Immunoblot analysis of natural and vaccine-induced IgG responses to rubella virus proteins expressed in insect cells.

BACKGROUND: The three structural proteins of rubella virus (RV), the capsid protein C and the envelope glycoproteins E1 and E2, were produced individually in soluble form in Sf9 insect cells using the baculovirus system. All proteins were equipped with a polyhistidine tag at their C-terminal ends to enable gentle purification by metal ion affinity chromatography. In addition, the E1 and E2 proteins were engineered to display the FLAG epitope tag at their N-terminal ends. STUDY DESIGN: The diagnostic potential of the recombinant purified proteins was evaluated by immunoblot and enzyme immuno assays (EIA) using a total of 57 well-characterised serum samples obtained at various time points after natural RV infection, congenital rubella syndrome (CRS), MMR vaccination or from controls with past RV immunity. In addition, acute and convalescent phase serum pools from a total of 20 patients were evaluated. Authentic RV proteins were used as a reference. RESULTS: The recombinant E1 and C proteins were predominant in eliciting the immune response in both postnatal and vaccinal RV infections, being much weaker in the vaccinal ones. The IgG response to the recombinant C protein was very strong after the first month post infection and decreased with time. The immune response against the recombinant E2 protein, however, was generally poor, but notably stronger after congenital infection. Together, the results showed that the individual recombinant protein antigens could be suitable for diagnosis of RV infection and for study of the immune response to rubella vaccination.     

Differential IgG avidity to rubella virus structural proteins.

Avidity maturation of IgG antibody responses directed against the structural proteins of rubella virus (E1, E2, and C) as well as whole rubella virus (RV) was assessed at sequential time intervals in 7 individuals following serologically confirmed wild rubella infection. Individual structural proteins were purified from tissue culture supernatants by differential centrifugation, followed by preparative SDS-PAGE under non-reducing conditions. Avidity of IgG anti-rubella responses was measured by using the 8 M urea elution technique and results expressed as an elution ratio [ER(%)]. A low mean ER(%) of 23% was determined for E1-specific IgG responses during the 10-20 day period following onset of clinical rubella, with subsequent maturation of avidity ER(%) values to 52%, 75%, and 84% at 3 months, 1 year, and 2 years, respectively, post-rubella. In contrast, IgG anti-E2 responses showed minimal avidity maturation with ER(%) values of 20%, 29%, 30%, and 31% over the same time intervals. Similarly, responses to the capsid protein (C) remained at low avidity ER(%) values of 21%, 29%, 36%, and 35% over the 2 year follow-up period. The avidity maturation values for IgG directed against whole RV preparations paralleled observations for E1-specific responses with ER(%) values of 23%, 52%, 85%, and 87%, respectively. These data support the need to assess individual protein-specific antibody avidities in order to more fully understand viral-specific immune responses.     

Diagnosis of recent primary rubella virus infections: significance of glycoprotein-based IgM serology, IgG avidity and immunoblot analysis

Reliable serodiagnosis of rubella virus (RV) infections requires discrimination of specific IgM induced by primary rubella from persistent, reactivated or non-specific IgM reactivity. Sera from 130 pregnant women with recent or past RV infection/vaccination, persistent IgM or negative rubella serology, 26 patients with other acute infections and 5 patients with rheumatoid factor-positivity were analyzed for RV-specific IgM by ELISA coated with whole-virus lysate or native glycoprotein, followed by determination of IgG avidity and E2-specific IgG using lysate-coated ELISA and non-reducing immunoblot. Compared to a reference μ-capture IgM ELISA, the sensitivity for diagnosing recent rubella infection/vaccination was 90.0% and 100% for the lysate-based and glycoprotein-based IgM ELISA, respectively. With respect to women with past RV infections or negative histories of RV infection/vaccination, both assays were 97.5-100% specific, whereas for patients with other acute infections the glycoprotein substrate provided a specificity of 92.3% compared to only 80.8% using whole-virus antigen. Analyzing anti-RV IgG avidity and anti-E2 IgG reactivity allowed the time point of primary infection to be determined unambiguously in >86% of samples. In conclusion, using RV glycoprotein antigen improves the specificity of indirect IgM ELISA. In cases of RV-specific IgM reactivity, recent primary rubella infection can be confirmed or excluded efficiently by specific IgG avidity and immunoblot analysis.     

