- Email: [email protected]
Maturation of IgG avidity to individual rubella virus structural proteins
Journal of Clinical Virology 22 (2001) 47 – 54 www.elsevier.com/locate/jcv
Maturation of IgG avidity to individual rubella virus structural proteins Jasminka Nedeljkovic a, Tanja Jovanovic b, Christian Oker-Blom c,* a Institute of Immunology and Virology, Torlak, Belgrade, Yugosla6ia Institute of Microbiology and Immunology, School of Medicine, Belgrade, Yugosla6ia c Department of Biological and En6ironmental Science, Uni6ersity of Jy6a¨skyla¨, FIN-40531 Jy6a¨skyla¨, Finland b
Received 1 December 2000; accepted 5 March 2001
Abstract Background: the structural proteins of rubella virus, the capsid protein C and the envelope glycoproteins E1 and E2 were produced in lepidopteran insect cells using baculovirus expression vectors. The C-terminal ends of the corresponding proteins were fused to a polyhistidine tag for easy and gentle purification by metal ion affinity chromatography. Objecti6es: to investigate the maturation of natural and vaccinal IgG avidity against individual authentic and recombinant rubella virus (RV) structural proteins. Study design: the analysis was carried out using a modified immunoblotting technique where the purified baculovirus-expressed proteins were compared with authentic rubella virus proteins. Altogether, 47 well-characterised serum samples from both naturally infected patients and vaccines were studied. Results: after natural RV infection, IgG antibodies specific for the E1 protein were predominant not only in terms of levels, but also in terms of rate and magnitude of avidity maturation. The avidity development of the IgG antibodies was much slower in vaccines than in patients after a natural RV infection. Conclusions: together, our results indicate that IgG avidity determination in conjunction with immunoblot analysis is useful in the diagnosis of a RV infection. The recombinant proteins showed similar reactivity patterns in the immunoblot analyses as compared with the authentic viral structural proteins, suggesting suitability for serodiagnostics. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Expression; Rubella virus; Structural proteins; Baculovirus; Avidity
1. Introduction Avidity or functional affinity indicates the relative average strength of interaction between the antibody binding sites or paratopes and their * Corresponding author. Tel.: + 358-14-2602285; fax: + 358-14-2602321. E-mail address: [email protected] (C. Oker-Blom).
respective antigenic determinants or epitopes (Hedman et al., 1993; Steward, 1986). The normal primary immune response to infectious agents is characterised by a rapid increase in IgG titre and the maturation from low to high IgG avidity. Introduction of avidity tests for a variety of infectious agents, such as rubella virus (RV), cytomegalovirus (CMV), toxoplasma (Toxo), and
1386-6532/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 6 - 6 5 3 2 ( 0 1 ) 0 0 1 6 1 - 5
48
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
Epstein-Barr virus (EBV) have significantly improved the serological diagnosis of the corresponding pathogens, enabling differentiation of primary infection (low-avidity antibodies) from reinfection or reactivation (high-avidity antibodies). However, avidity studies with individual viral proteins in different infections have shown that individual epitopes can give rise to antibodies of different avidity within the same individual (Kangro et al., 1991; McRae et al., 1991; Radkowski et al., 1993; de Ory et al., 1993; Chargelegue et al., 1993a,b, 1995; Thomas et al., 1996; Thoma, 1997). Avidity assays currently available for confirmation of recent RV infection, e.g. during pregnancy, employ whole virus as target antigen. Since RV contains three structural proteins (Oker-Blom et al., 1983), these assays only permit measurement of average avidity for all epitopes functionally active in the virion. Despite the success of rubella vaccination programs, long-term follow-up studies have shown significantly lower antibody levels in vaccines than in individuals with natural immunity (Christenson et al., 1994), resulting in a greater risk of reinfection in vaccines (Forsgren and Soren, 1985). The present study was designed to compare and characterise the maturation of natural and vaccinal IgG avidity against the individual structural proteins of RV. The results presented here provide further insight into the nature of the immune response in rubella infection, which is of importance in the development of novel subunit vaccines and for more precise virological serodiagnosis.
2. Materials and methods
2.1. Patient sera IgG avidity was studied using sera from individuals with primary natural or vaccinal rubella infection. Paired sera of 11 patients with primary RV infection were collected during a rubella epidemic in 1994. Patients with clinically and virologically confirmed primary infection (positive IgM, seroconversion of hemagglutination inhibi-
tion (HI) and IgG antibodies) were resampled 6 or more months later (IgM negative, IgG and HI positive). For comparison, sera of seven healthy children with rubella infection in the distant past (IgM negative, IgG and HI positive) were also examined. The vaccines were nine earlier seronegative children immunised with Morupar MMR vaccine (Wistar RA 27/3, Biocine Sclavo, Italy). The sera were obtained 1 month (IgM positive and IgG and HI seroconversion) and 12 months (IgM negative, IgG and HI positive) after immunisation.
2.2. Modified immunoblot assay Whole RV and the individual recombinant structural proteins were produced, purified, separated by SDS-PAGE and transferred to nitrocellulose strips under reducing conditions as described (Nedeljkovic et al., 1999). Serum samples diluted 1:50 in Tris–buffered saline (TBST, 0.1% TWEEN-20) containing 2% skimmed milk were incubated for 1 h at room temperature. The strips were washed twice with TBST and treated for 15 min with PBS (pH 7.4) containing 0.1% Tween 20 (PBST), either in the presence or absence of urea. Comparison of different (2–8 M) urea concentrations revealed that with recombinant proteins, low- and high-avidity antibodies were most efficiently distinguished by 8 M urea. The corresponding molar urea concentration for the authentic proteins was 7. After the urea treatment, strips were washed 6 times with TBST and incubated with alkaline phosphatase-conjugated anti-human IgG antibodies (1:15 000; Sigma). The strips were finally washed three times with TBST and the enzyme activity identified with NBT and BCIP (Sigma). After urea elution, significant (+ + + /+ ) or complete (+ + + /− or + + /−) loss of RV specific bands was assessed as the presence of low-avidity IgG antibodies. Accordingly, a mirror pattern (+ + + /+ + + or + + /+ +) or minimal loss of reactivity (+ + + /+ + or + + /+ ) after treatment with urea, was assessed as the presence of high- or moderate-avidity IgG antibodies. In addition, an IgG avidity ELISA using authentic RV antigen was used. The IgG
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
avidity was calculated and the results presented as avidity indexes (AI).
