An ELISA Using Synthetic Mycolic Acid-Based Antigens with DIVA Potential for Diagnosing Johne's Disease in Cattle

Simple Summary Johne's disease, caused by Mycobacterium avium subsp. paratuberculosis (MAP), is endemic in cattle herds in the UK and throughout the world. It is a wasting disease with a protracted incubation period, during which clinical signs are not apparent, and is responsible for considerable economic losses. Infection can be transmitted silently from animal to animal, and MAP can find its way into the food chain through the products of infected animals. Ante-mortem diagnosis of the disease is normally achieved through detection of the organism (by faecal culture or PCR) or serological tests, but agreement as to a positive diagnosis is often poor and the certainty of identifying all positive animals is low, particularly in early-stage infections. Mycobacterial cells contain high-molecular-weight lipids, mycolic acids (MAs), which are recognised by antibodies in active infections. Here, we show that a simple assay using different classes of single synthetic MAs can detect antibodies in serum from cattle with MAP infections and distinguish uninfected animals from animals positive both to faecal culture and a commercial serodiagnostic assay. It does so without interference from vaccination either for MAP (a Gudair vaccination) or for bovine TB (BCG vaccination). When the assay is optimised to distinguish negative samples from those positive only by faecal PCR, it identifies additional positives that are not seen using a commercial serodiagnostic assay; this suggests that, with further development, it may provide a method for detecting infections before the commercial serodiagnostic assay can.Abstract The problem: Ante-mortem diagnosis of Johne's disease, caused by Mycobacterium avium subsp. paratuberculosis (MAP), is normally achieved through faecal culture, PCR, or serological tests, but agreement as to which samples are positive for Johne's disease is often poor and sensitivities are low, particularly in early-stage infections. The potential solution: Mycobacterial cells contain very complex characteristic mixtures of mycolic acid derivatives that elicit antibodies during infection; this has been used to detect infections in humans. Here, we explore its application in providing an assay differentiating infected from vaccinated animals (DIVA assay) for Johne's disease in cattle. Method: Antibody responses to different classes of mycolic acid derivatives were measured using ELISA for serum from cattle positive for MAP by both faecal PCR and commercial serum ELISA, or just by PCR, and from animals from herds with no history of Johne's disease, bovine tuberculosis reactors, BCG-vaccinated, BCG-vaccinated and M. bovis-infected, and Gudair-vaccinated animals. Results: The best-performing antigens, ZAM295 and ST123-the latter a molecule present in the cells of MAP but not of Mycobacterium bovis-achieved a sensitivity of 75% and 62.5%, respectively, for serum from animals positive by both faecal PCR and a commercial MAP serum ELISA, at a specificity of 94% compared to 80 no-history negatives. Combining the results of separate assays with two antigens (ST123 and JRRR121) increased the sensitivity/specificity to 75/97.5%. At the same cut-offs, animals vaccinated with Gudair or BCG vaccines and bTB reactors showed a similar specificity. The specificity in BCG-vaccinated but M. bovis-infected animals dropped to 85%. Combining the results of two antigens gave a sensitivity/specificity of 37.5/97.5% for the full set of 80 PCR-positive samples, detecting 30 positives compared 16 for IDEXX. Conclusion: Serum ELISA using synthetic lipids distinguishes effectively between MAP-negative cattle samples and those positive by both PCR and a commercial MAP serodiagnostic, without interference by Gudair or BCG vaccination. It identified almost twice as many PCR positives as the commercial serodiagnostic, offering the possibility of earlier detection of infection.