- Email: [email protected]
Comparison of immunogenicity of recombinant and plasma derived hepatitis B antigen in guinea-pigs
Comparison of immunogenicity of recombinant and plasma derived hepatitis B antigen in guinea-pigs Zdenka Valinger*, Andja Tre~ec and Jelka Toma~,i~ Four different preparations o f hepatitis B antigen (HBsAg) were tested in parallel with respect to their ability to elicit anti-HBs in guinea-pigs. We cbmpared the effect of low doses (5 tzg single dose and 30 pg total amount) of human plasma derived and recombinant HBsAg, in purified forms or prepared as vaccines, respectively. The highest titre of anti-HBs developed was detected in guinea-pigs immunized with plasma derived HBsAg, followed the by response in animals immunized with plasma derived vaccine. Purified recombinant antigen and recombinant vaccine exhibited poor immunogenicity compared with plasma derived antigen preparations. The increase in the dose of recombinant antigen (150 izg total amount) resulted in markedly improved response in two out of ten animals. A nti-HBs immunoglobulin fractions from pooled sera within each group of animals were isolated, partially purified, coupled to polystyrene beads and subsequently used in immunoradiometric assay with monoclona112 Sl.anti_HBs" Only the anti-HBs isolated from pooled sera of animals immunized with purified plasma derived antigen met the requirements for high sensitivity and antigen specificity in such assay. Comparable properties were exhibited by the anti-HBs obtained from one guinea-pig immunized with a larger amount of recombinant antigen. With these two anti-HBs immunoglobulin preparations, amounts as low as 1 ng of HBsAg could be detected.
Keywords:HBsAg;immunoassay; guinea-pig; plasma-derived; recombinant
Introduction Solid phase radioimmunoassay (immunoradiometric assay) as well as enzyme linked immunosorbent assay (ELISA) of hepatitis B antigen (HBsAg) are very sensitive and reliable methods for detection of HBsAg in human plasma and serum, with detection limit below 1 nanogram. Various combinations of monoclonal and polyclonal anti-HBs preparations have been used as 'capture' or 'tracer' antibodies in such non-competitive, 'sandwich' assays. Polyclonal anti-HBs of high titres and avidity could be obtained by hyperimmunizing the animals with human plasma derived HBsAg. However, the isolation of HBsAg from the plasma of persons infected with HBV have always presented a problem due to a time consuming isolation and purification procedure and necessary safety precautions. Therefore, a comparatively simple, large scale production of HBsAg using recombinant DNA techniques seemed to be a very convenient solution. Recombinant HBsAg preparations, available in sufficient quantities, enabled the numerous studies of characteristics and immunogenicity, as well as the comparison of effectiveness with plasma derived antigen 1-12. Although both antigen preparations elicit anti-HBs response in humans and in various experimental animals, the quantitative differences have been reported in several papers, favouring plasma derived antigen, especially for inducing higher early immune response 1,2,4-7. Institute of Immunology, Department of Radioimmunology, PO Box 266, 41000 Zagreb, Yugoslavia. *To whom correspondence should be addressed. (Received 5 February 1990; revised 27 March 1990; accepted 25 April 1990) 0264-410X/90/060585-05 © 1990Butterworth-HeinemannLtd
In this paper we have presented the results of parallel experiments with plasma derived and recombinant HBsAg with respect to their ability to elicit anti-HBs antibody in hyperimmunized guinea-pigs. We have also compared the quality and characteristics of respective anti-HBs immunoglobulins as measured in terms of sensitivity in immunoradiometric assay of HBsAg.
Materials and methods Antigens Antigens used in this study were: (a) commercial samples of the Hevac B Pasteur vaccine (lot llD2) containing 5/~g of the HBsAg adsorbed onto AI(OH)3 in a volume of 1 ml; (b) SKRIT hepatitis B vaccine trial (Smith Kline RIT-Belgium, HBV 058), containing 20/~g of recombinant HBsAg in 1 ml; (c) highly purified HBsAg from the plasma of chronic carriers (adw subtype, 243 #g ml-1, 644753, 7/19/77) (the antigen preparation 13 was a kind gift from Dr J.L. Gerin, obtained through National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, USA); (d) purified recombinant HBsAg (Smith Kline RIT-Belgium, lot 132, 236/~gm1-1 in PBS) (the antigen preparation was a kind gift from Dr P. Hauser). Immunization protocols Laboratory animals used were guinea-pigs, female (250-300 g), five to seven per group. In parallel experiments, groups of guinea-pigs were injected subcutaneously on day 0 with 5 #g doses of each vaccine, respectively. The animals were boostered on days 7, 28, 51, 72 and 93, subcutaneously, with the same
Vaccine, Vol. 8, December 1990 585
Comparison of plasma-derived and recombinant HBsAg: Z. Valinger et al.
doses of vaccine preparations. Animals were bled by cardiac puncture 8 days after each booster, and the sera analysed individually for the titres ofanti-HBs antibodies. The same protocol was followed for immunization with purified antigen preparations (2.5 or 5#g doses); first injections were administered in each hind foot pad, emulsified in complete or incomplete Freund antigen. Booster injections were given subcutaneously as 50% emulsions in incomplete Freund adjuvant. Six combinations with respect to the dose and adjuvant used are indicated in Figure 2, In additional experiments, larger doses of recombinant antigen were used; animals were immunized intradermally with 50 #g recombinant non-adsorbed HBsAg prepared in 0.5ml PBS and emulsified with equal volume of complete Freund adjuvant. Second injections were on day 21 with 100/~g recombinant HBsAg in incomplete Freund adjuvant. Animals were bled by cardiac puncture on day 21 and the sera were tested individually; final bleeding was carried out on day 35.
Anti-HBs assays Antibody titres were determined by radioimmunoassay using the Ausab kit (Abbott Laboratories, Chicago, IL, USA). Individual titres were expressed as a maximal dilution of the serum giving SIN ratio greater than 2.1. The concentration of anti-HBs (IU ml- 1) was assayed by comparison with WHO standard anti-HBs preparation (100 IU ml- ~) obtained from Central Laboratory of The Netherlands Red Cross Blood Transfusion Service, Amsterdam. IU ml- ~ in pooled or individual sera within each group were determined by parallel line bioassay method, as recommended by WHO 14'15.
buffer, pH 7.4 and pretreated with 1% bovine serum albumin. 'Fractions containing radioactive protein were pooled and diluted 2:1 with fetal calf serum. This stock solution was stored at 4°C and further diluted before use.
