Researchers in the U.S. Department of Agriculture’s Agricultural Research Service (USDA-ARS) have discovered that the best way to protect citrus crops from a devastating disease is to let a dog follow its nose.
Citrus growers in Florida and California, where citrus is a $3.35 billion industry, would benefit significantly from rapid and reliable detection of the bacterium that causes citrus greening, known as Huanglongbing (HLB). Both states are experiencing severe epidemics of HLB, which stunts the growth of trees and fruit and ultimately leads to tree death.
The gold standard diagnostic test for CLas (short for Candidatus Liberibacter asiaticus), the bacterium that causes HLB, is polymerase chain reaction (PCR)—a technique used in many COVID-19 tests. However, because CLas is not always uniformly distributed throughout a citrus tree, small samples of leaves collected for PCR testing may not contain any CLas even if the sampled tree is infected. Thus, PCR misses up to 80% of infected trees due to insufficient sampling. Additionally, detecting infected trees by PCR can take months to years following infection—delays that allow the disease to spread. In every geographic location where PCR has been used as the primary method for CLas detection, efforts to halt progression of the citrus greening outbreak have failed.
To address this conundrum, researchers from USDA-ARS studied the techniques used to train dogs to detect drugs and explosives, then adapted those techniques for the detection of citrus greening and transferred them to industry partner F1K9. USDA-ARS researchers developed new methods to capture and present scents to the dogs during training, and F1K9 developed new methods to focus canine attention onto these scent sources.
Citrus greening detector dogs have demonstrated 99% accuracy. This is possible because detector dogs simultaneously sample the scent from the entire tree (a process that takes only about two seconds), rather than the small samples used for PCR. In tests, canine detection occurred between two weeks and a few months after infection—much earlier in the disease’s progression than when relying on small samples and PCR.
Since being commercialized by F1K9 in 2018, the citrus greening detector dogs have surveyed more than 31,500 citrus trees in California and Florida. In locations where dog-identified trees were removed, disease incidence the following year decreased by 95% or more.
Disease modeling shows that reliance on sampling and PCR detection results in exponential spread of the disease and rapid loss of orchards and profitability, whereas detector dogs are capable of continuously protecting citrus orchards against citrus greening over many years.
Although this technology was developed and transferred to aid the U.S. citrus industry, it has the capacity to be applied to other agricultural commodities. The research team has already demonstrated efficient canine detection of citrus canker, plum pox and viral watermelon vine decline, and is extending the technology to wine grapes as well.