2.3. A6idity ELISA Commercially available microtitre plates containing whole RV or control antigen were treated with the sera (1:100) for 1 h at 37°C according to the instructions provided by the manufacturer (Dade –Behring, Marburg, Germany). After removal of the serum, one half of the plate was soaked three times for 5 min in PBST and the other half in the same buffer containing 8 M urea. The wells were washed six times and the assay was further carried out according to the manufacturer’s instructions. The absorbance was measured at 450 nm and the avidity indexes (AI) were calculated as follows: the absorbance of wells containing the RV antigen minus the absorbance of wells containing the control antigen in the presence of urea was divided by the respective absorbance difference in the absence of urea and multiplied by 100.
3. Results The IgG avidity to authentic and recombinant rubella virus proteins was determined by immunoblot analysis in the presence and absence of urea (Fig. 1).
49
antibodies were of low avidity regardless of when the samples were taken after the onset of the rubelliform rash. The anti-C antibodies detected by the authentic proteins also displayed a time-dependent increase in urea resistance. Interestingly, in all cases the immune response towards the C protein was completely abolished in the acute phase sera in the presence of urea when the recombinant proteins were used. Together, a similar pattern between the avidity maturation of the anti-E1/anti-C antibodies in the immunoblots and the IgG avidity ELISA assays could be observed. At 28–30 days after onset of rash, IgG avidity against whole RV was relatively low; AI = 21–33 (mean 27.9). By 6 and more months after infection, the AIs with whole RV had increased to a mean of 73.63 (54–94).
3.2. Past immunity Table 2 shows the effect of urea in subjects with a rubella infection in the remote past. All such sera contained high-avidity IgG antibodies, with a mean AI of 78.5 when the IgG avidity ELISA was used. The anti-E1 antibodies, detected by the immunoblot assay and recombinant RV proteins, were almost unaffected by urea treatment. The
3.1. Natural RV infection In serum samples obtained 28– 30 days after onset of rash, urea wash usually caused a significant or complete removal of antibodies from the E1 and C proteins (Table 1). The proportion of urea-resistant anti-E1 IgG antibodies increased with time, displaying moderate or high avidity in most convalescent phase serum samples. In contrast, anti-E1 IgG antibodies with a greater susceptibility to the urea treatment was observed in three cases (5, 6 and 9). Anti-E2 antibodies were detected only in patients 5 and 6 with the authentic RV proteins and in patients 1 and 6 with the recombinant proteins (Table 1). In all three cases, the anti-E2 IgG
Fig. 1. Immunoblot pattern of the avidity maturation to baculovirus expressed rubella virus proteins in acute-phase (A) and convalescent-phase (R) sera carried out in the presence ( +) and absence ( −) of urea. The control lane on the left shows the recombinant proteins identified by rabbit anti-RV antiserum (Oker-Blom et al., 1983; Nedeljkovic et al., 1999).
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
50
Table 1 Patients with natural RV infectiona Patient serum
1 2 3 4 5 6 7 8 9 10 11
a
A C A C A C A C A C A C A C A C A C A C A C
Recombinant RV proteins
Authentic RV proteins
E1
E2
C
E1
E2
C
ELISA
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
(AI)
+++/+ +++/++ +++/+ +++/++ ++/− +++/++ +++/+ ++/+ ++/+ +++/+ +/− ++/− ++/+ +++/+++ ++/9 +++/++ +++/+ +++/++ ++/+ +++/++ ++/9 +++/+++
+/− +/− −/− −/− −/− −/− −/− −/− −/− −/− −/− +/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/−
+++/− 9/− +++/− 9/− −/− −/− +++/− 9/9 +++/− 9/− +++/− +/− +++/− +++/− +++/− +/− +++/− 9/− +/− 9/− −/− 9/−
++/+ ++/++ +/− ++/+ +/− ++/+ ++/9 ++/+ −/− ++/9 9/− 9/− ++/9 +++/++ n.t. n.t. ++/− 9/9 +/− ++/++ +/− +++/+++
−/− −/− −/− −/− −/− −/− −/− −/− 9/− 9/− 9/− 9/− −/− 9/− n.t. n.t. +/− −/− −/− −/− −/− −/−
++/+ +++/++ ++/− ++/+ +/− +/− ++/+ ++/+ +++/9 +++/9 +++/++ +/9 +++/+ +++/++ n.t. n.t. ++/− ++/+ +/− +/9 +/− ++/+
31 88 21 90 28 64 33 90 32 67 24 64 33 57 26 54 33 89 21 59 25 65
n.t., not tested sera; 9 Urea, in the absence and presence of urea; A, acute phase sera; C, convalescent-phase sera.
situation with authentic RV proteins was, however, different. In patients 2, 3, and 5, urea treatment caused a significant elution of antibodies from the E1 protein. The anti-E2 antibodies, however, showed minimal avidity maturation both with authentic and recombinant proteins i.e. they were almost completely washed out by the exposure to urea. On the contrary, anti-C antibodies were clearly detectable by the authentic C-protein. No change in the reactivity after the urea treatment was observed except in subjects 5 and 7 (Table 2). However, only two serum samples showed a weak reactivity with the recombinant C-protein. No avidity maturation regarding the anti-C antibodies could, therefore, be determined in this category using the recombinant proteins.
3.3. Vaccines In order to characterise the IgG antibody avidity to the individual RV proteins after a vaccinal RV infection, serum samples of nine earlier seronegative vaccines were investigated. The serum samples were collected 1 and 12 months after immunisation. At 1 month, the RVspecific antibodies exhibited particular susceptibility to treatment with urea (Table 3). When authentic proteins were used, the anti-E1 antibodies were completely removed by urea except in two cases (1 and 8), where no initial reactivity against the authentic proteins could be observed. The anti-E1 antibodies detected by recombinant proteins were significantly (6 and 9), or completely
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
(1, 2, 3, 5, and 8) removed in all vaccines except in two (4 and 7). The anti-C antibodies of five vaccines (2, 5, 6, 7, and 9) analysed with the authentic proteins and of seven vaccines (1, 2, 3, 5, 6, 7, and 9) with the recombinant counterparts were also completely undetectable after urea treatment. The presence of low-avidity antibodies was also observed at one month with whole RV in the IgG avidity ELISA, where the avidity indexes ranged from 3 to 20 (mean 11.2). At 12 months after immunisation, the mean IA value increased from 11.2 to 48.4 (3.5– 61). The avidity maturation to the individual proteins showed a similar, time-dependent pattern of development as was seen in patients with natural rubella infection, but with a significantly smaller proportion of IgG antibodies resistant to the urea treatment. The obvious increase of anti-E1 reactivity as a function of time was not equivalent to the increase in urea resistance. Low-avidity antiE1 antibodies were still predominant at 12 months after immunisation, i.e. in seven out of nine vaccines. Only in two cases, a moderate avidity of the anti-E1 antibodies could be observed. IgG antibodies directed against the E2 and C proteins appeared not to display any increase in the urea resistance. In summary, these results show that the humoral response to RV in vaccines matured with time, but significantly slower than in patients with natural infection.