Detection of rubella virus-specific immunoglobulin M antibodies with a baculovirus-expressed E1 protein.

The structural proteins of rubella virus (RV) were expressed in insect cells by using the baculovirus expression vector system. The recombinant E1 envelope glycoprotein was purified by immunoaffinity chromatography and used to detect RV-specific immunoglobulin M antibodies in a time-resolved fluoroimmunoassay. Correlation analysis between the reactivities of antibodies against this recombinant E1 and the reactivities against authentic RV antigen shows that purified E1 can detect RV antibodies of the immunoglobulin M type.     

Immunoaffinity purification of baculovirus-expressed rubella virus E1 for diagnostic purposes.

Three monoclonal antibodies, termed 4E10, 1E11:10, and 2D9:1, were generated against rubella virus. Immunoblot analysis with purified authentic rubella virus or recombinant baculovirus-expressed rubella virus structural proteins E1, E2, and C demonstrated that they were directed against the E1 envelope glycoprotein of the rubella virus particle. By using the yeast Ty virus-like particle system, it was possible to map the binding site of 1E11:10 within amino acids 236 to 286 of the E1 protein and the binding sites of 2D9:1 and 4E10 outside this region. Immunoaffinity purification with these monoclonal antibodies made it evident that they are useful for obtaining large quantities of pure baculovirus-expressed rubella virus envelope protein E1. The diagnostic potential of this immunoaffinity-purified recombinant rubella virus E1 protein compared with that of authentic rubella virus is demonstrated.     

Expression of recombinant E2 and C proteins of rubella virus in insect cells

Summary We have constructed a recombinant baculovirus expressing the rubella virus E2 (42–45 KDa) and C (34 KDa) proteins. Sf9 cells infected with recombinant virus were able to synthesize and process the two proteins coded by a unique precursor gene. By immunoblot and immunoprecipitation analysis with polyclonal and monoclonal antibodies, a precursor polyprotein (66 KDa) and two other proteins migrating with an apparent molecular weight of 42 KDa and 36 KDa were recognized as E2 glycoprotein and C protein, respectively. The recombinant E2 protein appeared to be glycosylated since it was susceptible to tunicamycin. The results indicate that the RV polyprotein coding for E2 and C is expressed and proteolytically cleaved in insect cells. This baculovirus expression system provides a useful alternative approach for the production of rubella virus antigens and should allow the purification of large quantities of the RV proteins for further biochemical and immunological studies.     

Significance of avidity and immunoblot analysis for rubella IgM-positive serum samples in pregnant women

A total of 512 IgM-positive serum samples from 449 pregnant women referred by microbiologists and medical laboratories for additional testing and final interpretation were collected over a 3-year period. Employment of an IgM capture enzyme immunoassay (EIA) confirmed only 31% of the initial EIA-IgM-positive samples. In order to discriminate acute rubella virus infections, which are associated with increased risk of fetal infection and embryopathy, from persistent or non-specific IgM, IgG avidity index and the presence of IgG with specificity to rubella virus E2 glycoprotein was determined by Western immunoblot. In only six patients (1.3%), a primary infection with rubella virus was diagnosed on the basis of IgM positivity, low avidity IgG, absence of E2-specific IgG in immunoblot concordant with clinical findings as well as consistent changes in follow-up samples. The serological results were not compatible with rubella re-infection. The infection status in 14 patients (3.1%) remained inconclusive even when both avidity assay and immunoblot were used, while restriction to either test did not allow a conclusive interpretation in 11.6% of patients. The use of both assays is clearly better, and therefore, recommended in IgM-positive samples.     