Radioimmunoassay of HBsAg. Coating of polystyrene beads with purified anti-HBs IgG (0.1-0.15mgml -I as optimal coating concentrations) was carried out in 0.025ra Tris-HCl buffer, pH 8.5, containing 0.02% sodium azide, according to standard procedures used in solid phase immunoassays. Assay protocol was the same as customarily used in commercial kits for HBsAg. Negative and positive controls and samples containing predetermined amounts of HBsAg were incubated with beads coated with various polyclonal anti-HBs preparations. After incubation at room temperature overnight, the beads were washed and incubated with 125I-anti-HBs at 45°C for 1 h. The results are expressed as the quotient (SIN) of the radioactivity (counts min-l) of the sample (S) and the mean radioactivity of the negative control samples (N). Hepatitis HBsAg Sensitivity Panel (Abbott) was used for testing of sensitivity of each assay combination. Results
Immunization with the vaccines The results of immunization with two different vaccine preparations are presented in Figure 1 (titres of anti-HBs)
ixlO6
Solid phase radioimmunoassay Isolation of anti-HBs lgG. Immunoglobulin G (IgG) fraction containing polyclonal anti-HBs was purified from individual or pooled guinea-pig sera by ammonium sulphate precipitation (50% saturation) followed by chromatography on DEAE Sephadex A-25 column, using 0.01 M Tris-HCl buffer, pH 8.5 as eluent. The IgG fractions were concentrated by Amicon pressure dialysis and further fractionated on Sephadex G-200 column using 0.01 M Tris-HCl buffer pH 8.5 as eluent. The purified IgG fractions containing polyclonal anti-HBs were used for coating polystyrene beads.
IxlO 5
g
T
12Sl-anti-HBs. Monoclonal IgG (mouse) antibody against hepatitis B surface antigen (MAS 105c, Sera-Lab, UK) was purified by ammonium sulphate precipitation (50% saturation) and subsequent chromatography on DEAE Sephacel using Tris-HCl buffer, pH 8.0 and NaC1 gradient (0-0.3M16. Isolated monoclonal anti-HBs was labelled with 1251 by modification of a method of Hunter and Greenwood 17 and Purcell et al. is. Briefly, 20#I 0.5M phosphate buffer, pH 7.4, 37MBq (lmCi) ~25I (Amersham, UK), 20#g protein (IgG) and 50#g chloramine T (in 0.25 M phosphate buffer, pH 7.5) were mixed in a small plastic tube and the reaction was allowed to proceed for 20 s. The reaction was terminated by addition of 50#g sodium metabisulphite in 0.25M phosphate buffer. Further, 20 #g potassium iodide, 50 #g sucrose in water containing 0.03% phenol red were added. The reaction mixture was fractionated on Sephadex G-50 column equilibrated and eluted with 0.025 ra phosphate
586
Vaccine, Vol. 8, December 1990
<
IxlO 4
"~ 58
l 59 Time (doys)
I 80
I IOI
Figure 1 Effect of different HBsAg vaccine preparations on anti-HBs titres in guinea-pigs (geometric mean titres, GMT). /k /k Titres of animals immunized with Hevac B (Pasteur) human plasma derived vaccine. [] [] Titres of animals immunized with hepatitis B vaccine (Smith Kline-RIT, Belgium) recombinant vaccine. Animals were immunized on day 0 and boostered on days 7, 28, 51, 72 and 93, with appropriate vaccine preparations containing 5/~g of antigen as a single dose. Bars indicate s.d. *p<0.05 versus the group that received recombinant antigen
Comparison of plasma-derived and recombinant HBsAg: Z. Vatinger Table 1
Results of immunization with vaccine preparations
Antigen used for immunization"
Titre (GMT) *
Anti-HBs (IU ml-1) °
Hevac B (Pasteur) vaccine plasma derived Recombinant vaccine Purified HBsAg - plasma-derived Purified HBsAg - recombinant
304 000 30 000 750 000 190000
345 51 489 186
'Each animal received 5#g of the respective antigen or vaccine preparation as a single dose. Total amount of the antigen per animal was 30#g. The vaccine preparations were used as supplied by the manufacturer. The purified antigen preparations were emulsified in CFA (first injection) and booster injections were injected emulsified in IFA, as described in Materials and methods. *The titres were determined for each serum separately. GMT = geometric mean titre. :The concentration of anti-HBs (IU ml TM) was determined in sera pooled within respective experimental group
and Table 1 (concentrations of anti-HBs). The animals were injected with respective vaccines according to the immunization protocol described in Materials and methods and bled as indicated in Figure I. The immune response was followed up to 101 days. At all intervals examined, the sera obtained from animals immunized with human plasma derived vaccine (5/~g single dose, 30/~g total amount of antigen) exhibited significantly higher values with respect to titres and concentration of anti-HBs, as compared with the recombinant vaccine.
et al.
guinea-pig immunized with large dose of recombinant antigen and exhibiting very high anti-HBs titre, which was processed separately. Immunoglobulin G fractions were isolated from pooled sera and partially purified using conventional gel filtration and ion exchange chromatography; this protein material was coupled to polystyrene beads and further used as solid phase component in sandwich immunoassay of HBsAg. Radioactive component was the same for all combinations investigated: monoclonal ~25I-anti-HBs. The sensitivity of each combination was examined using Hepatitis HBsAg Sensitivity Panel (Abbott). The results are presented in Table 2. The higher sensitivity was obtained with anti-HBs isolated from animals immunized with purified plasma derived antigen. Satisfactory sensitivity (1 ng HBsAg) was also achieved with anti-HBs elicited in one guinea-pig with high dose of recombinant antigen. Along the same line were the results of immunoassay using different ~25i_anti.HBs preparations, namely labelled components from Ausria (Abbott) kit and Auk-3 (Sorin-Biomedica, Saluggia, Italy) kit (data not shown).
Discussion The purpose of this study was to investigate the possibilities to induce formation of high titre anti-HBs
Immunization with purified antigens The results of immunization with purified plasma derived and recombinant antigen preparations, respectively, are presented in Figure 2 and Table 1. The immunization protocol was the same as described for both vaccine preparations. Two different single doses of each antigen were used (2.5 and 5 #g) in combination with complete (CFA) or incomplete (IFA) Freund adjuvant as indicated in Figure 2. At all intervals examined the titres were significantly higher in animals immunized with plasma derived antigen, both for lower and higher doses. Even in experiments with 2.5 #g doses in IFA, the response was higher than in any combination with recombinant antigen. In addition experiments following suggestions of Dr Hauser (Smith-Kline RIT, personal communication) the animals were immunized with markedly higher doses of recombinant antigen - only two injections, of 50 and 100#g were administered, following the immunization protocol described in Materials and methods. The results of two subsequent experiments, comprising groups of five animals, are presented in Figure 3. Very high variations in individual anti-HBs response were obtained. The mean anti-HBs titres were higher than those obtained with low doses of the same antigen, but still not comparable with the response to plasma derived purified antigen. Only one guinea-pig developed anti-HBs of very high titre, which could be used in immunoradiometric assay on HBsAg, as shown below.