51
4. Discussion The affinity maturation of the humoral response against the individual structural proteins of RV after both natural and vaccinal infections was studied. The antigens were produced either individually in insect cells using recombinant baculoviruses (Nedeljkovic et al., 1999) or by conventional methods in rubella virus infected mammalian cell cultures (Oker-Blom et al., 1983; Toivonen et al., 1983). The IgG avidity maturation to authentic RV has been described in detail earlier (Hedman and Seppa¨ la¨ , 1988; Morgan-Capner and Thomas, 1988). Our results showed that the E1 glycoprotein was predominant not only in eliciting the immune response during a rubella virus infection, but also in avidity maturation. Anti-C antibodies, detectable by authentic proteins in the convalescent phase of wild-type RV infections, as well as in individuals with RV infection in the distant past, displayed moderate avidity maturation, while the avidity maturation of anti-E2 antibodies was minimal. Maturation of anti-E1 and anti-C antibodies correlated with the maturation of the immune response against whole RV according to the ELISA results. High affinity antibodies are superior to low affinity antibodies in a number of biological reactions including neutralisation of virus infectivity in some test systems (Steward and Steensgaard, 1983).
Table 2 Past immunity seraa Patient
1 2 3 4 5 6 7 a
Recombinant RV proteins
Authentic RV proteins
E1
E2
C
E1
E2
C
ELISA
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
(AI)
+++/++ +++/++ ++/++ +++/+++ +++/++ +++/++ +++/+++
+/− −/− −/− −/− +/− −/− −/
+/− −/− −/− −/− −/− −/− +/+
+++/++ +/− +/9 ++/9 ++/9 +++/++ +++/++
−/− +/− +/− −/− −/− +++/+ 9/−
++/++ +/+ −/− +/9 +/− +++/+++ +/−
80 73 72 85 91 72 77
9 Urea, in the absence and presence of urea.
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
52 Table 3 Vaccinesa Patient
1
Months
1 12 1 12 1 12 1 12 1 12 1 12 1 12 1 12 1 12
2 3 4 5 6 7 8 9
a
Recombinant RV proteins
Authentic RV proteins
E1
E2
C
E1
E2
C
ELISA
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
(AI)
++/− +++/+ +/− ++/+ +/− +++/+ −/− +/− 9/− ++/9 +/− 9/− −/− −/− −/− −/− −/− −/−
−/− −/− −/− −/− −/− −/− −/− −/− +/− +/− +/− 9/− −/− −/− −/− −/− −/− −/−
9/− −/− ++/− ++/− +/− ++/− −/− −/− +/− +/− +/− +/− +/− −/− −/− +/− +/− −/−
−/− 9/− +/− ++/− 9/− ++/+ 9/− +/− +/− +/− +/ 9 ++/9 ++/− ++/9 −/− +++/++ +/− +++/++
−/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− ++/+ −/− −/−
−/− +/− +/− +/− −/− +/− −/− 9/9 +/− ++/+ +/− −/− +/− +/− −/− +/+ ++/− +/−
12 48 2 53 14 51 12 35 20 38 12 51 3 42 16 61 10 57
9 Urea, in the absence and presence of urea.
The present study also showed that detection of low-avidity anti-E1 antibodies distinguishes a recent from a remote RV infection. In our earlier report (Nedeljkovic et al., 1999) we demonstrated a predominance of anti-E1 IgG antibodies and an acute-phase predominance of anti-C IgG antibodies by using baculovirus-expressed recombinant RV proteins. It, therefore, appears that an additional urea treatment as shown here, could be used to increase the precision in confirming a primary RV infection during pregnancy. Thus, detection of low-avidity anti-E1 antibodies together with non-reactive anti-C antibodies by recombinant RV proteins as antigens could be of diagnostic value. In a small number of convalescent sera, as well as in some sera of past rubella infection, low-avidity anti-E1 antibodies were detected. The reason for this is not clear. One possibility could be that these persons were not able to select high affinity anti-E1 response. As the high avidity antibodies are necessary for an efficient, neutralising re-
sponse, the question appears whether these individuals possess a greater risk of getting a rubella reinfection. To assess the exact significance of this lack in avidity maturation, these patients should be followed up for a longer period of time with more detailed studies on the nature of their humoral and cellular immunity. To our knowledge, this is the first study in which the assessment of avidity to all three structural proteins of RV has been made simultaneously by using immunoblot assays. Mauracher et al., 1992 have earlier followed the maturation to the individual RV proteins by using a modified ELISA assay, employing individual RV proteins as antigen. In addition, they showed that the kinetics of anti-E1 antibodies correlated with avidity to whole RV. But avidity to the E2 and C proteins remained low for at least 2 years. In our study, primary vaccine-induced immune responses to whole RV, as well as to the individual RV structural proteins showed a similar pattern of maturation to those from patients with
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
natural rubella infection, but with a gradual increase in avidity over a much longer time than after natural infection. The prolonged avidity maturation after vaccinal infection was especially pronounced when the samples were studied by immunoblot analysis. Such delayed maturation of antibody response to the RV vaccine has earlier been described (Hedman et al., 1989; Enders and Knotek, 1989). Antibodies of high affinity were more efficient in virus neutralisation, as well as in protection of immunised cattle (Steward et al., 1991). In connection with this, the question arises whether vaccine-induced low-avidity anti-RV antibodies are sufficient to prevent a reinfection in vaccines in the first months after vaccination. The answer is, not always, because reinfection in vaccines has been reported (Horstmann et al., 1970). Taking into account all facts mentioned above as well as the fact that vaccine-induced immunity declines as a function of time (Christenson et al., 1994), the idea of a third vaccine dose at childbearing age should be considered. On the other hand, more prolonged persistence of low-avidity antibodies after immunisation might be an indirect clue of post vaccinal reactions related to persistent viral infections (Chantler et al., 1982; Tingle et al., 1985). As long as global eradication of rubella has not been achieved, further insight regarding the nature of the immune response to rubella infection is needed to enable the development of more efficient vaccines. But, it is clear that in assessment of effectiveness of a potential new rubella vaccine, it is necessary to determine the specificity and avidity of antibodies generated against the individual structural proteins in addition to the quantity of specific antibodies. Together, our results show that the baculovirus-expressed recombinant RV proteins are suitable not only for determination of the specificity, but also for the avidity of antibodies against individual RV proteins. More detailed investigations in terms of affinity maturation to the individual RV proteins will be of importance in the development of new effective subunit or synthetic vaccines, as well as for more precise virological diagnosis of the disease.