Detection of rubella virus-specific immunoglobulin G (IgG), IgM, and IgA antibodies by immunoblot assays.

Immunoblot (IB) assays were developed for detection of rubella virus (RV)-specific immunoglobulin G (IgG), IgM, and IgA antibodies in human serum following natural infection or immunization. IB assays performed under nonreducing conditions were compared with those performed under reducing conditions and with immunoprecipitation assays. Significant loss of antigenicity (greater than 90%) of RV E1 and E2 proteins was observed when IB assays were performed in the presence of 2-mercaptoethanol as compared with assays under nonreducing conditions. In contrast, the antigenicity of RV capsid protein was not influenced by reducing agents. Sensitivity of IB for RV-specific IgG antibodies was determined to be 0.01 IU/ml under nonreducing conditions. In the determination of RV-specific IgM and IgA antibodies by IB, pretreatment of serum with protein G to remove competing high-affinity RV-specific IgG or rheumatoid factor significantly improved assay sensitivity. IB assays were observed to be superior to immunoprecipitation assays in their ability to better define the specificities of RV-specific antibodies and to detect antibodies of all immunoglobulin classes. However, the conformational sensitivity of RV protein antigenicity should be an important consideration in the interpretation of RV-specific antibodies by IB assays.     

Maturation of immunoglobulin G avidity after rubella vaccination studied by an enzyme linked immunosorbent assay (Avidity-ELISA) and by haemolysis typing

Two tests were introduced recently for assessment of the avidity of rubella immunoglobulin antibodies. In the quantitative test—avidity—enzyme linked immunosorbent assay (ELISA)—IgG antibodies obtained from individuals shortly after primary infection with rubella virus are distinguished from those with past immunity by their antigen-elution characteristics. This method uses agents that disrupt hydrophobic bonds in proteins [Kamoun PP (1988): Denaturation of globular proteins by urea: Breakdown of hydrophobic bonds? Trends in Biological Sciences 13:424–425.]. In the semiquantitative, presumptive test—haemolysis typing—the lowavidity rubella-IgG antibodies are distinguished from the high-avidity antibodies by the quality of their haemolytic zones in a radial haemolysis test. In the present study, both tests were applied to sera taken before and after vaccination with two different strains (Cendehill or RA 27/3) of live attenuated rubella virus. It was found that after vaccination of previously nonimmune subjects, IgG synthesized during the first 2 months had a very low avidity; IgG avidity increased dramatically during the subsequent 4 months and less markedly between 6 and 12 months after vaccination. On the contrary, the initially high IgG avidity of previous immune vaccinees remained at an elevated level postvaccination. These results provide a basis for identification of recent primary rubella virus infections, or vaccination reactions, by the avidity of specific IgG and also for their separation from rubella reinfections.     

Humoral immune response to primary rubella virus infection

An assay capable of distinguishing between the immune response generated by recent exposure to rubella virus and the immune response existing as a result of past exposure or immunization is required for the diagnosis of primary rubella virus infection, especially in pregnant women. Avidity assays, which are based on the premise that chaotropic agents can be used to selectively dissociate the low-avidity antibodies generated early in the course of infection, have become routinely used in an effort to accomplish this. We have thoroughly investigated the immunological basis of an avidity assay using a viral lysate-based assay and an enzyme-linked immunosorbent assay (ELISA) based on a peptide analogue of the putative immunodominant region of the E1 glycoprotein (E1208-239). The relative affinities of the antibodies directed against E1208-239 were measured by surface plasmon resonance and were found to correlate well with the avidity index calculated from the ELISA results. We found that the immune response generated during primary rubella virus infection consists of an initial low-affinity peak of immunoglobulin M (IgM) reactivity followed by transient peaks of low-avidity IgG3 and IgA reactivity. The predominant response is an IgG1 response which increases in concentration and affinity progressively over the course of infection. Incubation with the chaotropic agent used in the avidity assay abolished the detection of the early low-affinity peaks of IgM, IgA, and IgG3 reactivity while leaving the high-affinity IgG1 response relatively unaffected. The present study supported the premise that avidity assays based on appropriate antigens can be useful to confirm primary rubella virus infection.     

