Evaluation of the immune status of dogs vaccinated against rabies by an enzyme-linked immunosorbent assay using crude preparations of insect cells infected with a recombinant baculovirus encoding the rabies virus glycoprotein gene

Abstract

Evaluation of the effectiveness of vaccination of animals against rabies is not routinely implemented. In cases where it is carried out, the rapid fluorescent focus inhibition test (RFFIT) or the fluorescent antibody virus neutralization (FAVN) test are the recommended tests. However, both of these tests require handling of live rabies virus (RABV), and are cumbersome to perform. In view of this, the enzyme-linked immunosorbent assay (ELISA) has been proposed as a surrogate test; however, availability of appropriate antigen is a major impediment for the development of ELISAs to detect anti-rabies antibodies. The most widely used antigen is the RABV glycoprotein (G) purified from cell culture-propagated virus, which requires a biosafety level 3 containment. The alternative is to use recombinantly expressed G, which needs to be to be properly glycosylated and folded to serve as the best antigen. The most suitable system for its production is the baculovirus expression system (BVES). However, purification of RABV G is challenging. We therefore tested partially purified preparations in the form of extracts of insect cells infected with baculovirus expressing RABV G, against sera from vaccinated dogs in an indirect ELISA. The results showed good concordance against RFFIT, with sensitivity and specificity of 90.48% and 80.00%, respectively. The system may be used for quick screening to determine the presence and an approximate level of antibodies, and can be modified to enable monitoring of mass dog vaccination programs, as well as to facilitate certification of dogs intended for international travel and transportation.

Fig 1. Expression of G by recombinant baculovirus.

A. Uninfected (Uninf.) or infected (Inf.) Sf-21 cell cultures were separated into cells and supernatants and subjected to PCR or RT-PCR using primers for full-length G gene. B. Cell extracts of uninfected (Mock) and infected (three different clones designated 1.4, 4.4 and 5.17) were subjected to SDS-PAGE and Coomassie blue staining. The position of protein markers is shown on the right. C. Extracts from control (untransfected) and transfection control (Cellfectin only) as well as cells infected with a control baculovirus (Bac-Null) or a baculovirus encoding RABV G (Bac-RABV-G) were subjected to immunoblotting using anti-His antibodies. The position of protein markers is shown on the right.

Fig 2. Purification of G expressed through baculovirus.

A. Extracts of Sf21 cells infected with baculovirus encoding the RABV G gene were passed over a Ni-NTA column, and fractions (numbered 1 through 8) were subjected to SDS-PAGE and Coomassie blue staining. The position of protein markers is shown on the right. B. Pooled fractions (2–4 and 5–8) were subjected to immunoblotting using anti-His antibodies. C. The pooled fraction 5–8 was subjected to immunoblotting using sera from dogs which were either naïve (non-immune) or vaccinated (immune) against rabies. The expected position of the G protein is shown on the left.

Fig 3. Non-linear regression analysis of RFFIT vs iELISA.

A. Dog sera with an antibody titre of 8.0 IU/mL as estimated by RFFIT were pooled and serially diluted, and plotted against positive predictive values obtained by iELISA for the test samples. B. Values in IU/mL for the iELISA, derived based on the non-linear regression analysis from Fig 3, were plotted against values in IU/mL of RFFIT.

Fig 4. Area under the curve for RFFIT vs iELISA.

The distribution of the frequency of the number of samples at each of the dilutions is plotted for RFFIT and iELISA. The dashed vertical line represents cut-off between positive and negative results at 0.5 IU/mL for RFFIT.