53
Acknowledgements We are grateful to Klaus Hedman (Department of Virology, Haartman Institute and HUCH Diagnostic, University of Helsinki) for valuable advice and for proof-reading the manuscript and to Aimo Salmi (Department of Virology, University of Turku) for providing the authentic RV control antigen. We also would like to thank Patrik Michel (Department of Biological and Environmental Science, University of Jyva¨ skyla¨ ) for technical assistance. This project has been financed by the Ella and Georg Ehrnrooth Foundation and the Academy of Finland.
References Chantler JK, Ford DK, Tingle AJ. Persistent rubella infection and rubella-associated arthritis. Lancet 1982;1:1323 – 5. Chargelegue D, Colvin BT, O’Toole CM. A 7-year analysis of anti-Gag (p17 and p24) antibodies in HIV-1-seropositive patients with haemophilia, immunoglobulin G titre and avidity are early predictors of clinical course. AIDS 1993a;7:87 – 90. Chargelegue D, O’Toole CM, Colvin BT. A longitudinal study of IgG antibody response to HIV-1 p17 gag protein in HIV-1+ patients with haemophilia, titre and avidity. Clin Exp Immunol 1993b;93:331 – 6. Chargelegue D, Stanley CM, O’Toole CM. The affinity of IgG antibodies to gag p24 and p17 in HIV-1 infected patients correlates with disease progression. Clin Exp Immunol 1995;99:175 – 81. Christenson B, Bo¨ ttiger M. Long-term follow-up study of rubella antibodies in naturally immune and vaccinated young adults. Vaccine 1994;12:41 – 5. de Ory F, Antonaya J, Fernandez MV, Echevarria JM. Application of low-avidity immunoglobulin G studies to diagnosis of Epstein-Barr virus infectious mononucleosis. J Clin Microbiol 1993;31:1669 – 71. Enders G, Knotek F. Rubella IgG total antibody avidity and IgG subclass-specific antibody avidity assay and their role in the differentiation between primary rubella and rubella reinfection. Infection 1989;17:218 – 26. Forsgren M, Soren L. Subclinical rubella reinfection in vaccinated women with rubella-specific IgM response during pregnancy and transmission of virus to the fetus. Scand J Infect Dis 1985;17:337 – 41. Hedman K, Seppa¨ la¨ I. Recent rubella virus infection indicated by a low avidity of specific IgG. J Clin Immunol 1988;8:214 – 21. Hedman K, Hietala J, Tiilikainen A, Hartikainen-Sorri AL, Raiha K, Suni J, Vaananen P, Pietilainen M. Maturation of immunoglobulin G avidity after rubella vaccination
54
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
studied by an enzyme linked immunosorbent assay (avidity-ELISA) and by haemolysis typing. J Med Virol 1989;27:293 –8. Hedman K, Lappalainen M, So¨ derlund M, Hedman L. Avidity of IgG in serodiagnosis of infectious diseases. Rev Med Microbiol 1993;4:123 –9. Horstmann DM, Liebhaber H, Le Bouvier GL, Rosenberg DA, Halsted SB. Rubella, reinfection of vaccinated and naturally immune response exposed in an epidemic. New Engl J Med 1970;283:771 –8. Kangro HO, Manzoor S, Harper DR. Antibody avidity following varicella-zoster virus infections. J Med Virol 1991;33:100 – 5. Mauracher CA, Mitchell LA, Tingle AJ. Differential IgG avidity to rubella virus structural proteins. J Med Virol 1992;36:202 – 8. McRae B, Lange JA, Ascher MS, de Wolf F, Sheppard HW, Goudsmit J, Allain JP. Immune response to HIV p24 core protein during the early phases of human immunodeficiency virus infection. AIDS Res Hum Retrovir 1991;7:637 – 43. Morgan-Capner P, Thomas HI. Serological distinction between primary rubella and reinfection. Lancet 1988;1:1397. Nedeljkovic J, Jovanovic T, Mladjenovic S, Hedman K, Peitsaro N, Oker-Blom C. Natural and vaccine-induced IgG response to rubella virus proteins expressed in insect cells. J Clin Virol 1999;14:119 –31. Oker-Blom C, Kalkkinen N, Ka¨ a¨ ria¨ inen L, Pettersson RF. Rubella virus contains one capsid protein and three envelope glycoproteins, E1, E2a and E2b. J Virol 1983;46:964 – 73.
.
Radkowski M, Laskus T, Goch A, Slusarczyk J. Affinity of anti-GP41 antibody in patients infected with human immunodeficiency virus type 1. Eur J Clin Invest 1993;23:455 – 8. Steward MW. Overview, introduction to methods used to study the affinity and kinetics of antibody-antigen reactions. In: Weir DM, editor. Handbook of Experimental Immunology in Four Volumes: Immunochemistry, vol. 1, Fouth ed. Oxford: Blackwell Scientific Publication, 1986. Steward MW, Steensgaard J. Antibody Affinity, Thermodynamic Aspects and Biological Significance. Boca Ration, FL: CRC Press, 1983. Steward MW, Stanley CM, Dimarchi R, Mulcahy G, Doel TR. High-affinity antibody induced by immunization with a synthetic peptide is associated with protection of cattle against foot-and-month disease. Immunology 1991;72:99 – 103. Thoma HI. Relative functional affinity of specific anti-core IgG in different categories of hepatitis B virus infection. J Med Virol 1997;51:189 – 97. Thomas HI, Wilson S, O’Toole CM, Lister CM, Saeed AM, Watkins RP, Morgan-Capner P. Differential maturation of avidity of IgG antibodies to gp41, p24 and p17 following infection with HIV-1. Clin Exp Immunol 1996;103:185 – 91. Toivonen V, Vainionpa¨ a¨ R, Salmi A, Hyypia¨ T. Glycopolypeptides of rubella virus. Arch Virol 1983;77:91 – 5. Tingle AJ, Chantler JK, Pot KH, Paty DW, Ford DK. Postpartum rubella immunization: association with development of prolonged arthritis, neurological sequelae, and chronic rubella viremia. J Infect Dis 1985;152:606 – 12.
