Inhibition ELISA for hepatitis B surface antigen (HBsAg) using monoclonal idiotype-anti-idiotype interaction

Journal of Immunological Methods, 125 (1989) 273-278 Elsevier

273

JIM 05409

Inhibition ELISA for hepatitis B surface antigen (HBsAg) using monoclonal idiotype-anti-idiotype interaction W o n Bae K i m 1, Yeong Woo Jo 2, Eung Chil Choi 1 and Byong K a k K i m 1 1 Department of Microbial Chemistry, College of Pharmacy, Seoul National University, Seoul 151- 742, Korea, and 2 Research Laboratories, Dong-A Pharmaceutical Co. Ltd., Seou1130-070, Korea

(Received 3 October 1988, revised received 15 May 1989, accepted 11 September 1989)

Monoclonal anti-idiotypic antibody (Anti-Id) to the common (a) epitope of hepatitis B surface antigen (HBsAg) were raised and used to detect serum HBsAg in an inhibition enzyme-linked immunosorbent assay (inhibition ELISA). HRP-labeUed Id 8D-3-6 was reacted with Anti-Id 4-8H coated on the solid phase in the presence of HBsAg. The ability of the antigen to inhibit the binding of labelled Id 8D-3-6 to anti-Id 4-8H was determined and the results correlated well with those obtained by radioimmunoassay. This assay requires only one washing step, takes 2 h and covers the range 10 ng/ml to 1/~g HBsAg/ml. The inhibition ELISA is a more convenient, rapid and relatively sensitive assay which can be used to measure the level of a wide range of serum HBsAg. Key words: ELISA, inhibition; Hepatitis B surface antigen; Monoclonal anti-idiotype antibody

Introduction

Detection of serum hepatitis B surface antigen (HBsAg) has been shown to be of value in clinical practice, particularly for the diagnosis of hepatitis B viral infection. This has been accomplished largely by radioimmunoassay (Hollinger et al., 1971; SkinhiSj et al., 1977). However, the disadvantages of using radioactive materials, such as the disposal of waste and the limited half-life of labelled reagents, have led to the development of

Correspondence to: B.K. Kim, College of Pharmacy, Seoul National University, Seoul 151-742, Korea. Abbreviations: HRP, horseradish peroxidase; HBsAg, hepatitis B surface antigen; ELISA, enzyme-linked immunosorbent assay; OPD, o-phenylenediamine.2HC1; OD, optical density; RIA, radioimmunoassay; MAb, monoclonal antibody; Id, idiotypic antibody; Anti-Id, anti-idiotypic antibody.

enzyme immunoassay techniques as alternatives to radioimmunoassay. Several enzyme immunoassays employing the two-step sandwich method (Wei et al., 1977; Adachi et al., 1978; Hyland et al., 1979) have been described and have proved to be sensitive and highly adaptable, but they are time-consuming, multi-step procedures. More recently, a two-site, one-step enzyme immunoassay has been introduced (Liu et al., 1985; Kim et al., 1987). Although this method provides similar sensitivity to the two-step sandwich method whilst significantly reducing the number of steps and total assay time, it is known to be subject to the so-called 'hook effect' (Liu et al., 1985). In order to avoid such limitations, it was decided to investigate an ELISA based on the inhibition of monoclonal Id-Anti-Id interactions. In the present paper, we describe the production of a monoclonal Anti-Id and its application in an inhibition ELISA for the measurement of serum HBsAg.

0022-1759/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

274

Materials and methods

Reagents Horseradish peroxidase, o-phenylenediamine. 2HC1 and glutaraldehyde were obtaiaed from the Sigma Chemical Co., St. Louis, MO, U.S.A. HBsAg standard was from Green Cross Pharm. Co., Japan. All the remaining reagents were purchased from various commercial sources and were of the highest purity available. Monoclonal Id and HRP-labelled antibody The production and characterization of MAb 8D-3-6 directed against the common (a) epitope of hepatitis B surface antigen has already been described (Kim et al., 1987). The antibody was purified from the ascitic fluid by Protein A-Sepharose CL-4B chromatography as described previously (Ey et al., 1978). HRP labelling of 8D-3-6 was performed according to Nakane and Kawaoi (1974). Production of monoclonal Anti-Id Monoclonal Anti-Id was obtained from BALB/c mice immunized with the purified MAb 8D-3-6. BALB/c female mice were immunized intraperitoneally with 100/~g of the idiotype 8D3-6 in Freund's complete adjuvant. Mice were boosted 3 weeks later and again after a further 3 weeks with MAb 8D-3-6 in Freund's incomplete adjuvant. A final boost with 50/xg of MAb 8D-3-6 in 50 mM phosphate-buffered saline (PBS, 50 mM, pH 7.2) was given at least 3 weeks later. Spleen cells of the immunized mice were then fused 4 days later with SP2/0-Ag14 myeloma cells using polyethylene glycol 1500 (KiShler and Milstein, 1975). Clones secreting Anti-Id were identified by binding ELISA and then subcloned by limiting dilution. Quantities of the MAb were produced by growing the hybridomas in ascites of pristane-primed BALB/c mice (Hoogenraad et al., 1983). Subsequently, the ascitic fluids were collected and antibodies were purified from the ascitic fluids by Protein A-Sepharose CL-4B chromatography as described above. Screening of Anti-ld The presence of Anti-Id in culture supernatants was detected by a direct binding assay using MAb