Testing o f anti-HBs immunoglobulins in solid phase radioimmunoassay After the final booster dose of the antigen and subsequent bleeding of the animals, serum samples were first tested individually and then pooled within each experimental group. The exception was one serum of
Ixlo 6
8 E o ixlO 5
/
I
/
/
/
/
/
/
/
I~lO 4
f j //"t I
f t
38
/
/f
,I 59
I 80
I I01
Time (days) Rgure 2
Effect of various HBsAg preparations on anti-HBs titres in guinea-pigs (geometric mean titres, GMT). Titres of animals immunized with human plasma derived antigen: O O, 5/.tg, CFA; H , 2.5/~g, CFA; 0 - - - - 0 , 2.5#g, IFA. Titres of animals immunized with recombinant antigen: ~ - - - - / k , 5#g, CFA; • • , 2.5/1g. CFA; • . . . . • , 2.5/tg, IFA. Animals were immunized and boostered as described in Figure 1. Only the first dose was administered in CFA in appropriate combinations. Bars indicate s.d. *p < 0.05 versus the group that received recombinant antigen (5/zg, CFA). **p<0.05 versus the group that received recombinant antigen (2.5/~g, IFA)
V a c c i n e , Vol. 8, D e c e m b e r 1990
587
Comparison of plasma-derived and recombinant HBsAg: Z. Valinger et al.
in guinea-pigs. Immunogenicity of plasma derived and recombinant antigens, respectively, formulated in alum as vaccines, or in purified form administered in Freund adjuvant, was tested in parallel experiments. According to our results, plasma derived antigen elicited higher titres of anti-HBs than recombinant antigen; only anti-HBs induced by purified plasma
b
a 5x IO6
Ix 106
E t9
.&l xlO 5 E o
I
I x 10 4
O
21
28
55 O Ti me ( doys )
21
28
35
Rgure 3 Individual anti-HBs titres of guinea-pigs immunized with recombinant HBsAg. (a) Titres of first group of five guinea-pigs; (b) titres of second group of five guinea-pigs. Each symbol represents the titre values for one guinea-pig. Animals were immunized on day 0 with 50/tg dose, followed by only one booster dose on day 21
Table 2
derived antigen met the specificity and sensitivity requirements for application in solid phase ('sandwich') radioimmunoassay. The immunogenicity of respective antigen and vaccine preparations used in this study was evaluated by the amount of anti-HBs elicited in guinea-pigs. For quantitative determination the Ausab kit (Abbott) was used based on human plasma derived HBsAg as a capture antigen. This fact might partly account for the lower titres obtained with recombinant antigen. The analytical bias in favour of plasma derived HBsAg and vaccine could influence the results but a major factor affecting the immunogenicity will be the structure and source of the antigen. HBsAg, the surface component of hepatitis B virus is a glycosylated protein composed of pre-S and S gene encoded regions. HBsAg in a purified form could be either isolated from human plasma of chronic carriers, or prepared by recombinant techniques. In the process of isolation and purification from plasma, the structure of native HBsAg could be affected; for example, digestion with pepsin would remove pre-S proteins 19 and various virus inactivation procedures lower the immunogenicity. The structures of recombinant antigens correspond to the components of hepatitis B virus envelope; some preparations, however, are not glycosylated and usually do not contain pre-S region. The role and importance of pre-S domains in the process of antibody induction and virus neutralization have been investigated and reported by several groups 2°-24. In general, antigen preparations which represent the surface antigen only partially could exhibit lower immunogenicity and induce the formation of anti-HBs with restricted affinity and avidity for the viral particle or for the complete HBsAg. We have had no official data concerning the structure and composition of any of the antigen preparations used in our experiments. We could only assume, according to the literature data 25'26 that purified recombinant antigen, obtained in yeast, as well as the recombinant vaccine preparation contain only S-gene encoded region and not pre-S domains. On the other hand, HBsAg incorporated in HEVAC (Pasteur) vaccine was isolated from human plasma and presumably, the methods used for its isolation and purification would not have modified the native state of the antigen 27'=8. Moreover, Budkowska
Results of sensitivity testing using different anti-HBs immunoglobulin preparations in immunoradiometric assay of HBsAg Factor
(S/N) obtained
in RIA
Anti-HBs b
HBsAg ~ Concentration (ng m l - ')
HPD HBsAg
Recombinant HBsAg (R-4)
Recombinant HBsAg (group)
Hevac vaccine
Recombinant vaccine
ad
1.65 1.25 0.93
4.18 3.50 2.80
3.45 2.42 2.14
2.30 1.50 1.10
0.98 0.92 1.25
1.66 1.45 1.39
ay
1.83 1.26 0.94
4.50 2.89 2.01
2.95 2.14 1.51
2.11 1.29 0.83
1.83 1.28 0.85
1.96 1.35. 0.68
Subtype
"Part of Hepatitis HBsAg Sensitivity Panel (Abbott) ~Anti-HBs immunoglobulin fractions were isolated from sera of guinea-pigs immunized with different HBsAg preparations or vaccines respectively, as indicated in separate headings R-4 denotes the guinea-pig immunized with large dose of recombinant HBsAg and giving the highest titre of anti-HBsAg (5000000) 'Group' denotes the remaining nine guinea-pigs receiving the large dose of recombinant antigen
588
Vaccine, Vol. 8, D e c e m b e r 1990
C o m p a r i s o n of p l a s m a - d e r i v e d a n d r e c o m b i n a n t H B s A g : Z. Valinger et al.
et al. 24 have shown that Pasteur vaccine induces anti-pre-S2 antibodies, and therefore contains pre-S2 domain. It is conceivable that pre-S2 epitopes would also contribute to the vaccine immunogenicity. On the basis of the data available on the purified HBsAg, adw subtype, obtained through NIH and isolated as described by Gerin et al. 13, it could be expected that this antigen also consists of pre-S and S region. It should be noted that the purified plasma derived antigen, the recombinant antigen, as well as the antigen in recombinant vaccine were of ad HBsAg subtype and Pasteur vaccine contained both ad and ay subtypes. The difference in subtype composition of vaccines used in this study might account for differences in respective S I N values indicated in Table 2, i.e. for smaller SIN values obtained with anti-HBs immunoglobulins induced by Pasteur vaccine. To the best of our knowledge, direct comparison of immunogenicity of purified plasma derived versus recombinant antigen preparations has not been reported so far. On the other hand, comparison of the effects of plasma derived versus recombinant vaccines have been investigated and described in numerous reports 1-12. According to the published results, vaccination with the preparations based on recombinant antigens elicited an anti-HBs response similar to that of plasma derived vaccines. However, a closer analysis of quantitative response, i.e. geometric mean titres of anti-HBs reveals the difference in favour of plasma derived vaccines. More pronounced difference was observed in early responseL2"'v-7; after several booster injections the anti-HBs titres of the same order of magnitude were obtained with both vaccine preparations in several trials. Our results corroborate the findings that the HBsAg isolated from human plasma elicits higher immune response than the antigen produced by recombinant DNA technology, providing that the native structure of plasma derived antigen was not affected to the great extent in the course of isolation procedure. It is to be expected, that such antigen preparations, which have retained native structure, would elicit faster and higher immune response than recombinant antigens devoid of pre-S proteins. It is also to be expected that more economic immunization trials using very low doses of HBsAg vaccines would be more successful with plasma derived vaccines, especially those containing HBsAg comprising all components and structural features of native HBsAg.
