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

Preliminary multiplex microarray IgG immunoassay for the diagnosis of toxoplasmosis and rubella

BACKGROUND

During pregnancy, toxoplasmosis and rubella can cause serious damage to the mother and the foetus through vertical transmission. Early diagnosis enables implementation of health measures aimed at preventing vertical transmission and minimising damage caused by these diseases.

OBJECTIVE

Here, we report the development of a multiplex assay for simultaneous detection of IgG antibodies produced during toxoplasmosis and rubella infection.

METHODS

This assay is based on xMap technology. Initially, by singleplex assays, we evaluated the following antigens: one Toxoplasma gondii lysate; two antigenic extracts of T. gondii (TOX8131 and TOX8122); fragments of T. gondii antigens [SAG-1 (amino acids 45-198), GRA-7 (24-100), GRA-1 (57-149), ROP-4, and MIC-3 (234-306)]; two chimeric antigens composed of fragments of SAG-1, GRA-7, and P35 (CTOX and CTOXH); and fragments of Rubella virus antigens [E-1 (157-176, 213-239, 374-390), E-2 (31-105), and C (1-123)].

FINDINGS

A multiplex assay to simultaneously diagnose toxoplasmosis and rubella was designed with the best-performing antigens in singleplex and multiplex assays, which included CTOXH, T. gondii lysate, TOX8131, E-1, and E-2. The multiplex assay showed 100% sensitivity and specificity for anti-T. gondii IgG detection and 95.6% sensitivity and 100% specificity for anti-R. virus IgG detection.

MAIN CONCLUSIONS

We found that, despite the difficulties related to developing multiplex systems, different types of antigens (extracts and recombinant proteins) can be used to develop high-performance diagnostic tests. The assay developed is suitable to screen for prior T. gondii and R. virus infections, because it is a rapid, high-throughput, low-cost alternative to the current standard diagnostic tools, which require multiple individual tests.

Expression and characterization of rubella virus glycoprotein E1 in yeast cells

OBJECTIVES

To express E1 glycoprotein of rubella virus (RuV) strain JR23 in yeast and develop a diagnostic assay using expressed E1 protein as coating antigen in comparison with other diagnostic assays.

METHODS

cDNA of E1 open reading frame of RuV was PCR amplified using plasmid pMD18-T-E1 as template and cloned into plasmid pBluscriptII SK+. After being confirmed by PCR, restriction endonuclease digestion and sequencing, pBluscriptII SK(+)-E1 plasmid DNA was digested by restriction endonuclease EcoR I and Xba I, and a fragment of 1.5 kb was isolated and cloned into a yeast expression pGAPZ(alpha)A, resulting in pGAPZ(alpha)A-E1. After confirmation by sequencing, pGAPZ(alpha)A- E1 was transformed into yeast GS115 cells with LiCl method. E1 protein expression in GS115 was analyzed by SDS-PAGE and Western blot. An indirect ELISA was developed using the recombinant E1 protein as coating antigen for detecting RuV E1 antibodies in 90 serum samples. To compare the specificity, sensitivity and reproducibility of the assay with other methods, the same serum samples were also assayed by RuV culture medium as coating antigen (Jingmei kit and German RECI kit). Statistical analyses were performed to compare the differences among these methods and to determine which coating antigen source, the recombinant E1 protein or RuV-infected culture medium, is more suitable for the assay.

RESULTS

A fragment of 1.5 kb, corresponding to the full open reading frame of E1, was PCR amplified and cloned in yeast expression vector. The clone was confirmed by restriction digestion, PCR and sequencing. E1 as a secretive protein was successfully expressed by GS115. Its molecular weight was about 58 kDa. SDS-PAGE showed that the recombinant protein was expressed efficiently and constantly in Pichia pastoris GS115 cells. The expression level reached a peak 48 h after culturing and stabilized thereafter. E1 protein was detected in both supernatant and cells. Western blot showed that the secretive E1 protein in the supernatants could react with both the anti-RuV-positive serum and a monoclonal antibody against E1. However, E1 protein derived from cells could only react with the anti-RuV-positive serum, polyclonal antibody, but not the monoclonal antibody. Compared with the German RECI kit, the sensitivity, specificity, and accordance rate of the assay using recombinant E1 protein as coating antigen were 67.11, 71.43 and 67.78%, respectively, while those of the assay using RuV-infected culture medium as coating antigen were 50, 78.57 and 54.44%, respectively. Compared with the German RECI kit, the sensitivity, specificity, and accordance rate of the ELISA assay using the Jingmei kit were 84.71, 71.43 and 82.22%, respectively. The data indicated that recombinant E1 protein derived from the yeast expression system can serve as a better source than RuV-infected cell medium as coating antigen for detecting antibodies against RuV in the indirect ELISA assay. Statistical analysis of the data generated from two independent experiments using recombinant E1 protein as coating antigen indicated that the assay was very consistent with no statistically significant difference between the two experiments (p > 0.05). 76 out of 90 serum samples were detected positive using the German RECI kit, while 68, 55 and 41 samples were positive using the Jingmei kit, recombinant E1 and RuV-infected cell medium, respectively. Statistical analyses indicated that the positive rates were significantly different among all four assays (p < 0.05) except for one pair (German RECI kit and the Jingmei kit: p > 0.05). Comparing the positive rates obtained from the assay using recombinant E1 and that using RuV-infected cell medium, it seems that the recombinant E1 protein is a better source than RuV culture medium as coating antigen in the indirect ELISA assay for detection of RuV antibody.

