) A total of 192 serum samples sampled from January to June 1987, giving low-level blocking reactivity by the RIA with an inhibition range of 90 to 30%, were tested for anti-hbc IgGl. (iv) A total of 19 serum samples were kindly provided by Antero Danersund (Blood Center, Akademiska Sjukhuset, Uppsala, Sweden). These sera were repeatedly found to be positive for anti-hbc in tests of blood from 1,343 donors at the Blood Center by the EIA (Hepanostika Uni-Form; Organon, Oss, The Netherlands) from December 1987 to January (v) Seven serum samples which, when tested by EIA (Corzyme; Abbott) and RIA (Corab; Abbott) at different blood centers in central Sweden, gave low-level anti-hbc results and were sent to the Department of Virology for confirmation were used. These five groups of sera are referred to as the titrated sera, the consecutive sera, the low-level sera, the blood donor sera, and the reference sera, respectively. Methods. An EIA was developed for the determination of anti-hbc of different class and subclass specificities. As a positive control, an anti-hbc IgG-positive serum sample diluted 1:100 in phosphate-buffered saline with 0.05% Tween 20 and 1% bovine serum albumin was used. For the testing of 849

2 850 SALLBERG AND MAGNIUS J. CLIN. MICROIBIOL. TABLE 1. Optical densities obtained by EIA for the indicated classes and subclasses of anti-hbc with negative sera Test no. No. of Mean (SD) OD" of anti-hbc: *negative sera IgG1 IgG3 IgM IgAl (0.008) (0.010) (0.029) (0.015) (0.019) (0.007) (0.043) (0.010) (0.010) (0.007) (0.040) (0.012) (0.027) (0.035) (0.033) (0.007) OD, Optical density. anti-hbc IgGl and IgG3, the samples were diluted 1:100, and for the testing of anti-hbc IgM and IgAl, they were diluted 1:1,000. Serum samples negative for all HBV markers were used as negative controls. All four different EIAs were performed by the following procedure. One hundred microliters of each sample dilution was added in duplicate to a microtiter plate coated with recombinant HBc antigen (Enzygnost anti-hbc; Behringwerke AG, Marburg, Federal Republic of Germany). The plate was then left to incubate at 4 C overnight. After the plates were washed five times in phosphate-buffered saline with 0.05% Tween 20, l of ascitic fluid containing monoclonal class- and subclass-specific antibodies was added to each well, and the plate was incubated at 37 C for 90 mins. The monoclonal antibodies, which were purchased from Seward Laboratory (London, United Kingdom), were as follows: IgGl, clone NL16; IgG3, clone ZG4; IgM, clone AF6; IgAl, clone N1F2. These were diluted 1:5,000, 1:5,000, 1:32,000, and 1:2,000, respectively. After the plates were washed, 100,ul of biotinylated rabbit anti-mouse IgGl (Zymed Laboratories Inc., San Francisco, Calif.) diluted 1:6,000 in phosphate-buffered saline with 0.05% Tween 20 with 1% bovine serum albumin was added, and the plate was incubated at 37 C for 90 min. After the plate was washed, 100 Ftl of horseradish peroxidase-conjugated streptavidin (Zymed Laboratories) diluted 1:6,000 in phosphate-buffered saline with 0.05% Tween 20 with 1% bovine serum albumin was added to each well, and the plates were incubated at air.bient temperature for 30 min. One hundred microliters of ortho-phenylenediamine was added to each well after the removal of excess horseradish peroxidase-conjugated streptavidin, and the plate was left to incubate at room temperature for 30 min. The enzyme reaction was terminated with 100,ul of 0.5 M H2SO4, and the plate was read with a double-beam spectrophotometer (SLT 210.1; Kontron) at 492 nm. HBs antigen (14) and anti-hbs (10), anti-hbc (17), and anti-hbc (7) IgM were determined by commercial assays (Abbott) as described previously. HBV DNA in serum was assayed by a dot blot procedure as described by Norder et al. (H. Norder, C. Brattstrom, and L. O. Magnius, J. Med. Virol., in press). RESULTS The performance of the EIA for anti-hbc IgGl, IgG3, IgAl, and IgM on the negative sera used for cutoff determinations is given in Table 1. The relative sensitivity of the