Acknowledgements The authors are grateful to Dr J.L. Gerin for his generosity in providing the HBsAg isolated from human plasma and to Dr P. Hauser for the donation of recombinant HBsAg. They are also grateful to Dr S. Smerdel for helpful discussions and to Dr M. Uroi6 for carrying out animal experiments.
3 4
5
6
7
8
9
10 11
12
13
14
15 16
17
18
19
20
21
22
23
24
25
26
References 1
2
Heijtink, R.A., Kruining, J., Bakker, M. and Schlam, S.W. Immune response after vaccination with recombinant hepatitis B vaccines as compared to that after plasma-derived vaccine. Antiviral Res. 1985 (Suppl. 1), 273 Kuwert, E., Scheiermann, N., Gesemann, M., Paar, D., Safary, A., Simeon, E., Hauser, P. and Andre, F. Dose range study in healthy volunteers of a hepatitis B vaccine produced in yeast. Antiviral Res. 1985 (Suppl. 1), 281
27
28
Deinhardt, F. and Jilg, W. Vaccines against hepatitis. Ann. Inst. PasteurlVirol. 1986, 137E, 79 Hollinger, F.B., Troisi, C.L. and Pepe, P.E. Anti-HBs responses to vaccination with human hepatitis B vaccine made by recombinant DNA technology in yeast. J. Infect. Dis. 1986, 153, 156 Scheiermann, N., Gesemann, K.M., Kreutzfelder, E. and Paar, D. Effects of a recombinant yeast-derived hepatitis B vaccine in healthy adults. Postgrad. Med. J. 1987, 63 (Suppl. 2), 115 Just, M., Berger, R. and Just, V. Reactogenicity and immunogenicity of a recombinant hepatitis B vaccine compared with plasma-derived vaccine in young adults. Postgrad. Med. J. 1987, 63 (Suppl. 2), 121 Crovari, P., Crovari, P.C., Petrilli, R.C., Icardi, G.C. and Bonanni, P. Immunogenicity of a yeast-derived hepatitis B vaccine (Energix-B) in healthy young adults. Postgrad. Med. J. 1987, 63 (Suppl. 2), 161 Waters, J.A., O'Rourke, S.M., Richardson, S.C., Papaevangelou, G. and Thomas, H.C. Qualitative analysis of the humoral immune response to the 'a' determinant of HBs antigen after inoculation with plasma-derived or recombinant vaccine. J. Med. Virol. 1987, 21, 155 Wiedermann, G., Scheiermann, N., Goubau, P., Ambrosch, F., Gesemann, M., De Bel, C. et al. Multicentre dose range study of a yeast-derived hepatitis B vaccine. Vaccine 1987, 5, 179 Jilg, W., Schmidt, M. and Deinhardt, F. Immune response to hepatitis B revaccination. J. Med. Virol. 1988, 24, 377 Scheiermann, N., Gesemann, K.M, Paar, D. and Maurer, C. Reactogenicity and immunogenicity of a new recombinant yeastderived hepatitis B vaccine. Zbl. Bakt. Hyg. A. 1988, 269, 411 Stephenne, I. Recombinant vs plasma derived hepatitis B vaccines: issues of safety, immunogenicity and cost effectiveness. Vaccine 1988,6,299 Gerin, J.L., Faust, R.M. and Holland, P.V. Biophysical characterization of the adr subtype of hepatitis B antigen and preparation of anti-r sera in rabbits. J. Immunol. 1975, 115, 100 Ferguson, M., Seagroatt, V. and Schild, G.C. A collaborative study to establish the International Reference Reagent for hepatitis B vaccine containing plasma-derived hepatitis B surface antigen. J. Biol. Stand. 1989, 17, 151 Finney, D.J. Statistical Methods in Biological Assay, 3rd edn. Griffin and Co. Ltd, London, 1978 Manil, L., Motte, P., Pernas, P., Troalen, F., Bohuon, C. and Bellet, D. Evaluation of protocols for purification of mouse myeloma antibodies. J. Immunol. Methods 1986, 98, 25 Hunter, W.M. and Greenwood, F.C. Preparation of iodine-131 labelled human growth hormone of high specific activity. Nature 1982, 194, 495 Purcell, R.H., Wong, D.C., Alter, H.J. and Holland, P.V. Microtiter solid-phase radioi.mmunoassay for hepatitis B antigen. Appl. Microbiol. 1973, 26, 478 Neurath, R.A., Kent, S.B.H., Strick, N., Taylor, P. and Stevens, C.E. Hepatitis B virus contains pre-S gene coded domains. Nature 1985, 315, 154 Neurath, A.R., Strick, N., Kent, S.B.H., Offensperger, W., Wahl, S., Christman, J.K. and Acs, G. Enzyme-linked immunoassay of pre-S gene-coded sequences in hepatitis B vaccines. J. Virol. Meth. 1985, 12, 185 Mora, I., Porres, J.C. and Carreno, V. Receptors for polymerized human serum albumin in plasma derived hepatitis B vaccines. Vaccine 1987, 5, 279 Coursaget, P., Adamowicz, P., Bourdil, C., Yvonnet, B., Buisson, Y., Barres, J.L. et al. Anti-pre-S2 antibodies in nautral hepatitis B virus infection and after immunization. Vaccine 1988, 6, 357 Neurath, A.R., Seto, B. and Strick, N. Antibodies to synthetic peptides from the preSl region of the hepatitis B virus (HBV) envelope (env) protein are virus-neutralizing and protective. Vaccine 1989, 7, 234 Budkowska, A., Dubreuil, P. and Pillot, J. Detection of antibodies to pre-S2 encoded epitopes of hepatitis B virus by monoclonal antibody-enzyme immunoassay. J. Irnmuno/. Methods 1987,102, 85 Petre, J., Van Wijnendaele, F., De Neys, B., Conrath, K., Van Opstal, O., Hauser, P. eta/. Development of a hepatitis B vaccine from transformed yeast cells. Postgrad. Med. J. 1987, 63 (Suppl. 2), 73 Hauser, P., Voet, P., Simeon, E., Thomas, H.C., Petre, J., De Wilde, M. and Stephenne, J. immunological properties of recombinant HBsAg produced in yeast. Postgrad. Med. J. 1987, 6,3(Suppl. 2), 83 Barin, F., Andre, M., Goudeau, A., Coursaget, P. and Maupas, P. Large scale purification of hepatitis B surface antigen (HBsAg). Ann. Microbiol. (Inst. Pasteur) 1978, 129, 87 Adamowicz, P., Chabanier, G., Hyafil, F., Lucas, G., Prunet, P., Reculard, P. and Vinas, R. Elimination of serum proteins and potential virus contaminants during hepatitis B vaccine preparation. Vaccine 1984, 2, 209
Vaccine, Vol. 8, D e c e m b e r 1990
589
Keywords:HBsAg;immunoassay; guinea-pig; plasma-derived; recombinant