Maturation of IgG avidity to individual rubella virus structural proteins Jasminka Nedeljkovic a, Tanja Jovanovic b, Christian Oker-Blom c,* a Institute of Immunology and Virology, Torlak, Belgrade, Yugosla6ia Institute of Microbiology and Immunology, School of Medicine, Belgrade, Yugosla6ia c Department of Biological and En6ironmental Science, Uni6ersity of Jy6a¨skyla¨, FIN-40531 Jy6a¨skyla¨, Finland b
Received 1 December 2000; accepted 5 March 2001
Abstract Background: the structural proteins of rubella virus, the capsid protein C and the envelope glycoproteins E1 and E2 were produced in lepidopteran insect cells using baculovirus expression vectors. The C-terminal ends of the corresponding proteins were fused to a polyhistidine tag for easy and gentle purification by metal ion affinity chromatography. Objecti6es: to investigate the maturation of natural and vaccinal IgG avidity against individual authentic and recombinant rubella virus (RV) structural proteins. Study design: the analysis was carried out using a modified immunoblotting technique where the purified baculovirus-expressed proteins were compared with authentic rubella virus proteins. Altogether, 47 well-characterised serum samples from both naturally infected patients and vaccines were studied. Results: after natural RV infection, IgG antibodies specific for the E1 protein were predominant not only in terms of levels, but also in terms of rate and magnitude of avidity maturation. The avidity development of the IgG antibodies was much slower in vaccines than in patients after a natural RV infection. Conclusions: together, our results indicate that IgG avidity determination in conjunction with immunoblot analysis is useful in the diagnosis of a RV infection. The recombinant proteins showed similar reactivity patterns in the immunoblot analyses as compared with the authentic viral structural proteins, suggesting suitability for serodiagnostics. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Expression; Rubella virus; Structural proteins; Baculovirus; Avidity
1. Introduction Avidity or functional affinity indicates the relative average strength of interaction between the antibody binding sites or paratopes and their * Corresponding author. Tel.: + 358-14-2602285; fax: + 358-14-2602321. E-mail address: [email protected] (C. Oker-Blom).
respective antigenic determinants or epitopes (Hedman et al., 1993; Steward, 1986). The normal primary immune response to infectious agents is characterised by a rapid increase in IgG titre and the maturation from low to high IgG avidity. Introduction of avidity tests for a variety of infectious agents, such as rubella virus (RV), cytomegalovirus (CMV), toxoplasma (Toxo), and
1386-6532/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 6 - 6 5 3 2 ( 0 1 ) 0 0 1 6 1 - 5
48
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
Epstein-Barr virus (EBV) have significantly improved the serological diagnosis of the corresponding pathogens, enabling differentiation of primary infection (low-avidity antibodies) from reinfection or reactivation (high-avidity antibodies). However, avidity studies with individual viral proteins in different infections have shown that individual epitopes can give rise to antibodies of different avidity within the same individual (Kangro et al., 1991; McRae et al., 1991; Radkowski et al., 1993; de Ory et al., 1993; Chargelegue et al., 1993a,b, 1995; Thomas et al., 1996; Thoma, 1997). Avidity assays currently available for confirmation of recent RV infection, e.g. during pregnancy, employ whole virus as target antigen. Since RV contains three structural proteins (Oker-Blom et al., 1983), these assays only permit measurement of average avidity for all epitopes functionally active in the virion. Despite the success of rubella vaccination programs, long-term follow-up studies have shown significantly lower antibody levels in vaccines than in individuals with natural immunity (Christenson et al., 1994), resulting in a greater risk of reinfection in vaccines (Forsgren and Soren, 1985). The present study was designed to compare and characterise the maturation of natural and vaccinal IgG avidity against the individual structural proteins of RV. The results presented here provide further insight into the nature of the immune response in rubella infection, which is of importance in the development of novel subunit vaccines and for more precise virological serodiagnosis.
2. Materials and methods
2.1. Patient sera IgG avidity was studied using sera from individuals with primary natural or vaccinal rubella infection. Paired sera of 11 patients with primary RV infection were collected during a rubella epidemic in 1994. Patients with clinically and virologically confirmed primary infection (positive IgM, seroconversion of hemagglutination inhibi-
tion (HI) and IgG antibodies) were resampled 6 or more months later (IgM negative, IgG and HI positive). For comparison, sera of seven healthy children with rubella infection in the distant past (IgM negative, IgG and HI positive) were also examined. The vaccines were nine earlier seronegative children immunised with Morupar MMR vaccine (Wistar RA 27/3, Biocine Sclavo, Italy). The sera were obtained 1 month (IgM positive and IgG and HI seroconversion) and 12 months (IgM negative, IgG and HI positive) after immunisation.
2.2. Modified immunoblot assay Whole RV and the individual recombinant structural proteins were produced, purified, separated by SDS-PAGE and transferred to nitrocellulose strips under reducing conditions as described (Nedeljkovic et al., 1999). Serum samples diluted 1:50 in Tris–buffered saline (TBST, 0.1% TWEEN-20) containing 2% skimmed milk were incubated for 1 h at room temperature. The strips were washed twice with TBST and treated for 15 min with PBS (pH 7.4) containing 0.1% Tween 20 (PBST), either in the presence or absence of urea. Comparison of different (2–8 M) urea concentrations revealed that with recombinant proteins, low- and high-avidity antibodies were most efficiently distinguished by 8 M urea. The corresponding molar urea concentration for the authentic proteins was 7. After the urea treatment, strips were washed 6 times with TBST and incubated with alkaline phosphatase-conjugated anti-human IgG antibodies (1:15 000; Sigma). The strips were finally washed three times with TBST and the enzyme activity identified with NBT and BCIP (Sigma). After urea elution, significant (+ + + /+ ) or complete (+ + + /− or + + /−) loss of RV specific bands was assessed as the presence of low-avidity IgG antibodies. Accordingly, a mirror pattern (+ + + /+ + + or + + /+ +) or minimal loss of reactivity (+ + + /+ + or + + /+ ) after treatment with urea, was assessed as the presence of high- or moderate-avidity IgG antibodies. In addition, an IgG avidity ELISA using authentic RV antigen was used. The IgG
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
avidity was calculated and the results presented as avidity indexes (AI).