8D-3-6 and normal mouse Ig. An aliquot of 100/~1 of a 10 # g / m l solution of MAb 8D-3-6 or normal mouse IgG in 0.1 M sodium carbonate buffer (pH 9.6) was fixed overnight at 4°C on an immunoplate (Nunc, Roskide, Denmark). After washing with a washing buffer (PBS containing 0.02% Tween 20), 100/~1 of the hybridoma culture supernatants were incubated in the MAb 8D-3-6 or normal mouse Ig-coated plates at 37 °C for 2 h. Following extensive washing with the washing buffer, 100/tl of HRP-labelled MAb 8D-3-6 were added. After 2 h incubation, the plate was again washed with the washing buffer, and 100 /xl of 14.5 mM o-phenylenediamine (OPD) in 0.1 M sodium citrate buffer (pH 5.0) containing 5 mM H202 as a substrate were added and the plate incubated for 30 min at room temperature. The reaction was stopped by the addition of 100/~1 of 2 N sulfuric acid, and bound antibody activity was detected by measuring the optical density at a wavelength of 492 nm. Hybridomas producing antibodies were expanded and further characterized in idiotype binding assay.

Idiotype binding assay Binding of HRP-labelled 8D-3-6 to HBsAg was tested in the presence of possible competitors of Anti-Id hybridoma supernatant. 100/~1 of 3/xg/ml of HBsAg were coated as previously described. 50 /~1 of HRP-labelled 8D-3-6 were incubated overnight at room temperature with an equal volume of various dilutions of hybridoma culture supernatants. The mixture was added to HBsAg-coated plates. At the end of 1 h incubation at 37 o C, the plates were washed and bound enzyme activity was developed and detected as described above. Mouse monoclonal antibody to unrelated antigen, MAb 2F-7 to whole HCG (Kim et al., 1987), was used as a specificity control. Construction of an inhibition ELISA for HBsAg Each plate was coated with 100 /~1 of Anti-Id (10/xg/ml) in 0.1 M sodium carbonate buffer (pH 9.6) as described above. Any remaining binding sites were blocked by incubating with 0.1% of bovine serum albumin in PBS at 37°C for 2 h. After washing, the plate was dried and stored at 4 ° C until use. 50 /~1 of the HBsAg standards or serum samples were incubated with an equal

275

volume of HRP labelled MAb 8D-3-6 for various lengths of time. The reactions were then transferred to the Anti-Id 4-8H coated plate and incubated at 37 °C for 1 h. The plates were washed with the washing buffer and the substrate added. Peroxidase activity was developed and detected as described above.

2.0

£

1.0 "O

Antisera Antisera to HBsAg were obtained from three rabbits (R1-R3), three mice (M1-M3) and six guinea pigs (G1-G6) injected with purified HBsAg (Kim et al., 1987) in complete Freund's adjuvant and boosted (with purified HBsAg) in incomplete Freund's adjuvant. Ten human sera (H1-H10) positive for anti-HBs and negative for HBsAg were obtained from the Korea Red Cross Blood Bank, Seoul, Korea. The anti-HBs activity of each serum was determined by commercial enzyme immunoassay, Ausab EIA (Abbott Lab., North Chicago, IL, U.S.A.).