Antibody response to wild rubella virus structural proteins following immunization with RA 27/3 live attenuated vaccine

Summary Antibody response to individual structural proteins (E 1, E 2, and C) of the M-33 wild rubella virus strain was assayed by an immunoblot technique in 12 girls, following immunization with RA 27/3 live attenuated rubella vaccine. Of the 12 immunized subjects, before vaccination 9 had no demonstrable rubella specific antibodies while the remaining 3 had a low level of rubella specific antibodies, reacting only with the E 1 protein. At one month after vaccination all the immunized subjects presented anti-E 1, anti-E 2, and anti-C specific antibodies. However, at 1–2 weeks after vaccination the 9 girls who were seronegative before immunization still had no detectable antibodies to any of the rubella virus structural proteins, while the 3 subjects whose preimmunization sera had reacted with the E 1 protein presented an accelerated immune response, showing anti-E 2 and anti-C specific antibodies and a more intensely marked anti-E 1 specific band.     

Avidity maturation following immunization with two human cytomegalovirus (CMV) vaccines: a live attenuated vaccine (Towne) and a recombinant glycoprotein vaccine (gB/MF59)

Two human cytomegalovirus (CMV) vaccines have been previously evaluated for their immunogenicity: a recombinant gB/MF59 vaccine and an attenuated strain of CMV (Towne). In healthy adults, we measured the antibody avidity maturation indices that occurred after vaccination with each. For Towne, administered as a single dose, the rise in IgG antibody avidity to CMV glycoprotein gB occurred slowly and continued for 24 months post-immunization. For gB/MF59, administered as two priming doses and a booster dose given at 6 months, the booster rapidly induced IgG antibodies to gB whose avidity was maximal at 7 months after the initial priming dose. Both vaccines induced antibody levels and avidity maturation indices that equaled those induced by wild-type virus suggesting that both vaccines may be effective in controlling CMV infections.     

Sex differences in antibody- and cell-mediated immune response to rubella re-immunisation

Antibody (AMI) and cell-mediated (CMI) immunity to rubella virus (RV) were evaluated in healthy adolescent males (n = 11) and females (n = 28) undergoing routine reimmunisation with RA27/3 strain RV as a component of measles-mumps-rubella (MMR) vaccine. Blood samples were collected just before and at 2, 4 and 10 weeks after MMR. While there were no sex differences before MMR and at week 10 after vaccination, levels of specific IgG determined by whole RV enzyme immunoassay were found to be significantly higher in males at weeks 2 and 4, suggesting brisker onset of recall AMI. Analysis of RV protein-specific IgG by immunoblot assays also revealed that while there were no notable sex differences in the distribution of E1-specific antibodies, more males produced E2-specific antibodies whereas more females produced C-specific antibodies after immunisation. Analysis of CMI with whole inactivated RV and a panel of RV synthetic peptides in lymphocyte proliferation assays revealed a brisker onset of CMI in males which paralleled that observed for AMI. The numbers of RV antigens recognised by males were significantly higher at weeks 2 and 4. Also, mean and median stimulation indices measured at weeks 2 and 4 for certain peptides, including two known to contain overlapping antibody neutralisation domains and T-cell epitopes, E1(213-239) and E1(254-285), were also found to be significantly higher in male subjects. These observations suggest that there are hormonal influences on RV-specific immunity which might result in differential handling of RV and, hence, may partially explain why females are more predisposed to adverse outcomes of rubella infection and immunisation.     

Cellular and humoral immune responses to rubella virus structural proteins E1, E2, and C.