2.3. A6idity ELISA Commercially available microtitre plates containing whole RV or control antigen were treated with the sera (1:100) for 1 h at 37°C according to the instructions provided by the manufacturer (Dade –Behring, Marburg, Germany). After removal of the serum, one half of the plate was soaked three times for 5 min in PBST and the other half in the same buffer containing 8 M urea. The wells were washed six times and the assay was further carried out according to the manufacturer’s instructions. The absorbance was measured at 450 nm and the avidity indexes (AI) were calculated as follows: the absorbance of wells containing the RV antigen minus the absorbance of wells containing the control antigen in the presence of urea was divided by the respective absorbance difference in the absence of urea and multiplied by 100.
3. Results The IgG avidity to authentic and recombinant rubella virus proteins was determined by immunoblot analysis in the presence and absence of urea (Fig. 1).
49
antibodies were of low avidity regardless of when the samples were taken after the onset of the rubelliform rash. The anti-C antibodies detected by the authentic proteins also displayed a time-dependent increase in urea resistance. Interestingly, in all cases the immune response towards the C protein was completely abolished in the acute phase sera in the presence of urea when the recombinant proteins were used. Together, a similar pattern between the avidity maturation of the anti-E1/anti-C antibodies in the immunoblots and the IgG avidity ELISA assays could be observed. At 28–30 days after onset of rash, IgG avidity against whole RV was relatively low; AI = 21–33 (mean 27.9). By 6 and more months after infection, the AIs with whole RV had increased to a mean of 73.63 (54–94).
3.2. Past immunity Table 2 shows the effect of urea in subjects with a rubella infection in the remote past. All such sera contained high-avidity IgG antibodies, with a mean AI of 78.5 when the IgG avidity ELISA was used. The anti-E1 antibodies, detected by the immunoblot assay and recombinant RV proteins, were almost unaffected by urea treatment. The
3.1. Natural RV infection In serum samples obtained 28– 30 days after onset of rash, urea wash usually caused a significant or complete removal of antibodies from the E1 and C proteins (Table 1). The proportion of urea-resistant anti-E1 IgG antibodies increased with time, displaying moderate or high avidity in most convalescent phase serum samples. In contrast, anti-E1 IgG antibodies with a greater susceptibility to the urea treatment was observed in three cases (5, 6 and 9). Anti-E2 antibodies were detected only in patients 5 and 6 with the authentic RV proteins and in patients 1 and 6 with the recombinant proteins (Table 1). In all three cases, the anti-E2 IgG
Fig. 1. Immunoblot pattern of the avidity maturation to baculovirus expressed rubella virus proteins in acute-phase (A) and convalescent-phase (R) sera carried out in the presence ( +) and absence ( −) of urea. The control lane on the left shows the recombinant proteins identified by rabbit anti-RV antiserum (Oker-Blom et al., 1983; Nedeljkovic et al., 1999).
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
50
Table 1 Patients with natural RV infectiona Patient serum
1 2 3 4 5 6 7 8 9 10 11
a
A C A C A C A C A C A C A C A C A C A C A C
Recombinant RV proteins
Authentic RV proteins
E1
E2
C
E1
E2
C
ELISA
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
(AI)
+++/+ +++/++ +++/+ +++/++ ++/− +++/++ +++/+ ++/+ ++/+ +++/+ +/− ++/− ++/+ +++/+++ ++/9 +++/++ +++/+ +++/++ ++/+ +++/++ ++/9 +++/+++
+/− +/− −/− −/− −/− −/− −/− −/− −/− −/− −/− +/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/−
+++/− 9/− +++/− 9/− −/− −/− +++/− 9/9 +++/− 9/− +++/− +/− +++/− +++/− +++/− +/− +++/− 9/− +/− 9/− −/− 9/−
++/+ ++/++ +/− ++/+ +/− ++/+ ++/9 ++/+ −/− ++/9 9/− 9/− ++/9 +++/++ n.t. n.t. ++/− 9/9 +/− ++/++ +/− +++/+++
−/− −/− −/− −/− −/− −/− −/− −/− 9/− 9/− 9/− 9/− −/− 9/− n.t. n.t. +/− −/− −/− −/− −/− −/−
++/+ +++/++ ++/− ++/+ +/− +/− ++/+ ++/+ +++/9 +++/9 +++/++ +/9 +++/+ +++/++ n.t. n.t. ++/− ++/+ +/− +/9 +/− ++/+
31 88 21 90 28 64 33 90 32 67 24 64 33 57 26 54 33 89 21 59 25 65
n.t., not tested sera; 9 Urea, in the absence and presence of urea; A, acute phase sera; C, convalescent-phase sera.
situation with authentic RV proteins was, however, different. In patients 2, 3, and 5, urea treatment caused a significant elution of antibodies from the E1 protein. The anti-E2 antibodies, however, showed minimal avidity maturation both with authentic and recombinant proteins i.e. they were almost completely washed out by the exposure to urea. On the contrary, anti-C antibodies were clearly detectable by the authentic C-protein. No change in the reactivity after the urea treatment was observed except in subjects 5 and 7 (Table 2). However, only two serum samples showed a weak reactivity with the recombinant C-protein. No avidity maturation regarding the anti-C antibodies could, therefore, be determined in this category using the recombinant proteins.
3.3. Vaccines In order to characterise the IgG antibody avidity to the individual RV proteins after a vaccinal RV infection, serum samples of nine earlier seronegative vaccines were investigated. The serum samples were collected 1 and 12 months after immunisation. At 1 month, the RVspecific antibodies exhibited particular susceptibility to treatment with urea (Table 3). When authentic proteins were used, the anti-E1 antibodies were completely removed by urea except in two cases (1 and 8), where no initial reactivity against the authentic proteins could be observed. The anti-E1 antibodies detected by recombinant proteins were significantly (6 and 9), or completely
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
(1, 2, 3, 5, and 8) removed in all vaccines except in two (4 and 7). The anti-C antibodies of five vaccines (2, 5, 6, 7, and 9) analysed with the authentic proteins and of seven vaccines (1, 2, 3, 5, 6, 7, and 9) with the recombinant counterparts were also completely undetectable after urea treatment. The presence of low-avidity antibodies was also observed at one month with whole RV in the IgG avidity ELISA, where the avidity indexes ranged from 3 to 20 (mean 11.2). At 12 months after immunisation, the mean IA value increased from 11.2 to 48.4 (3.5– 61). The avidity maturation to the individual proteins showed a similar, time-dependent pattern of development as was seen in patients with natural rubella infection, but with a significantly smaller proportion of IgG antibodies resistant to the urea treatment. The obvious increase of anti-E1 reactivity as a function of time was not equivalent to the increase in urea resistance. Low-avidity antiE1 antibodies were still predominant at 12 months after immunisation, i.e. in seven out of nine vaccines. Only in two cases, a moderate avidity of the anti-E1 antibodies could be observed. IgG antibodies directed against the E2 and C proteins appeared not to display any increase in the urea resistance. In summary, these results show that the humoral response to RV in vaccines matured with time, but significantly slower than in patients with natural infection.