Results Monoclonal Anti-ld to M A b 8D-3-6 The culture supernatants of approximately 600 hybridomas obtained from two fusions were tested for Anti-Id activity in a binding ELISA using MAb 8D-3-6 and normal mouse Ig-coated plates. Three culture supernatants binding to MAb 8D-3-6 and not to normal mouse Ig were detected. The corresponding hybridomas, named 1-6F, 2-5E, 4-8H were further assayed as possible competitors of Id binding to HBsAg. Two hybridoma supernatants (4-8H, 1-6F) inhibited HRP labelled Id binding to HBsAg (Fig. 1), suggesting that the Anti-Ids are paratope-related Anti-Ids. The hybridomas were recloned and expanded as ascites. Construction of an inhibition E L I S A for H B s A g The development of the assay for quantification of HBsAg was based on the inhibition of the Anti-Id binding to its idiotype by HBsAg. HRP-labelled MAb 8D-3-6 was utilized as a probe, and Anti-Id 4-8H served as a solid-phase antibody. Optimal concentrations of both MAb 8D-3-6 and Anti-Id 4-8H were determined on the basis of 8D-3-6 reactivity with 4-8H in the solid phase. An

o 0.0 J

I

I

I

I

I

i

I

I

I

I

2o 2-1 2-2 2-3 2-4. 2-s 2-b 2-7 2-s 2-0 2-1o Dilution of hybridorna supernatants

Fig. 1. Binding of HRP-labeUed idiotype to HBsAg. Two-fold dilutions of anti-idiotype secreting hybridoma supernatants, 4-8H (o), 1-6F (11), 2-5E (A) and unrelated anti-HCG hybridoma supernatant (o), were incubated with HRP-labelled idiotype before adding to HBsAg-coated wells of plates.

increased amount of labelled MAb 8D-3-6 and Anti-Id 4-8H resulted in increased capacity for binding labelled MAb 8D-3-6. When 4-8H was coated at higher concentrations than 10 /~g/ml, maximal levels of binding were observed. In order to confirm the optimal assay system, concentrations of labelled MAb 8D-3-6 and Anti-Id 4-8H corresponding to an absorbance range of 1.5-1.7 in the steepest part of the sigmoidial binding curves were used in a further study. Fig. 2 shows dose-response curves at various preincubation times. The inhibition exerted by HBsAg on the Id-Anti-Id interaction was tested after different times of incubation and was expressed as a percentage of (B 0 - B ) / B o, where B represents the amount of binding of the HRP labelled Id 8D-3-6 to Anti-Id 4-8H in the presence of different concentrations of HBsAg, and B 0 represents the maximum binding of HRP-labelled MAb 8D-3-6 to Anti-Id 4-8H in the absence of HBsAg. As shown in Fig. 2, the sensitivity improved when preincubation times were increased from 0 to 30 min. However, preincubation times longer than 30 min did not affect the dose-response curve. Also, the minimum amount detectable by the inhibition ELISA was 10 n g / m l of HBsAg in the 30 rain preincubation curve, when the sensitivity was defined as the HBsAg concentration exerting 10%

276 100

TABLE I R E L A T I O N S H I P B E T W E E N S H A R E D Id LEVELS A N D Id INHIBITION VALUES Serum sources

.ID

" 50 c m

0

1

1'o

i

,oo

HBsAg c o n c e n t r a t i o n ,

lO'OO ng/ml

Fig. 2. Influence of preincubation times of HRP-labelled idiotype 8D-3-6 and HBsAg on the dose-response curve. HRPlabelled idiotype 8D-3-6 was incubated with various concentrations of HBsAg immediately (o), 30 re_in (El), 2 h (A), or 18 h (e) before addition to anti-idiotype 4-8H coated microplatewells.

inhibition (two standard deviations above the mean value of ten normal human sera).

Inhibition of anti-idiotype binding to idiotype by polyclonal anti-HBs The interaction of polyclonal anti-HBs with Anti-Id 4-8H inhibited the binding of labelled MAb 8D-3-6, because the Anti-Id 4-8H was masked by the binding of its corresponding Id in polyclonal anti-HBs. This was tested by adding a high titer of rabbit, mouse, guinea pig and human polyclonal anti-HBs. The mixture of the polyclonal anti-HBs and the labelled MAb 8D-3-6 were preincubated for 30 min and assayed by the inhibition ELISA. Anti-HBs sera obtained from three mice, three rabbits and five guinea pigs inhibited the Id-Anti-Id interaction to differing degrees, whereas human anti-HBs failed to inhibit the Id-Anti-Id interaction (Table I). These results suggested the presence of shared idiotypes in tested animals, but not in humans. A direct binding assay was therefore performed to ascertain whether shared idiotypes were detected using anti-HBs from different species. In the direct binding assay, Anti-Id was fixed to the solid phase; biotinlabelled HBsAg and HRP-labeUed avidin were used with the commercial 'Ausab' enzyme immunoassay. The antiserum giving the highest Id