EIA for anti-hbc IgGl was 2 to 17 times higher than that of the RIA at the endpoint determination on the seven titrated serum samples (Table 2). On testing of the 445 consecutive serum specimens by the RIA and for anti-hbc IgGl, 74 were positive and 369 were negative by both assays. Two serum samples gave discordant results. One of the discordant serum samples was positive only by the RIA and one was positive only for anti-hbc IgGl. The sample that was positive only by the RIA was found to be positive for anti-hbc IgM and IgAl (Table 3). The sample that was positive only for anti-hbc IgGl became exhausted and could not be tested further. When the 192 low-level serum samples were tested by the RIA and for anti-hbc IgGl, 151 were positive and 31 were negative by both assays and 10 gave discordant results. Of five discordant samples that were positive only by the RIA, two had detectable anti-hbc IgM, anti-hbc IgAl, or both (Table 3). Neither of these two samples had detectable anti-hbs. One sample, which gave 72% inhibition in the RIA, was negative in all confirmatory assays, including that for anti-hbs (Table 3). The remaining two samples had detectable anti-hbc IgG3 in combination with anti-hbs (Table 3). All five serum samples that were positive only for anti-hbc IgGl were positive for anti-hbs (Table 3). Results obtained by the RIA for these five serum samples indicated that there was from 46 to 31% inhibition. Of the 19 blood donor serum samples tested, 16 were found to be positive by the RIA in our laboratory. Of these 19 serum samples, 10, all of which were positive by the RIA, were positive for anti-hbc IgGl, anti-hbc IgG3, or both. Three of the blood donor serum samples were positive only by the RIA and for anti-hbc IgM, anti-hbc IgAl, or both. Anti-HBc IgM was found in one sample that was negative by the RIA. Five samples, three of which were positive by the RIA, were negative for anti-hbc IgG1, IgG3, IgM, and IgAl and for anti-hbs (Table 3). The results from the determination of the classes and subclasses of anti-hbc in the 10 blood donor serum samples that were negative for anti-hbs are given in Table 3. Two of the seven reference serum samples were positive for anti-hbc IgGl. The results for the remaining five serum samples are given in Table 3. All discrepant samples were negative for HBs antigen by the RIA and for HBV DNA by the dot blot hybridization assay. The results concerning the classes and subclasses of anti-hbc for all investigated discrepant samples in relation TABLE 2. Endpoint titers for anti-hbc IgGl by EIA of seven serum samples positive for anti-hbc by RIA Identification RIA Endpoint titer by: Anti-HBc IgGl EIA Relative sensitivity' 5194/86 1, , /87 2, , / , / , /87 5,100 3,300, / , / , " Since the serum was diluted 1:100 in the EIA, the relative sensitivity was corrected by use of the indicated factor.

3 VOL. 27, 1989 ANTI-HBc IgGl EIA AND SPECIFICITY OF ANTI-HBc 851 TABLE 3. Results from extended testing of 25 serum samples that gave discrepant results between competitive assays for anti-hbc and EIA for anti-hbc IgGl Inhibition (%) Anti-HBs A42() by EIA for anti-hbc with the indicated specificity: RIA IgM Serum group by anti-hbc RIA RIA S/N`' IgG3 IgM IgA1 capture S/N Consecutive b 0.928" 1.0 Low level b NT" ' NT b NT " " 0.130" NT " " b " 0.738" " NT " NT Blood donor ' NT " 0.170b NT NT " NT NT " NT Reference sera NT 60" " " " " " S/N, Sample to negative ratio. " Value greater than the mean optical density of negative controls + 7 standard deviation units. ' NT, Not tested. d Value greater than the optical density of negative controls + 3 standard deviation units. e Two samples derived from the same patient taken 1 month apart. to anti-hbs results are given in Table 4. There was a highly significant association between the presence of anti-hbc IgGl, anti-hbc IgG3, or both and that of anti-hbs (P < ; the Fisher exact test). No association was found between anti-hbc IgM, anti-hbc IgAl, or both and anti- HBs. The rate of positivity for anti-hbc IgGl in relation to