Better understanding of cell-mediated immune responses to rubella virus would provide the basis for the development of safe and effective vaccines against rubella and would aid in analysis of the pathophysiology of congenital rubella syndrome. We have expressed individual rubella virus structural proteins, E1, E2 and C, via vaccinia virus recombinants. Using the expressed recombinant proteins as antigens, we were able to demonstrate antigen-specific lymphocyte proliferative responses in control individuals and individuals with congenital rubella syndrome. Among the two human groups studied, E1 glycoprotein proved to be a better immunogen than E2 or C. For the control individuals, significant differences in proliferative responses to the structural proteins E1, E2, and C were observed. These differences were not significant in individuals with congenital rubella syndrome. In parallel to the lymphoproliferative responses, immunoglobulin G responses were also found directed mainly to the E1 glycoprotein. These results suggest that E1 may be the most important rubella virus antigen to study in determining the domains required for constructing subunit vaccines against rubella.     

Immunogenicity properties of authentic and heterologously synthesized structural protein VP2 of infectious pancreatic necrosis virus

The sole coat protein VP2 of infectious pancreatic necrosis virus (IPNV) was isolated and purified from intact virions, propagated in CHSE-214 cells. Likewise was the full-length VP2 protein isolated and purified upon cloning and expression of the corresponding complete gene in E. coli. The two purified proteins of different synthetic origins carrying identical primary structures were utilized in an immunization program using a rabbit model. Sera obtained against both immunogens react equally well with authentic and recombinant VP2 in Western blots and ELISAs. Also, the total net binding forces as determined by avidity index (AI) calculations were high and of similar stature, exceeding 80. An IPNV infection of susceptible and permissive CHSE-214 cells could only be neutralized by IgG preparations obtained from rabbits immunized with authentic VP2. Only such antibodies were able to aggregate and sediment radiolabeled virions in glycerol gradients upon rate zonal centrifugations. The presence of sugar moieties on the authentic protein is suggested to be of pivotal importance in eliciting an immune response capable of preventing infection in cell cultures in vitro.     

Diagnostic potential of baculovirus-expressed rubella virus envelope proteins.

The envelope glycoproteins E1 and E2 of rubella virus were abundantly expressed in Spodoptera frugiperda Sf9 insect cells by using a baculovirus expression vector. The recombinant protein products were purified by immunoaffinity chromatography and characterized by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblotting, and enzyme immunoassay (EIA). The purified recombinant antigen consisted of the envelope polypeptides, corresponding to the viral E1 and E2 proteins, and a polyprotein precursor (molecular mass, 90 to 95 kDa). The antigen was reactive with human convalescent-phase sera in immunoblot analysis, and the reactivity correlated well (r = 0.861) with that of a whole-virus antigen when tested by EIA by using a total of 106 rubella virus immunoglobulin G-positive and -negative serum specimens. When the sera from patients with recent rubella virus infection were tested with the recombinant glycoproteins by EIA, the correlation was not as close (r = 0.690). However, all of the 26 serum specimens were reactive with the recombinant antigen. The results demonstrate that these bioengineered antigens have a potential for use in routine diagnostic assays of rubella virus immunity and recent infection.     

The association of rubella virus in congenital cataract - a hospital-based study in India.