CONCLUSIONS

The recombinant yeast expression vector of RuV E1 glycoprotein was constructed successfully. The E1 protein as a secretive protein was successfully expressed by GS115 and maintained its antigenicity very well. As coating antigen, the recombinant E1 protein served a better source than RuV culture medium in the indirect ELISA method for the detection of RuV antibody.

Preparation of immunoblot test stripes from a Rubella virus-like particles dye crystal complex as antigen

Stably transfected Chinese hamster ovary (CHO24S) cells were the source for Rubella virus-like particles (RVLP) containing all structural proteins (E1, E2, C and their dimers). RVLP are secreted from the CHO24S cells into the medium and the time-point for collecting the medium with the highest yield of >100 kDa proteins (with 17 mg protein from 10 ml cell culture supernatant) was after 2 days of incubation. Different methods for RVLP isolation from the cell culture supernatants were assessed by SDS-PAGE and Western blotting (using sera positive or negative for Rubella virus (RV)-specific antibodies or an anti-E1 monoclonal antibody). A combination of membrane filtration with a rapid, novel gradient ultracentrifugation step (using Coomassie brilliant blue G crystals as adsorbens for RVLP that facilitated virus isolation) was the most suitable technique. 132 RV-positive human sera (RV IgG > 20 IU/ml by commercial ELISA) were tested by our "self made" immunoblot test stripes (using RVLP adsorbed to dye crystals as antigen) for the presence or absence of antibodies specific for RV structural proteins. 57.6% of these sera had antibodies against E1, E2 and C, 31% against E1 and C, and 1.5% against E1 only, whereas 3.8% had no RV specific antibodies and only 6.0% were equivocal which demonstrated that these "self made" test stripes can reliably differentiate RV antibody specificities.

Mimicking rubella virus particles by using recombinant envelope glycoproteins and liposomes

The envelope glycoproteins E1 and E2 of rubella virus (RV) were engineered to display the FLAG epitope tag and a polyhistidine tag, at their amino and carboxy termini, respectively. These modified envelope proteins were produced in Sf9 insect cells utilizing baculovirus expression vectors, the E1 and E2 vectors giving rise to protein products of about 58 and 42 kDa, respectively. The recombinant proteins were purified by immobilized metal-ion affinity chromatography and reconstituted into liposomes via their hydrophobic transmembrane anchors. The liposomes were prepared by detergent dialysis in the presence of europium-DTPA chelate, enabling the subsequent measurement of the binding of the resultant proteoliposomes to the antibodies by time resolved fluorescence. RV mimicking proteoliposomes were recognized by antibodies specific for the E1 and E2 proteins, as well as the FLAG epitope tag. This type of virosome may prove useful for studies on the basic biological events of an RV infection or as diagnostic reagents.

Expression, processing, and immunogenicity of the structural proteins of Venezuelan equine encephalitis virus from recombinant baculovirus vectors

Recombinant baculoviruses expressing the structural proteins of Venezuelan equine encephalitis virus (VEE) have been constructed and the intracellular processing, antigenicity, and immunogenicity of the expression products have been assessed. Baculoviruses expressing the entire structural protein region (C-E3-E2-6K-E1), or the complete glycoprotein region (E3-E2-6K-E1), generated products in Sf9 cells that were accurately and completely processed, and resulted in mature proteins that were antigenically and electrophoretically indistinguishable from authentic viral proteins. These products were highly immunogenic in BALB/c mice, induced efficient VEE neutralizing responses, and protected these animals against challenge with virulent VEE. Expression of individual glycoprotein regions (E3-E2 and 6K-E1) generated products that were accurately but incompletely processed, and induced non-neutralizing antibodies that reacted more efficiently with denatured than native VEE proteins. Nonetheless, immunization with the 6K-E1 expression product provided complete protection against VEE challenge.

Baculovirus-mediated large-scale expression and purification of a polyhistidine-tagged rubella virus capsid protein

The capsid protein of rubella virus was produced in baculovirus-infected Spodoptera frugiperda insect cells, with a polyhistidine affinity tag at the carboxy terminus. The RV capsid recombinant protein was produced in a 10-liter bioreactor and purified, under nondenaturing conditions, using immobilized metal-ion affinity chromatography. Immunoblot analyses indicated that the purified recombinant protein was intact and migrated with the expected molecular weight. The final yield was 5 mg of purified protein per liter of cell culture. Surface plasmon resonance was used to investigate the antigenic potential of the histidine tagged capsid protein in an antigen-antibody interaction study. A specific interaction between the two proteins was shown. Our results suggest that this strategy should be useful in interaction studies of other virus-specific proteins and antibodies.