Introduction Solid phase radioimmunoassay (immunoradiometric assay) as well as enzyme linked immunosorbent assay (ELISA) of hepatitis B antigen (HBsAg) are very sensitive and reliable methods for detection of HBsAg in human plasma and serum, with detection limit below 1 nanogram. Various combinations of monoclonal and polyclonal anti-HBs preparations have been used as 'capture' or 'tracer' antibodies in such non-competitive, 'sandwich' assays. Polyclonal anti-HBs of high titres and avidity could be obtained by hyperimmunizing the animals with human plasma derived HBsAg. However, the isolation of HBsAg from the plasma of persons infected with HBV have always presented a problem due to a time consuming isolation and purification procedure and necessary safety precautions. Therefore, a comparatively simple, large scale production of HBsAg using recombinant DNA techniques seemed to be a very convenient solution. Recombinant HBsAg preparations, available in sufficient quantities, enabled the numerous studies of characteristics and immunogenicity, as well as the comparison of effectiveness with plasma derived antigen 1-12. Although both antigen preparations elicit anti-HBs response in humans and in various experimental animals, the quantitative differences have been reported in several papers, favouring plasma derived antigen, especially for inducing higher early immune response 1,2,4-7. Institute of Immunology, Department of Radioimmunology, PO Box 266, 41000 Zagreb, Yugoslavia. *To whom correspondence should be addressed. (Received 5 February 1990; revised 27 March 1990; accepted 25 April 1990) 0264-410X/90/060585-05 © 1990Butterworth-HeinemannLtd
In this paper we have presented the results of parallel experiments with plasma derived and recombinant HBsAg with respect to their ability to elicit anti-HBs antibody in hyperimmunized guinea-pigs. We have also compared the quality and characteristics of respective anti-HBs immunoglobulins as measured in terms of sensitivity in immunoradiometric assay of HBsAg.
Materials and methods Antigens Antigens used in this study were: (a) commercial samples of the Hevac B Pasteur vaccine (lot llD2) containing 5/~g of the HBsAg adsorbed onto AI(OH)3 in a volume of 1 ml; (b) SKRIT hepatitis B vaccine trial (Smith Kline RIT-Belgium, HBV 058), containing 20/~g of recombinant HBsAg in 1 ml; (c) highly purified HBsAg from the plasma of chronic carriers (adw subtype, 243 #g ml-1, 644753, 7/19/77) (the antigen preparation 13 was a kind gift from Dr J.L. Gerin, obtained through National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, USA); (d) purified recombinant HBsAg (Smith Kline RIT-Belgium, lot 132, 236/~gm1-1 in PBS) (the antigen preparation was a kind gift from Dr P. Hauser). Immunization protocols Laboratory animals used were guinea-pigs, female (250-300 g), five to seven per group. In parallel experiments, groups of guinea-pigs were injected subcutaneously on day 0 with 5 #g doses of each vaccine, respectively. The animals were boostered on days 7, 28, 51, 72 and 93, subcutaneously, with the same
Vaccine, Vol. 8, December 1990 585
Comparison of plasma-derived and recombinant HBsAg: Z. Valinger et al.
doses of vaccine preparations. Animals were bled by cardiac puncture 8 days after each booster, and the sera analysed individually for the titres ofanti-HBs antibodies. The same protocol was followed for immunization with purified antigen preparations (2.5 or 5#g doses); first injections were administered in each hind foot pad, emulsified in complete or incomplete Freund antigen. Booster injections were given subcutaneously as 50% emulsions in incomplete Freund adjuvant. Six combinations with respect to the dose and adjuvant used are indicated in Figure 2, In additional experiments, larger doses of recombinant antigen were used; animals were immunized intradermally with 50 #g recombinant non-adsorbed HBsAg prepared in 0.5ml PBS and emulsified with equal volume of complete Freund adjuvant. Second injections were on day 21 with 100/~g recombinant HBsAg in incomplete Freund adjuvant. Animals were bled by cardiac puncture on day 21 and the sera were tested individually; final bleeding was carried out on day 35.
Anti-HBs assays Antibody titres were determined by radioimmunoassay using the Ausab kit (Abbott Laboratories, Chicago, IL, USA). Individual titres were expressed as a maximal dilution of the serum giving SIN ratio greater than 2.1. The concentration of anti-HBs (IU ml- 1) was assayed by comparison with WHO standard anti-HBs preparation (100 IU ml- ~) obtained from Central Laboratory of The Netherlands Red Cross Blood Transfusion Service, Amsterdam. IU ml- ~ in pooled or individual sera within each group were determined by parallel line bioassay method, as recommended by WHO 14'15.
buffer, pH 7.4 and pretreated with 1% bovine serum albumin. 'Fractions containing radioactive protein were pooled and diluted 2:1 with fetal calf serum. This stock solution was stored at 4°C and further diluted before use.
Radioimmunoassay of HBsAg. Coating of polystyrene beads with purified anti-HBs IgG (0.1-0.15mgml -I as optimal coating concentrations) was carried out in 0.025ra Tris-HCl buffer, pH 8.5, containing 0.02% sodium azide, according to standard procedures used in solid phase immunoassays. Assay protocol was the same as customarily used in commercial kits for HBsAg. Negative and positive controls and samples containing predetermined amounts of HBsAg were incubated with beads coated with various polyclonal anti-HBs preparations. After incubation at room temperature overnight, the beads were washed and incubated with 125I-anti-HBs at 45°C for 1 h. The results are expressed as the quotient (SIN) of the radioactivity (counts min-l) of the sample (S) and the mean radioactivity of the negative control samples (N). Hepatitis HBsAg Sensitivity Panel (Abbott) was used for testing of sensitivity of each assay combination. Results
Immunization with the vaccines The results of immunization with two different vaccine preparations are presented in Figure 1 (titres of anti-HBs)
ixlO6
Solid phase radioimmunoassay Isolation of anti-HBs lgG. Immunoglobulin G (IgG) fraction containing polyclonal anti-HBs was purified from individual or pooled guinea-pig sera by ammonium sulphate precipitation (50% saturation) followed by chromatography on DEAE Sephadex A-25 column, using 0.01 M Tris-HCl buffer, pH 8.5 as eluent. The IgG fractions were concentrated by Amicon pressure dialysis and further fractionated on Sephadex G-200 column using 0.01 M Tris-HCl buffer pH 8.5 as eluent. The purified IgG fractions containing polyclonal anti-HBs were used for coating polystyrene beads.