BACKGROUND: The association of rubella virus (RV) with congenital cataract has been well established. Since the data on association of RV with congenital cataract in India are scanty, a study was done based on virus isolation from lens aspirates in patients undergoing therapeutic lensectomy and serology. OBJECTIVE: To determine the incidence of the association of rubella virus with congenital cataract. STUDY DESIGN: The lens aspirates collected during the 9-year period (from 1990 to 1998), from 70 children up to 12 years of age with congenital cataract were processed for the isolation of rubella virus by conventional viral isolation culture method using BHK-21 and Vero cell lines. Identification of the virus was confirmed by immunofluorescence using human anti-rubella virus specific hyperimmune serum. Serum samples were collected from 55 out of these 70 children and the presence of antibodies to RV was detected by ELISA test. RESULTS: RV was isolated from lens aspirates in seven (10%) out of the 70 children with congenital cataract. Of the 55 sera tested, 22 had both anti-rubella IgM and IgG antibodies, in 13 only anti-RV IgG antibodies, in seven only IgM antibodies and the rest of the 13 samples did not have detectable levels of rubella antibodies. Among the children who had IgM antibodies, 12 (24.5%) were below the age of 6 months. CONCLUSION: It can be concluded based on virus isolation that 10% of patients with congenital cataract were due to rubella infection and the detection of 24.5% anti-RV IgM antibodies in children below 6 months old shows the possible association of rubella virus with congenital cataract.     

Vaccinia-vectored expression of the rubella virus structural proteins and characterization of the E1 and E2 glycosidic linkages.

The maturation of rubella virus (RV) glycoprotein E2 from the single intracellular species (E2i; MW = 40 kDa) to the heterodisperse virion species (E2v; MW = 42 to 47 kDa), was studied by pulse-chase radiolabeling in Vero cells infected with RV or with recombinant vaccinia viruses (VVs) which express the entire RV structural protein open reading frame (VV-CE2E1) or glycoprotein E2 independently (VV-E2). The RV proteins expressed by the recombinant VVs comigrated with authentic RV intracellular proteins. In pulse-chase experiments, performed in both RV- and VV-CE2E1-infected cells, the amount of pulse-labeled E2i was substantially reduced during a 3- to 4-hr chase; during the same chase the amount of pulse-labeled E1 and C did not change. The concomitant appearance of the E2v forms was not observed. In contrast, in VV-E2-infected cells, no reduction in the amount of E2i occurred after as long as a 10-hr chase. Western blots using anti-E2 monoclonal antibodies showed that E2i was the predominant E2 species in cells infected with RV, VV-CE2E1, and VV-E2. However, minor amounts of three discrete species which comigrated within the extent of the E2v smear were also detected in cells infected with all three viruses, indicating that some degree of intracellular processing to E2v did occur. The disappearance of E2i during pulse-chase radiolabeling without the concomitant appearance of detectable E2v and the predominance of this labile form under steady-state conditions as revealed by Western blot analysis suggested that E2i was selectively turned over in both RV- and VV-CE2E1-infected cells. Such turnover was not apparent in VV-E2-infected cells, indicating that association with C and E1 was necessary for turnover to occur. Endoglycosidase digestion experiments and glycan differentiation assays revealed that E2v contained O-linked glycans. The presence of O-glycans on E2v accounted for part of the difference in size between E2v and E2i. Both virion E1 and E2 were found to contain high-mannose, hybrid-type, and complex-type N-glycans. Heterogeneity existed in the extent of processing of these glycans among individual E1 and E2 molecules.     

Antibody response to the rubella virus structural proteins in infants with the congenital rubella syndrome

Forty-five serum samples from 31 newborns and infants with the congenital rubella syndrome (CRS) were tested by immunoprecipitation to determine their antibody spectra to each of the structural proteins of rubella virus. Most sera (37/45) contained little or no E2 protein-specific antibody, but some (6/45) precipitated a greater amount of the E2 glycoprotein than the E1 glycoprotein. The relative E1/E2 ratio was found to decrease with time when serial serum samples from the same patient were tested. No correlation between the IgG class hemagglutination inhibition antibody titers and the E1/E2 ratio could be demonstrated. However, in some serum specimens relatively high neutralizing antibody titers were correlated with immunoprecipitation of the E2 glycopolypeptide. None of the CRS sera reacted well with the C protein. The immunoprecipitation patterns found in CRS sera were qualitatively different from those observed in a series of 25 sera from young adults with conventional serologic evidence of rubella immunity following natural infection. All of the natural immune sera recognized each of the three structural polypeptides of rubella virus.