Development and validation of an immunoperoxidase antigen detection test for improved diagnosis of rabies in Indonesia

Abstract

Rabies continues to pose a significant threat to human and animal health in regions of Indonesia. Indonesia has an extensive network of veterinary diagnostic laboratories and the 8 National laboratories are equipped to undertake diagnostic testing for rabies using the commercially-procured direct fluorescent antibody test (FAT), which is considered the reference (gold standard) test. However, many of the Indonesian Provincial diagnostic laboratories do not have a fluorescence microscope required to undertake the FAT. Instead, certain Provincial laboratories continue to screen samples using a chemical stain-based test (Seller's stain test, SST). This test has low diagnostic sensitivity, with negative SST-tested samples being forwarded to the nearest National laboratory resulting in significant delays for completion of testing and considerable additional costs. This study sought to develop a cost-effective and diagnostically-accurate immunoperoxidase antigen detection (RIAD) test for rabies that can be readily and quickly performed by the resource-constrained Provincial laboratories. This would reduce the burden on the National laboratories and allow more rapid diagnoses and implementation of post-exposure prophylaxis. The RIAD test was evaluated using brain smears fixed with acetone or formalin and its performance was validated by comparison with established rabies diagnostic tests used in Indonesia, including the SST and FAT. A proficiency testing panel was distributed between Provincial laboratories to assess the reproducibility of the test. The performance of the RIAD test was improved by using acetone fixation of brain smears rather than formalin fixation such that it was of equivalent accuracy to that of the World Organisation for Animal Health (OIE)-recommended FAT, with both tests returning median diagnostic sensitivity and specificity values of 0.989 and 0.993, respectively. The RIAD test and FAT had higher diagnostic sensitivity than the SST (median = 0.562). Proficiency testing using a panel of 6 coded samples distributed to 16 laboratories showed that the RIAD test had good reproducibility with an overall agreement of 97%. This study describes the successful development, characterisation and use of a novel RIAD test and its fitness for purpose as a screening test for use in provincial Indonesian veterinary laboratories.

Fig 1. Flowchart showing current and proposed (dashed lines) laboratory testing regimen for diagnosis of rabies in Indonesia following competency assessment of individual provincial laboratories using the RIAD_acetone test.

Fig 2. Expression and purification of RABV NP and characterization of antiserum.

RABV NP inclusion bodies (IBs) (A) and gel-purified RABV NP (B) were resolved by SDS PAGE and stained with Coomassie blue. All lanes of gels stained with Coomassie blue were loaded with 10 μl of RABV NP in the dilutions or amounts indicated. Recombinant, gel-eluted His-tagged RABV NP was identified by immunoblotting with anti-His antibody (1:1,000) followed by sheep anti-mouse-HRP (1:2,000) (C). Sera from a pre- and post-immunized rabbit were diluted 1:10,000 and assessed for anti-RABV NP polyclonal antibody production by immunoblotting (D). All gels used for immunoblotting were loaded with 10 ng of RABV NP per well. Molecular mass markers were Mark 12 or See Blue Plus 2 (Invitrogen).

Fig 3. Comparison of RIAD_acetone test with FAT (inset) on canine brain smears infected with rabies virus (positive) or not (negative).

The presence of RABV antigen is indicated by brick red deposits (Positive) or green fluorescence (inset) in brain smears tested using the RIAD_acetone test or FAT, respectively. Magnification of the RIAD_acetone images are 63X and the FAT image is 20X. All smears were fixed in acetone.