IxlO 5
g
T
12Sl-anti-HBs. Monoclonal IgG (mouse) antibody against hepatitis B surface antigen (MAS 105c, Sera-Lab, UK) was purified by ammonium sulphate precipitation (50% saturation) and subsequent chromatography on DEAE Sephacel using Tris-HCl buffer, pH 8.0 and NaC1 gradient (0-0.3M16. Isolated monoclonal anti-HBs was labelled with 1251 by modification of a method of Hunter and Greenwood 17 and Purcell et al. is. Briefly, 20#I 0.5M phosphate buffer, pH 7.4, 37MBq (lmCi) ~25I (Amersham, UK), 20#g protein (IgG) and 50#g chloramine T (in 0.25 M phosphate buffer, pH 7.5) were mixed in a small plastic tube and the reaction was allowed to proceed for 20 s. The reaction was terminated by addition of 50#g sodium metabisulphite in 0.25M phosphate buffer. Further, 20 #g potassium iodide, 50 #g sucrose in water containing 0.03% phenol red were added. The reaction mixture was fractionated on Sephadex G-50 column equilibrated and eluted with 0.025 ra phosphate
586
Vaccine, Vol. 8, December 1990
<
IxlO 4
"~ 58
l 59 Time (doys)
I 80
I IOI
Figure 1 Effect of different HBsAg vaccine preparations on anti-HBs titres in guinea-pigs (geometric mean titres, GMT). /k /k Titres of animals immunized with Hevac B (Pasteur) human plasma derived vaccine. [] [] Titres of animals immunized with hepatitis B vaccine (Smith Kline-RIT, Belgium) recombinant vaccine. Animals were immunized on day 0 and boostered on days 7, 28, 51, 72 and 93, with appropriate vaccine preparations containing 5/~g of antigen as a single dose. Bars indicate s.d. *p<0.05 versus the group that received recombinant antigen
Comparison of plasma-derived and recombinant HBsAg: Z. Vatinger Table 1
Results of immunization with vaccine preparations
Antigen used for immunization"
Titre (GMT) *
Anti-HBs (IU ml-1) °
Hevac B (Pasteur) vaccine plasma derived Recombinant vaccine Purified HBsAg - plasma-derived Purified HBsAg - recombinant
304 000 30 000 750 000 190000
345 51 489 186
'Each animal received 5#g of the respective antigen or vaccine preparation as a single dose. Total amount of the antigen per animal was 30#g. The vaccine preparations were used as supplied by the manufacturer. The purified antigen preparations were emulsified in CFA (first injection) and booster injections were injected emulsified in IFA, as described in Materials and methods. *The titres were determined for each serum separately. GMT = geometric mean titre. :The concentration of anti-HBs (IU ml TM) was determined in sera pooled within respective experimental group
and Table 1 (concentrations of anti-HBs). The animals were injected with respective vaccines according to the immunization protocol described in Materials and methods and bled as indicated in Figure I. The immune response was followed up to 101 days. At all intervals examined, the sera obtained from animals immunized with human plasma derived vaccine (5/~g single dose, 30/~g total amount of antigen) exhibited significantly higher values with respect to titres and concentration of anti-HBs, as compared with the recombinant vaccine.
et al.
guinea-pig immunized with large dose of recombinant antigen and exhibiting very high anti-HBs titre, which was processed separately. Immunoglobulin G fractions were isolated from pooled sera and partially purified using conventional gel filtration and ion exchange chromatography; this protein material was coupled to polystyrene beads and further used as solid phase component in sandwich immunoassay of HBsAg. Radioactive component was the same for all combinations investigated: monoclonal ~25I-anti-HBs. The sensitivity of each combination was examined using Hepatitis HBsAg Sensitivity Panel (Abbott). The results are presented in Table 2. The higher sensitivity was obtained with anti-HBs isolated from animals immunized with purified plasma derived antigen. Satisfactory sensitivity (1 ng HBsAg) was also achieved with anti-HBs elicited in one guinea-pig with high dose of recombinant antigen. Along the same line were the results of immunoassay using different ~25i_anti.HBs preparations, namely labelled components from Ausria (Abbott) kit and Auk-3 (Sorin-Biomedica, Saluggia, Italy) kit (data not shown).
Discussion The purpose of this study was to investigate the possibilities to induce formation of high titre anti-HBs
Immunization with purified antigens The results of immunization with purified plasma derived and recombinant antigen preparations, respectively, are presented in Figure 2 and Table 1. The immunization protocol was the same as described for both vaccine preparations. Two different single doses of each antigen were used (2.5 and 5 #g) in combination with complete (CFA) or incomplete (IFA) Freund adjuvant as indicated in Figure 2. At all intervals examined the titres were significantly higher in animals immunized with plasma derived antigen, both for lower and higher doses. Even in experiments with 2.5 #g doses in IFA, the response was higher than in any combination with recombinant antigen. In addition experiments following suggestions of Dr Hauser (Smith-Kline RIT, personal communication) the animals were immunized with markedly higher doses of recombinant antigen - only two injections, of 50 and 100#g were administered, following the immunization protocol described in Materials and methods. The results of two subsequent experiments, comprising groups of five animals, are presented in Figure 3. Very high variations in individual anti-HBs response were obtained. The mean anti-HBs titres were higher than those obtained with low doses of the same antigen, but still not comparable with the response to plasma derived purified antigen. Only one guinea-pig developed anti-HBs of very high titre, which could be used in immunoradiometric assay on HBsAg, as shown below.