percent inhibition by the RIA is given in Table 5 for all investigated sera. When the RIA was used as the reference method, the specificity of the anti-hbc IgGl EIA was 98.5%, as found by parallel testing of 406 negative serum samples in the consecutive and low-level series. On testing of 232 positive serum samples in the consecutive and low-level series, the sensitivity was 97.4% when the RIA was used as the reference method. TABLE 4. Relation between outcome of tests for different anti- HBc immunoglobulin classes and subclasses and anti-hbs in 25 discrepant serum samples summarized from Table 3 IgM and/or IgAl result No. of sera positive for anti-hbs/no. of sera positive for anti-hbc of IgGl and/or IgG3 + - Total + 2/2 0/10 2/12 5/5 0/8 5/13 Total 7/7' 0/18" 7/25 " P < by the Fisher exact test. DISCUSSION In this study we found that low-level results for anti-hbc by the RIA are either associated with anti-hbc IgGl and/or anti-hbc IgG3 in combination with IgM and/or IgAl, or only with anti-hbc IgM and/or anti-hbc IgAl without anti-hbc IgGl and/or anti-hbc IgG3. We also found that sera with low-level anti-hbc associated with anti-hbc IgGl and/or anti-hbc IgG3 were mostly positive for anti-hibs and that those sera with low-level anti-hl3c which only had anti-hbc of the IgAl and IgM classes were mainly negative for anti-hbs. IgA and IgM are formed early in the immune response and may be used to diagnose ongoing viral infec- TABLE 5. Rate of positivity for anti-hbc IgGl in relation to percent inhibition by the RIA Rate of positivity (no. positive/no. tested) for the following sera: Inhibition by RIA Consecutive Low- Blood level donor Reference Total >85 61/61 34/34 7/7 0/0 102/ /5 56/58" 2/2 0/0 63/ /9 61/64" 1/6 0/4 70/ /16 5/36 0/3 1/2 6/57 <31 1/354 0/0 0/1 1/1 2/356 Total 75/ /192 10/19 2/7 243/663 " One sample was positive for anti-hbc IgG3.

4 852 SALLBERG AND MAGNIUS tions (19). However, we suggest that low-level anti-hbc with IgM and IgAl reactivity may be secondary to other immunological stimuli than HBV, such as another infection activating B cells to produce anti-hbc. This might explain why, on repeated testing of individuals with low-level anti- HBc, it has been found that they tend to become negative for anti-hbc with time (5). In addition to being a T-cell-dependent antigen, the HBc antigen may also be a T-cell-independent antigen (15). Therefore, unspecific activation of immature B cells capable of producing anti-hbc IgM might result in a positive anti- HBc test, which would not be indicative of exposure to HBV. In mice anti-hbc IgM production is T cell independent, but the switch from IgM to IgG is T cell dependent (15). Thus, HBc antigen-activated T cells should be a prerequisite for anti-hbc IgG production. It is therefore most likely that the presence of anti-hbc IgGl and anti-hbc IgG3, which requires a differentiated plasma cell, in contrast to the presence ofanti-hbc IgM and anti-hbc IgAl, is a sign of previous exposure to HBV. Since IgGl is the quantitatively dominating subclass of IgG directed to HBc antigen (18), this subclass is best suited for specificity confirmation of anti-hbc results from competitive assays. In three individuals with low-level anti-hbc in our study, the anti-hbc IgM and anti-hbc IgAl response was secondary to an acute or reactivated Epstein-Barr virus infection. IgA has been shown to be the next most responsible immunoglobulin after IgM for the elevation of total immunoglobulin levels in infectious mononucleosis caused by Epstein-Barr virus (2). The regulation of the two IgA subclasses has been shown to be less T cell dependent than that of the IgG subclasses (3). It has been claimed that before complete maturation is attained, B cells are susceptible to antigenic or mitogenic stimulation, which may give rise to IgM production, IgA production, or both but not to production of any of the IgG subclasses (24). Thus, several lines of evidence indicate that the switching to IgA production is under less stringent regulation than that to IgG is. If the presence