51
4. Discussion The affinity maturation of the humoral response against the individual structural proteins of RV after both natural and vaccinal infections was studied. The antigens were produced either individually in insect cells using recombinant baculoviruses (Nedeljkovic et al., 1999) or by conventional methods in rubella virus infected mammalian cell cultures (Oker-Blom et al., 1983; Toivonen et al., 1983). The IgG avidity maturation to authentic RV has been described in detail earlier (Hedman and Seppa¨ la¨ , 1988; Morgan-Capner and Thomas, 1988). Our results showed that the E1 glycoprotein was predominant not only in eliciting the immune response during a rubella virus infection, but also in avidity maturation. Anti-C antibodies, detectable by authentic proteins in the convalescent phase of wild-type RV infections, as well as in individuals with RV infection in the distant past, displayed moderate avidity maturation, while the avidity maturation of anti-E2 antibodies was minimal. Maturation of anti-E1 and anti-C antibodies correlated with the maturation of the immune response against whole RV according to the ELISA results. High affinity antibodies are superior to low affinity antibodies in a number of biological reactions including neutralisation of virus infectivity in some test systems (Steward and Steensgaard, 1983).
Table 2 Past immunity seraa Patient
1 2 3 4 5 6 7 a
Recombinant RV proteins
Authentic RV proteins
E1
E2
C
E1
E2
C
ELISA
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
(AI)
+++/++ +++/++ ++/++ +++/+++ +++/++ +++/++ +++/+++
+/− −/− −/− −/− +/− −/− −/
+/− −/− −/− −/− −/− −/− +/+
+++/++ +/− +/9 ++/9 ++/9 +++/++ +++/++
−/− +/− +/− −/− −/− +++/+ 9/−
++/++ +/+ −/− +/9 +/− +++/+++ +/−
80 73 72 85 91 72 77
9 Urea, in the absence and presence of urea.
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
52 Table 3 Vaccinesa Patient
1
Months
1 12 1 12 1 12 1 12 1 12 1 12 1 12 1 12 1 12
2 3 4 5 6 7 8 9
a
Recombinant RV proteins
Authentic RV proteins
E1
E2
C
E1
E2
C
ELISA
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
9 Urea
(AI)
++/− +++/+ +/− ++/+ +/− +++/+ −/− +/− 9/− ++/9 +/− 9/− −/− −/− −/− −/− −/− −/−
−/− −/− −/− −/− −/− −/− −/− −/− +/− +/− +/− 9/− −/− −/− −/− −/− −/− −/−
9/− −/− ++/− ++/− +/− ++/− −/− −/− +/− +/− +/− +/− +/− −/− −/− +/− +/− −/−
−/− 9/− +/− ++/− 9/− ++/+ 9/− +/− +/− +/− +/ 9 ++/9 ++/− ++/9 −/− +++/++ +/− +++/++
−/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− −/− ++/+ −/− −/−
−/− +/− +/− +/− −/− +/− −/− 9/9 +/− ++/+ +/− −/− +/− +/− −/− +/+ ++/− +/−
12 48 2 53 14 51 12 35 20 38 12 51 3 42 16 61 10 57
9 Urea, in the absence and presence of urea.
The present study also showed that detection of low-avidity anti-E1 antibodies distinguishes a recent from a remote RV infection. In our earlier report (Nedeljkovic et al., 1999) we demonstrated a predominance of anti-E1 IgG antibodies and an acute-phase predominance of anti-C IgG antibodies by using baculovirus-expressed recombinant RV proteins. It, therefore, appears that an additional urea treatment as shown here, could be used to increase the precision in confirming a primary RV infection during pregnancy. Thus, detection of low-avidity anti-E1 antibodies together with non-reactive anti-C antibodies by recombinant RV proteins as antigens could be of diagnostic value. In a small number of convalescent sera, as well as in some sera of past rubella infection, low-avidity anti-E1 antibodies were detected. The reason for this is not clear. One possibility could be that these persons were not able to select high affinity anti-E1 response. As the high avidity antibodies are necessary for an efficient, neutralising re-
sponse, the question appears whether these individuals possess a greater risk of getting a rubella reinfection. To assess the exact significance of this lack in avidity maturation, these patients should be followed up for a longer period of time with more detailed studies on the nature of their humoral and cellular immunity. To our knowledge, this is the first study in which the assessment of avidity to all three structural proteins of RV has been made simultaneously by using immunoblot assays. Mauracher et al., 1992 have earlier followed the maturation to the individual RV proteins by using a modified ELISA assay, employing individual RV proteins as antigen. In addition, they showed that the kinetics of anti-E1 antibodies correlated with avidity to whole RV. But avidity to the E2 and C proteins remained low for at least 2 years. In our study, primary vaccine-induced immune responses to whole RV, as well as to the individual RV structural proteins showed a similar pattern of maturation to those from patients with
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
natural rubella infection, but with a gradual increase in avidity over a much longer time than after natural infection. The prolonged avidity maturation after vaccinal infection was especially pronounced when the samples were studied by immunoblot analysis. Such delayed maturation of antibody response to the RV vaccine has earlier been described (Hedman et al., 1989; Enders and Knotek, 1989). Antibodies of high affinity were more efficient in virus neutralisation, as well as in protection of immunised cattle (Steward et al., 1991). In connection with this, the question arises whether vaccine-induced low-avidity anti-RV antibodies are sufficient to prevent a reinfection in vaccines in the first months after vaccination. The answer is, not always, because reinfection in vaccines has been reported (Horstmann et al., 1970). Taking into account all facts mentioned above as well as the fact that vaccine-induced immunity declines as a function of time (Christenson et al., 1994), the idea of a third vaccine dose at childbearing age should be considered. On the other hand, more prolonged persistence of low-avidity antibodies after immunisation might be an indirect clue of post vaccinal reactions related to persistent viral infections (Chantler et al., 1982; Tingle et al., 1985). As long as global eradication of rubella has not been achieved, further insight regarding the nature of the immune response to rubella infection is needed to enable the development of more efficient vaccines. But, it is clear that in assessment of effectiveness of a potential new rubella vaccine, it is necessary to determine the specificity and avidity of antibodies generated against the individual structural proteins in addition to the quantity of specific antibodies. Together, our results show that the baculovirus-expressed recombinant RV proteins are suitable not only for determination of the specificity, but also for the avidity of antibodies against individual RV proteins. More detailed investigations in terms of affinity maturation to the individual RV proteins will be of importance in the development of new effective subunit or synthetic vaccines, as well as for more precise virological diagnosis of the disease.