Testing o f anti-HBs immunoglobulins in solid phase radioimmunoassay After the final booster dose of the antigen and subsequent bleeding of the animals, serum samples were first tested individually and then pooled within each experimental group. The exception was one serum of
Ixlo 6
8 E o ixlO 5
/
I
/
/
/
/
/
/
/
I~lO 4
f j //"t I
f t
38
/
/f
,I 59
I 80
I I01
Time (days) Rgure 2
Effect of various HBsAg preparations on anti-HBs titres in guinea-pigs (geometric mean titres, GMT). Titres of animals immunized with human plasma derived antigen: O O, 5/.tg, CFA; H , 2.5/~g, CFA; 0 - - - - 0 , 2.5#g, IFA. Titres of animals immunized with recombinant antigen: ~ - - - - / k , 5#g, CFA; • • , 2.5/1g. CFA; • . . . . • , 2.5/tg, IFA. Animals were immunized and boostered as described in Figure 1. Only the first dose was administered in CFA in appropriate combinations. Bars indicate s.d. *p < 0.05 versus the group that received recombinant antigen (5/zg, CFA). **p<0.05 versus the group that received recombinant antigen (2.5/~g, IFA)
V a c c i n e , Vol. 8, D e c e m b e r 1990
587
Comparison of plasma-derived and recombinant HBsAg: Z. Valinger et al.
in guinea-pigs. Immunogenicity of plasma derived and recombinant antigens, respectively, formulated in alum as vaccines, or in purified form administered in Freund adjuvant, was tested in parallel experiments. According to our results, plasma derived antigen elicited higher titres of anti-HBs than recombinant antigen; only anti-HBs induced by purified plasma
b
a 5x IO6
Ix 106
E t9
.&l xlO 5 E o
I
I x 10 4
O
21
28
55 O Ti me ( doys )
21
28
35
Rgure 3 Individual anti-HBs titres of guinea-pigs immunized with recombinant HBsAg. (a) Titres of first group of five guinea-pigs; (b) titres of second group of five guinea-pigs. Each symbol represents the titre values for one guinea-pig. Animals were immunized on day 0 with 50/tg dose, followed by only one booster dose on day 21
Table 2
derived antigen met the specificity and sensitivity requirements for application in solid phase ('sandwich') radioimmunoassay. The immunogenicity of respective antigen and vaccine preparations used in this study was evaluated by the amount of anti-HBs elicited in guinea-pigs. For quantitative determination the Ausab kit (Abbott) was used based on human plasma derived HBsAg as a capture antigen. This fact might partly account for the lower titres obtained with recombinant antigen. The analytical bias in favour of plasma derived HBsAg and vaccine could influence the results but a major factor affecting the immunogenicity will be the structure and source of the antigen. HBsAg, the surface component of hepatitis B virus is a glycosylated protein composed of pre-S and S gene encoded regions. HBsAg in a purified form could be either isolated from human plasma of chronic carriers, or prepared by recombinant techniques. In the process of isolation and purification from plasma, the structure of native HBsAg could be affected; for example, digestion with pepsin would remove pre-S proteins 19 and various virus inactivation procedures lower the immunogenicity. The structures of recombinant antigens correspond to the components of hepatitis B virus envelope; some preparations, however, are not glycosylated and usually do not contain pre-S region. The role and importance of pre-S domains in the process of antibody induction and virus neutralization have been investigated and reported by several groups 2°-24. In general, antigen preparations which represent the surface antigen only partially could exhibit lower immunogenicity and induce the formation of anti-HBs with restricted affinity and avidity for the viral particle or for the complete HBsAg. We have had no official data concerning the structure and composition of any of the antigen preparations used in our experiments. We could only assume, according to the literature data 25'26 that purified recombinant antigen, obtained in yeast, as well as the recombinant vaccine preparation contain only S-gene encoded region and not pre-S domains. On the other hand, HBsAg incorporated in HEVAC (Pasteur) vaccine was isolated from human plasma and presumably, the methods used for its isolation and purification would not have modified the native state of the antigen 27'=8. Moreover, Budkowska
Results of sensitivity testing using different anti-HBs immunoglobulin preparations in immunoradiometric assay of HBsAg Factor
(S/N) obtained
in RIA
Anti-HBs b
HBsAg ~ Concentration (ng m l - ')
HPD HBsAg
Recombinant HBsAg (R-4)
Recombinant HBsAg (group)
Hevac vaccine
Recombinant vaccine
ad
1.65 1.25 0.93
4.18 3.50 2.80
3.45 2.42 2.14
2.30 1.50 1.10
0.98 0.92 1.25
1.66 1.45 1.39
ay
1.83 1.26 0.94
4.50 2.89 2.01
2.95 2.14 1.51
2.11 1.29 0.83
1.83 1.28 0.85
1.96 1.35. 0.68
Subtype
"Part of Hepatitis HBsAg Sensitivity Panel (Abbott) ~Anti-HBs immunoglobulin fractions were isolated from sera of guinea-pigs immunized with different HBsAg preparations or vaccines respectively, as indicated in separate headings R-4 denotes the guinea-pig immunized with large dose of recombinant HBsAg and giving the highest titre of anti-HBsAg (5000000) 'Group' denotes the remaining nine guinea-pigs receiving the large dose of recombinant antigen
588
Vaccine, Vol. 8, D e c e m b e r 1990
C o m p a r i s o n of p l a s m a - d e r i v e d a n d r e c o m b i n a n t H B s A g : Z. Valinger et al.
et al. 24 have shown that Pasteur vaccine induces anti-pre-S2 antibodies, and therefore contains pre-S2 domain. It is conceivable that pre-S2 epitopes would also contribute to the vaccine immunogenicity. On the basis of the data available on the purified HBsAg, adw subtype, obtained through NIH and isolated as described by Gerin et al. 13, it could be expected that this antigen also consists of pre-S and S region. It should be noted that the purified plasma derived antigen, the recombinant antigen, as well as the antigen in recombinant vaccine were of ad HBsAg subtype and Pasteur vaccine contained both ad and ay subtypes. The difference in subtype composition of vaccines used in this study might account for differences in respective S I N values indicated in Table 2, i.e. for smaller SIN values obtained with anti-HBs immunoglobulins induced by Pasteur vaccine. To the best of our knowledge, direct comparison of immunogenicity of purified plasma derived versus recombinant antigen preparations has not been reported so far. On the other hand, comparison of the effects of plasma derived versus recombinant vaccines have been investigated and described in numerous reports 1-12. According to the published results, vaccination with the preparations based on recombinant antigens elicited an anti-HBs response similar to that of plasma derived vaccines. However, a closer analysis of quantitative response, i.e. geometric mean titres of anti-HBs reveals the difference in favour of plasma derived vaccines. More pronounced difference was observed in early responseL2"'v-7; after several booster injections the anti-HBs titres of the same order of magnitude were obtained with both vaccine preparations in several trials. Our results corroborate the findings that the HBsAg isolated from human plasma elicits higher immune response than the antigen produced by recombinant DNA technology, providing that the native structure of plasma derived antigen was not affected to the great extent in the course of isolation procedure. It is to be expected, that such antigen preparations, which have retained native structure, would elicit faster and higher immune response than recombinant antigens devoid of pre-S proteins. It is also to be expected that more economic immunization trials using very low doses of HBsAg vaccines would be more successful with plasma derived vaccines, especially those containing HBsAg comprising all components and structural features of native HBsAg.
