Canine olfactory detection of a vectored phytobacterial pathogen, Liberibacter asiaticus, and intergration with disease control

Authors: Gottwald, Timothy; LUO, WEIQI - North Carolina State University; POSNY, DREW - North Carolina State University; Poole, Gavin; LOUWS, FRANK - North Carolina State University; McCollum, Thomas; Hartung, John; Bai, Jinhe; Duan, Ping; Taylor, Earl; DA GRAÇA, JOHN - Texas A&M University; SCHNEIDER, WILLIAM - F1k9; Polek, Marylou; HALL, DAVID - Retired ARS Employee

Submitted to: Proceedings of the National Academy of Sciences (PNAS)
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/26/2019
Publication Date: 2/3/2020
Citation: Gottwald, T.R., Luo, W., Posny, D., Poole, G.H., Louws, F., Mccollum, T.G., Hartung, J.S., Bai, J., Duan, Y., Taylor, E.L., Da Graça, J., Schneider, W., Polek, M., Hall, D. 2020. Canine olfactory detection of a vectored phytobacterial pathogen, Liberibacter asiaticus, and intergration with disease control. Proceedings of the National Academy of Sciences. 117(7)3492-3501. https://doi.org/10.1073/pnas.1914296117.
DOI: https://doi.org/10.1073/pnas.1914296117

Interpretive Summary: Dogs have been trained to detect HLB-infected trees [huaglongbing (HLB) is the most severe disease of citrus worldwide] with an accuracy of >99% and can be used as an early detection method to detect the disease in new areas and for commercial and residential survey. Detection by dogs is consistently months to years earlier than all other detection methods (molecular, serological, chemical) currently known. When deployed in commercial and residential settings, detection was real time requiring about 1-2 seconds per tree. A mathematical model was used to examine the effectiveness and cost/benefit of deployment of disease detection by dogs in a 60 acre orchard compared to 1) human visual detection, and 2) a common molecular detection method (PCR). Five years after first infection the orchard maintained using detector dogs was 95% healthy tree population and viable orchard production. Whereas, PCR detection resulted in removal of 84% of newly diseased trees and only ~9% healthy trees remaining. Visual detection resulted in the most inferior disease control resulting in ~47% disease detection necessitating tree removal, with the remaining trees all infected and no healthy trees remaining after 5 years. When linked with a rapid response, detection by dogs will be a powerful tool which will help minimize disease spread and optimize planting longevity.

Technical Abstract: Early detection and rapid response are crucial to avoid severe epidemics of exotic pathogens. However, most detection methods (molecular, serological, chemical) are logistically limited for large-scale survey of outbreaks due to intrinsic sampling issues and laboratory throughput. Evaluation of 20 canines for trained detection of a severe exotic phytobacterial arboreal pathogen demonstrated 0.9905 accuracy, 0.8570 sensitivity and 0.9960 specificity. Cryptic infections were detected within 2 weeks post-infection that remained subclinical visually and by molecular (qPCR) for 1-32 months. In field deployments, detection was real time; ~1-2 seconds per tree. Spatio-temporal epidemic simulations demonstrated that sustainable control of pathogen prevalence was possible when canine detection was followed by intervention, i.e., culling infected individuals, whereas, molecular (qPCR) and visual detection failed to suppress an exponential trajectory of infection .