of isolated IgM antibodies, IgAl antibodies, or both against other non- T-cell-dependent viral antigens is not diagnostic of exposure to the virus, this will have important implications for the interpretation of low-level results in other serological assays as well. Testing for anti-hbc is currently being introduced at different blood centers worldwide, not only to reduce HBVrelated posttransfusion hepatitis but also to identify blood donors who might be or become infected with other bloodtransmitted agents, such as human immunodeficiency virus and non-a, non-b hepatitis virus. The specificity of anti- HBc tests with regard to HBV exposure is thus very important not only for economic reasons but also because the exclusion of blood donors on the basis of false anti-hbc results, i.e., results not indicative of HBV exposure, can be psychologically harmful. Of 19 blood donors diagnosed as being anti-hbc positive by the EIA (Organon), only 10 were found to have been previously exposed to HBV, according to our results. In our series, the false-positive rate for the anti-hbc EIA (Organon) was thus 9 of 19 (47%). This might be compared with the similar proportion reported by Hanson and Polesky (6) of 52% of serum samples (of 482 samples tested) negative for anti-hbc and anti-hbs by RIA that were repeatedly positive for anti-hbc by a competitive EIA (Corzyme). They also reported that 79% of blood donors with anti-hbc by the RIA had anti-hbs compared with 38% of blood donors with anti-hbc, as detected by the EIA. On the basis of this proposed false-positive rate of 52%, they question the use of the competitive anti-hbc EIA for the screening of blood donors. According to Lee et al. (11), the specificity of commercial anti-hbc tests with recombinant HBc antigen is, however, high, when positive samples are neutralized with purified recombinant HBc antigen. The explanation for these seemingly contradictory findings might be that low-level anti-hbc, although immunologically specific, might often be unrelated to HBV exposure. In a study at 72 different test sites in the United States, Kline et al. (8) found that the prevalence rate of anti-hbc was 2.36%, with a range from 0.55 to 6.38%. In lowprevalence areas, the false-positive rate of anti-hbc may be close to the recorded positive rate, and any possible relation between HBV and other blood-transmitted agents causing posttransfusion hepatitis is likely to be overshadowed. Thus, the relative frequency of false-positive anti-hbc results may vary considerably among different areas. This may explain why conclusions drawn from studies at different blood centers have been conflicting regarding the value of anti- HBc as a surrogate test for reducing the incidence of non-a, non-b-related posttransfusion hepatitis (1, 8, 9, 22, 26). Reports relating anti-hbc to other blood-borne infections must be reevaluated in light of our findings of the significance of low-level results of anti-hbc assays. ACKNOWLEDGMENTS J. CLIN. MICROBIOL. We are indebted to M. Noah, Behringwerke AG, for providing us with HBc antigen-coated microtiter plates. We thank Lena Almgren for skillful technical assistance and M. Marsden for revising the English. LITERATURE CITED 1. Aymard, J. P., C. Janol, S. Gayet, C. Guillemin, P. Canton, P. Gaucher, and F. Streiff Post-transfusion non-a, non-b hepatitis after cardiac surgery. Vox Sang. 51: Bahna, S. L., C. H. Douglas, and C. A. Horwitz Sequential changes of the five immunoglobulin classes and other responses in infectious mononucleosis. Int. Arch. Allerg. Appl. Immunol. 74: Conley, M. E., and M. S. Bartlet In vitro regulation of IgA subclass synthesis. Il. The source of IgA2 plasma cells. J. Immunol. 133: Draelos, M., T. Morgan, R. Schifman, and R. E. Sampliner Significance of isolated antibody to hepatitis B core antigen determined by immune response to hepatitis B vaccination. J. Am. Med. Assoc. 258: Hadler, S. C., B. L. Murphy, C. A. Schable, W. L. Heyward, D. P. Francis, and M. A. Kane Epidemiological analysis of the significance of low-positive test results for antibody to hepatitis B surface and core antigens. J. Clin. Microbiol. 