Acknowledgements The authors are grateful to Dr J.L. Gerin for his generosity in providing the HBsAg isolated from human plasma and to Dr P. Hauser for the donation of recombinant HBsAg. They are also grateful to Dr S. Smerdel for helpful discussions and to Dr M. Uroi6 for carrying out animal experiments.
3 4
5
6
7
8
9
10 11
12
13
14
15 16
17
18
19
20
21
22
23
24
25
26
References 1
2
Heijtink, R.A., Kruining, J., Bakker, M. and Schlam, S.W. Immune response after vaccination with recombinant hepatitis B vaccines as compared to that after plasma-derived vaccine. Antiviral Res. 1985 (Suppl. 1), 273 Kuwert, E., Scheiermann, N., Gesemann, M., Paar, D., Safary, A., Simeon, E., Hauser, P. and Andre, F. Dose range study in healthy volunteers of a hepatitis B vaccine produced in yeast. Antiviral Res. 1985 (Suppl. 1), 281
27
28
Deinhardt, F. and Jilg, W. Vaccines against hepatitis. Ann. Inst. PasteurlVirol. 1986, 137E, 79 Hollinger, F.B., Troisi, C.L. and Pepe, P.E. Anti-HBs responses to vaccination with human hepatitis B vaccine made by recombinant DNA technology in yeast. J. Infect. Dis. 1986, 153, 156 Scheiermann, N., Gesemann, K.M., Kreutzfelder, E. and Paar, D. Effects of a recombinant yeast-derived hepatitis B vaccine in healthy adults. Postgrad. Med. J. 1987, 63 (Suppl. 2), 115 Just, M., Berger, R. and Just, V. Reactogenicity and immunogenicity of a recombinant hepatitis B vaccine compared with plasma-derived vaccine in young adults. Postgrad. Med. J. 1987, 63 (Suppl. 2), 121 Crovari, P., Crovari, P.C., Petrilli, R.C., Icardi, G.C. and Bonanni, P. Immunogenicity of a yeast-derived hepatitis B vaccine (Energix-B) in healthy young adults. Postgrad. Med. J. 1987, 63 (Suppl. 2), 161 Waters, J.A., O'Rourke, S.M., Richardson, S.C., Papaevangelou, G. and Thomas, H.C. Qualitative analysis of the humoral immune response to the 'a' determinant of HBs antigen after inoculation with plasma-derived or recombinant vaccine. J. Med. Virol. 1987, 21, 155 Wiedermann, G., Scheiermann, N., Goubau, P., Ambrosch, F., Gesemann, M., De Bel, C. et al. Multicentre dose range study of a yeast-derived hepatitis B vaccine. Vaccine 1987, 5, 179 Jilg, W., Schmidt, M. and Deinhardt, F. Immune response to hepatitis B revaccination. J. Med. Virol. 1988, 24, 377 Scheiermann, N., Gesemann, K.M, Paar, D. and Maurer, C. Reactogenicity and immunogenicity of a new recombinant yeastderived hepatitis B vaccine. Zbl. Bakt. Hyg. A. 1988, 269, 411 Stephenne, I. Recombinant vs plasma derived hepatitis B vaccines: issues of safety, immunogenicity and cost effectiveness. Vaccine 1988,6,299 Gerin, J.L., Faust, R.M. and Holland, P.V. Biophysical characterization of the adr subtype of hepatitis B antigen and preparation of anti-r sera in rabbits. J. Immunol. 1975, 115, 100 Ferguson, M., Seagroatt, V. and Schild, G.C. A collaborative study to establish the International Reference Reagent for hepatitis B vaccine containing plasma-derived hepatitis B surface antigen. J. Biol. Stand. 1989, 17, 151 Finney, D.J. Statistical Methods in Biological Assay, 3rd edn. Griffin and Co. Ltd, London, 1978 Manil, L., Motte, P., Pernas, P., Troalen, F., Bohuon, C. and Bellet, D. Evaluation of protocols for purification of mouse myeloma antibodies. J. Immunol. Methods 1986, 98, 25 Hunter, W.M. and Greenwood, F.C. Preparation of iodine-131 labelled human growth hormone of high specific activity. Nature 1982, 194, 495 Purcell, R.H., Wong, D.C., Alter, H.J. and Holland, P.V. Microtiter solid-phase radioi.mmunoassay for hepatitis B antigen. Appl. Microbiol. 1973, 26, 478 Neurath, R.A., Kent, S.B.H., Strick, N., Taylor, P. and Stevens, C.E. Hepatitis B virus contains pre-S gene coded domains. Nature 1985, 315, 154 Neurath, A.R., Strick, N., Kent, S.B.H., Offensperger, W., Wahl, S., Christman, J.K. and Acs, G. Enzyme-linked immunoassay of pre-S gene-coded sequences in hepatitis B vaccines. J. Virol. Meth. 1985, 12, 185 Mora, I., Porres, J.C. and Carreno, V. Receptors for polymerized human serum albumin in plasma derived hepatitis B vaccines. Vaccine 1987, 5, 279 Coursaget, P., Adamowicz, P., Bourdil, C., Yvonnet, B., Buisson, Y., Barres, J.L. et al. Anti-pre-S2 antibodies in nautral hepatitis B virus infection and after immunization. Vaccine 1988, 6, 357 Neurath, A.R., Seto, B. and Strick, N. Antibodies to synthetic peptides from the preSl region of the hepatitis B virus (HBV) envelope (env) protein are virus-neutralizing and protective. Vaccine 1989, 7, 234 Budkowska, A., Dubreuil, P. and Pillot, J. Detection of antibodies to pre-S2 encoded epitopes of hepatitis B virus by monoclonal antibody-enzyme immunoassay. J. Irnmuno/. Methods 1987,102, 85 Petre, J., Van Wijnendaele, F., De Neys, B., Conrath, K., Van Opstal, O., Hauser, P. eta/. Development of a hepatitis B vaccine from transformed yeast cells. Postgrad. Med. J. 1987, 63 (Suppl. 2), 73 Hauser, P., Voet, P., Simeon, E., Thomas, H.C., Petre, J., De Wilde, M. and Stephenne, J. immunological properties of recombinant HBsAg produced in yeast. Postgrad. Med. J. 1987, 6,3(Suppl. 2), 83 Barin, F., Andre, M., Goudeau, A., Coursaget, P. and Maupas, P. Large scale purification of hepatitis B surface antigen (HBsAg). Ann. Microbiol. (Inst. Pasteur) 1978, 129, 87 Adamowicz, P., Chabanier, G., Hyafil, F., Lucas, G., Prunet, P., Reculard, P. and Vinas, R. Elimination of serum proteins and potential virus contaminants during hepatitis B vaccine preparation. Vaccine 1984, 2, 209
Vaccine, Vol. 8, D e c e m b e r 1990
589