Anti-HBs activity a

Mouse M 1 M 2 M 3 Guinea pig G 1 G 2 G 3 G 4 G 5 G 6 Rabbit R 1 R 2 R 3 Human HI-10 ¢

Shared Id activity b

% Inhibition of Id binding to Anti-Id

12 800 6 400 12 800

640 320 640

25 22 36

64500 16 200 32 300 64100 32400 32700

320 640 1280 640 43 640

16 25 61 34 6 33

6400 6400 12 800

320 320 320

13 15 19

1250-3130 d

0

0--3 d

a Reciprocal dilution of antisera giving positive results when measured by enzyme immunoassay, A u s a b (Abbott Lab., North Chicago, IL, U.S.A.). b Shared idiotype activity was measured by direct ELISA. Antisera were added to the 4-8H coated wells of plates and incubated for 2 h at 40 o C. Bound antibodies were determined by a biotin-avidin amplified EIA, Ausab. Data are expressed as the reciprocal of dilution giving positive results. e Ten sera obtained from different individuals. d Range of values. 20 10 eo e~

e

***.

cD

1

oe

•

0.1

•~ o.oi o o ° o

°° o o o i

i

i

0.01

0.1

1

RIA

i

i

10 20

, ~Ug/ml

Fig. 3. Correlation between levels of serum HBsAg measured by R I A and by the inhibition ELISA ( y = 0 . 8 8 6 x - 0 . 0 2 4 , r = 0.984, n = 40). (O) H B s A g positive serum, ( o ) negative serum.

277 inhibition value showed the highest shared Id activity.

Correlation with radioimmunoassay (RIA) 40 HBsAg positive sera (10 n g / m l to 20 /xg H B s A g / m l ) and ten negative sera were obtained from the Korea Red Cross Blood Bank, Seoul, Korea. These sera were assayed using a commercial radioimmunoassay, Ausria II-125 (Abbott Lab., North Chicago, IL, U.S.A.) and by the inhibition ELISA described above with 30 rain preincubation. A good correlation was obtained between the inhibition ELISA and R I A when the levels of HBsAg in sera were determined (Fig. 3).

Discussion In this report we have described the development of a new inhibition ELISA based on an Id-Anti-Id interactions and its application to the detection of HBsAg. Recent studies have used the same principle to detect various targets (Potocnjak et al., 1982; Sluis et al., 1985; Thompson et al., 1985). A monoclonal Anti-Id was prepared and fixed to a solid phase before the addition of a preincubated mixture of the sera to be tested and labelled Id. In such an assay HBsAg present in the sera inhibit the binding of the Anti-Id on the solid phase to HRP-labelled Id. This inhibition ELISA is simple, rapid and easy to perform compared to the classical ELISA (Wei et al., 1977). Although not providing the sensitivity of a commercially available kit employing the principle of the solidphase double antibody sandwich method (Nath et al., 1983), the present assay offers several advantages. Washing steps and incubation time can be reduced to allow many samples to be assayed within a few hours. Our studies are particularly suited to quantitating a wide range of HBsAg concentrations from 10 n g / m l to 1/~g/ml. In the classical two-step ELISA method, the dose-response curve is linear for a narrow range of HBsAg concentrations. When the HBsAg concentration exceeds a certain limit, the hook effect is observed in the one-step ELISA method. Therefore multiple dilutions should be tested in both one-step and two-step ELISA procedures. In contrast, in the present study, almost all of the sam-

ples could be assayed without dilution. However, the inhibition ELISA does have a possible source of error - the presence in the sample of the same Id as the labelled Id. Findings by Thanavala (1985) showed that internal image monoclonal Anti-Ids were able to inhibit the binding of the HBsAg to its corresponding idiotype in human, goat and swine polyclonal anti-HBs. Because of this reactivity with a variety of polyclonal anti-HBs antisera, the inhibition ELISA described above might erroneously give a false positive result. As shown in Table I, anti-HBs antisera from three different species of animals inhibited the Id-Anti-Id interaction, whereas h u m a n anti-HBs antisera did not inhibit the Id binding to an appreciable extent. It has previously been demonstrated that the extent of shared Id varied among different animals in each species and a high inhibition value was not associated with a high anti-HBs titer (Kennedy et al., 1983). We found that the Id-Anti-Id interaction was related to shared Id activity, not to anti-HBs activity. In human anti-HBs antisera with high anti-HBs activity, shared Id activity was not detected. These results could be due to either the absence of the shared Id or the preferential expression of other Id, in which the shared Id represented a minor population of the total antiHBs. In conclusion, we have shown that a new immunoassay involving the principle of competitive inhibition of the Id-Anti-Id interaction can provide a level of sensitivity approaching that of the classical ELISA, and can be applied over the same working range. Also, our study demonstrates that the idiotype-anti-idiotype approach is a reasonable alternative to the established immunoassay procedures for HBsAg.