19: Hanson, M. R., and H. F. Polesky Evaluation of routine anti-hbc screening of volunteer blood donors; a questionable surrogate test for non-a, non-b hepatitis. Transfusion 27: Hawkes, R. A., C. R. Boughton, V. Ferguson, and N. I. Lehnman Use of immunoglobulin M antibody to hepatitis B core antigen in diagnosis of viral hepatitis. J. Clin. Microbiol. 11: Kline, W. E., R. J. Bowman, K. K. Ennis McCurdy, J. P. O'Malley, and S. G. Sandler Hepatitis B core antibody (anti-hbc) in blood donors in the United States: implications for surrogate testing. Transfusion 27: Koizol, D. E., P. V. Holland, D. W. Alling, J. C. Melpolder, R. E. Solomon, R. H. Purcell, L. M. Hudson, F. J. Shoup, H. Krakauer, and H. J. Alter Antibody to hepatitis B core antigen as a paradoxical marker for non-a, non-b hepatitis agents in donated blood. Ann. Intern. Med. 104: Lander, J., H. Alter, and R. Purcell Frequency of

5 VOL. 27, 1989 ANTI-HBc IgGl EIA AND SPECIFlCITY OF ANTI-HBc 853 antibody to hepatitis associated antigen as measured by a new radioimmunoassay technique. J. Immunol. 106: Lee, H. S., M. S. Rajagopalan, D. Chien, R. Cordell, H. A. Perkins, and G. N. Vyas Specificity of enzyme immunoassay for hepatitis B core antibody used in screening blood donors. Transfusion 27: Linde, G. A Subclass distribution of rubella virus-specific immunoglobulin G. J. Clin. Microbiol. 21: Linde, G. A., J. Anderson, G. Lundgren, and B. Wahren Subclass reactivity to Epstein-Barr virus capsid antigen in primary and reactivated EBV infections. J. Med. Virol. 21: Ling, C. M., and L. R. Overby Prevalence of hepatitis B virus antigen as revealed by direct radioimmune assay with 1251 antibody. J. Immunol. 109: Milich, D. R., and A. McLachan The nucleocapsid of hepatitis B virus is both a T-cell-independent and a T-celldependent antigen. Science 12: Moreil, A., B. Roth-Wicky, and F. Skvaril Immunoglobulin G subclass restriction of antibodies against hepatitis B surface antigen. Infect. lmm. 39: Overby, L. R., and C. M. Ling Radioimmune assay for anti-core as evidence for exposure to hepatitis B virus. Rush- Presbyterian St. Lukes Med. Bull. 15: Sallberg, M., H. Norder, A. Linde, and L. O. Magnius Subclass pattern of anti-hbc IgG in chronic hepatitis B: relation to HBeAg/anti-HBe and HBV-DNA, p Ini Viral hepatitis and liver disease. Alan R. Liss, Inc., New York. 19. Sarov, I., V. Insler, B. Sarov, R. Cevenini, F. Rumpianesi, M. Donati, D. Kleinman, B. Piura, J. Lieberman, N. Kimmel, M. Friedman, and M. La Placa Specific serum IgA antibodies in the diagnosis of active viral and chlamydia infection, p hi New horizons in microbiology. Elsevier Biomedical Press, Amsterdam. 20. Skvaril, F lgg subclasses in viral infections. Monogr. Allergy 19: Skvaril, F., and H. Joller-Jemeika IgG subclasses of anti-hbs antibodies in vaccinated and non-vaccinated individuals and in anti-hbs immunoglobulin preparations. Int. Arch. Allerg. Apple. Immunol. 73: Stevens, C. E., R. D. Aach, B. Hollinger, J. W. Mosley, W. Szmuness, R. Kahn, J. Werch, and V. Edwards Hepatitis B virus antibody in blood donors and the occurrence of non-a, non-b hepatitis in transfusion recipients. Ann. Intern. Med. 101: Sundqvist, V.-A., A. Linde, and B. Wahren Virus-specific immunoglobulin G subclasses in herpes simplex and varicellazoster virus infections. J. Clin. Microbiol. 20: Teale, J. M., and N. R. Klinman Control of the production of different classes of antibody, p In Fundamental immunology. Raven Press, New York. 25. Tedder, R. S., C. H. Cameron, R. Wilson-Croome, D. R. Howell, A. Colgrove, and J. A. J. Barbara Contrasting patterns and frequency of antibodies to the surface, core and e antigens of hepatitis B virus in blood donors and in homosexual patients. J. Med. Virol. 6: Zuck, T. F., W. C. Sherwood, and J. R. Bove A review of recent events related to surrogate testing of blood to prevent non-a, non-b posttransfusion hepatitis. Transfusion 27: Downloaded from on November 3, 2018 by guest