53
Acknowledgements We are grateful to Klaus Hedman (Department of Virology, Haartman Institute and HUCH Diagnostic, University of Helsinki) for valuable advice and for proof-reading the manuscript and to Aimo Salmi (Department of Virology, University of Turku) for providing the authentic RV control antigen. We also would like to thank Patrik Michel (Department of Biological and Environmental Science, University of Jyva¨ skyla¨ ) for technical assistance. This project has been financed by the Ella and Georg Ehrnrooth Foundation and the Academy of Finland.
References Chantler JK, Ford DK, Tingle AJ. Persistent rubella infection and rubella-associated arthritis. Lancet 1982;1:1323 – 5. Chargelegue D, Colvin BT, O’Toole CM. A 7-year analysis of anti-Gag (p17 and p24) antibodies in HIV-1-seropositive patients with haemophilia, immunoglobulin G titre and avidity are early predictors of clinical course. AIDS 1993a;7:87 – 90. Chargelegue D, O’Toole CM, Colvin BT. A longitudinal study of IgG antibody response to HIV-1 p17 gag protein in HIV-1+ patients with haemophilia, titre and avidity. Clin Exp Immunol 1993b;93:331 – 6. Chargelegue D, Stanley CM, O’Toole CM. The affinity of IgG antibodies to gag p24 and p17 in HIV-1 infected patients correlates with disease progression. Clin Exp Immunol 1995;99:175 – 81. Christenson B, Bo¨ ttiger M. Long-term follow-up study of rubella antibodies in naturally immune and vaccinated young adults. Vaccine 1994;12:41 – 5. de Ory F, Antonaya J, Fernandez MV, Echevarria JM. Application of low-avidity immunoglobulin G studies to diagnosis of Epstein-Barr virus infectious mononucleosis. J Clin Microbiol 1993;31:1669 – 71. Enders G, Knotek F. Rubella IgG total antibody avidity and IgG subclass-specific antibody avidity assay and their role in the differentiation between primary rubella and rubella reinfection. Infection 1989;17:218 – 26. Forsgren M, Soren L. Subclinical rubella reinfection in vaccinated women with rubella-specific IgM response during pregnancy and transmission of virus to the fetus. Scand J Infect Dis 1985;17:337 – 41. Hedman K, Seppa¨ la¨ I. Recent rubella virus infection indicated by a low avidity of specific IgG. J Clin Immunol 1988;8:214 – 21. Hedman K, Hietala J, Tiilikainen A, Hartikainen-Sorri AL, Raiha K, Suni J, Vaananen P, Pietilainen M. Maturation of immunoglobulin G avidity after rubella vaccination
54
J. Nedeljko6ic et al. / Journal of Clinical Virology 22 (2001) 47–54
studied by an enzyme linked immunosorbent assay (avidity-ELISA) and by haemolysis typing. J Med Virol 1989;27:293 –8. Hedman K, Lappalainen M, So¨ derlund M, Hedman L. Avidity of IgG in serodiagnosis of infectious diseases. Rev Med Microbiol 1993;4:123 –9. Horstmann DM, Liebhaber H, Le Bouvier GL, Rosenberg DA, Halsted SB. Rubella, reinfection of vaccinated and naturally immune response exposed in an epidemic. New Engl J Med 1970;283:771 –8. Kangro HO, Manzoor S, Harper DR. Antibody avidity following varicella-zoster virus infections. J Med Virol 1991;33:100 – 5. Mauracher CA, Mitchell LA, Tingle AJ. Differential IgG avidity to rubella virus structural proteins. J Med Virol 1992;36:202 – 8. McRae B, Lange JA, Ascher MS, de Wolf F, Sheppard HW, Goudsmit J, Allain JP. Immune response to HIV p24 core protein during the early phases of human immunodeficiency virus infection. AIDS Res Hum Retrovir 1991;7:637 – 43. Morgan-Capner P, Thomas HI. Serological distinction between primary rubella and reinfection. Lancet 1988;1:1397. Nedeljkovic J, Jovanovic T, Mladjenovic S, Hedman K, Peitsaro N, Oker-Blom C. Natural and vaccine-induced IgG response to rubella virus proteins expressed in insect cells. J Clin Virol 1999;14:119 –31. Oker-Blom C, Kalkkinen N, Ka¨ a¨ ria¨ inen L, Pettersson RF. Rubella virus contains one capsid protein and three envelope glycoproteins, E1, E2a and E2b. J Virol 1983;46:964 – 73.
.
Radkowski M, Laskus T, Goch A, Slusarczyk J. Affinity of anti-GP41 antibody in patients infected with human immunodeficiency virus type 1. Eur J Clin Invest 1993;23:455 – 8. Steward MW. Overview, introduction to methods used to study the affinity and kinetics of antibody-antigen reactions. In: Weir DM, editor. Handbook of Experimental Immunology in Four Volumes: Immunochemistry, vol. 1, Fouth ed. Oxford: Blackwell Scientific Publication, 1986. Steward MW, Steensgaard J. Antibody Affinity, Thermodynamic Aspects and Biological Significance. Boca Ration, FL: CRC Press, 1983. Steward MW, Stanley CM, Dimarchi R, Mulcahy G, Doel TR. High-affinity antibody induced by immunization with a synthetic peptide is associated with protection of cattle against foot-and-month disease. Immunology 1991;72:99 – 103. Thoma HI. Relative functional affinity of specific anti-core IgG in different categories of hepatitis B virus infection. J Med Virol 1997;51:189 – 97. Thomas HI, Wilson S, O’Toole CM, Lister CM, Saeed AM, Watkins RP, Morgan-Capner P. Differential maturation of avidity of IgG antibodies to gp41, p24 and p17 following infection with HIV-1. Clin Exp Immunol 1996;103:185 – 91. Toivonen V, Vainionpa¨ a¨ R, Salmi A, Hyypia¨ T. Glycopolypeptides of rubella virus. Arch Virol 1983;77:91 – 5. Tingle AJ, Chantler JK, Pot KH, Paty DW, Ford DK. Postpartum rubella immunization: association with development of prolonged arthritis, neurological sequelae, and chronic rubella viremia. J Infect Dis 1985;152:606 – 12.