References Adachi, H., Fukada, T., Funahushi, S., Kurahori, T. and Ishikawa, E. (1978) Sandwich enzyme immunoassay of hepatitis B surface antigen (HBsAg). Vox Sang. 35, 219. Ey, Y.-S.V., Prowse, S.J. and Jenkin, C.R. (1978) Isolation of pure IgG1, IgG2a, and IgG2b immunoglobulinsfrom mouse serum using Protein A-Sepharose. Immunochemistry 15, 429. Hollinger, F.B., Vomdam, V. and Dressman, G.R. (1971) Assay of Australia antigen and antibody employing

278 double-antibody and solid-phase radioimmunoassay techniques and comparison with the passive hemagglutination methods. J. Immunol. 107, 1099. Hoogenraad, N., Helma, T. and Hoogenraad, J. (1983) The effect of preinjection of mice with pristane on ascites tumour formation and monoclonal antibody production, J. Immunol. Methods 61,317. Hyland, C.A., Mason, E.C., Harden, P.A., Shaw, A.E., Mayrer, D. and Hope, S.L. (1979) Sensitivities of radioimmunoassay and enzyme-linked immunosorbent assay for detection of hepatitis B surface antigen. Vox Sang. 36, 137. Kennedy, R.C., lonescu-Matiu, I., Sanchez, Y. and Dressman, G.R. (1983) Detection of interspecies idiotype cross-reactions associated with antibodies of hepatitis B surface antigen. Eur. J. Immunol. 13, 232. Kim, G.W., Hong, S.Y., Shin, S.C., Lee, S.H. and Kim, W.B. (1987) Studies on the development of enzyme linked immunosorbent assay (ELISA) for hepatitis B surface antigen (HBsAg) by monoclonal antibodies of different affinity constants. Arch. Pharm. Res. 10, 18. Kim, H.J., Kim, G.W., Cha, S.H., Kim, W.B. and Byun, S.M. (1987) Purification of mouse monoclonal anti-HCG IgG from ascites fluid by hydroxylapatite chromatography. Biotech. Tech. 1, 269. K/Shier, G. and Milstein, C. (1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495. Liu, Y.-S.V. and Green, A. (1985) A monoclonal antibody enzyme immunoassay for detection of hepatitis B surface antigen with use of a biotin-avidin system. Clin. Chem. 31, 202.

Nakane, P. and Kawaoi, P. (1974) Peroxidase-labeled antibody. A new method of conjugation. J. Histochem. Cytochem. 22, 1084. Nath, N., Dodd, R.Y. and Fang, C.T. (1983) Enzyme-linked immunoassay for hepatitis B surface antigen: An evaluation of commercial test kits. Transfusion 23, 45. Potocnjak, P., Zavala, F., Nussenzweig, R. and Nussenzweig, V. (1982) Inhibition of idiotype-anti-idiotype interaction for detection of a parasite antigen: A new immunoassay. Science 215, 1637. Skinhtij, P., Nielsen, J.O. and Dietrichson, O. (1977) Serological evidence of hepatitis B infection in patients with chronic liver diseases: Radioimmunoassay of HBsAg and anti-HBs. Scand. J. Gastroenterol. 12, 615. Sluis, K., Magnusson, C.G.M. and Masson, P.L. (1985) Inhibition of idiotype-anti-idiotype reaction in particle counting immunoassay as a tentative assay of IgE. J. Immunol. Methods 76, 255. Thanavala, Y.M., Bond, A., Tedder, R., Hay, F.C. and Roitt, I.M. (1985) Monoclonal 'internal image' anti-idiotypic antibodies of hepatitis B surface antigen. Immunology 55, 197. Thompson, R.E., Hewitt, C.R., Piper, D.J., Hansen, W.P., Rubin, R.H., Tolkoff-Rubin, N.E., Barrett, M.C. and Nelles, M.J. (1985) Competitive idiotype-anti-idiotype immunoassay for adenosine deaminase binding protein in urine. Clin. Chem. 31, 1833. Wei, R., Knight, G.J., Zimmerman, D.H. and Bond, H.E. (1977) Solid-phase enzyme immunoassay for hepatitis B surface antigen. Clin. Chem. 23, 813.