Development and Evaluation of a Polymerase Chain

The primary objective of this research project was to develop and evaluate a more sensitive diagnostic assay for detecting Xylella fastidiosa (X.f), the bacterium that causes Pierce’s Disease (PD) of grapevines. The development of more sensitive diagnostic assays would greatly facilitate the identification of alternate plant reservoirs of PD and grape germplasm that is more resistant or tolerant of X.f. In 1995 we evaluated the utility of a new and very sensitive diagnostic test, called the polymerase chain reaction (PCR), for detecting X.f. in grapevines and other woody plants. PCR uses a special thermostable enzyme called Taq polymerase to amplify a specific fragment of X.f. DNA. The presence of the amplified DNA indicates that the sample was infected with the X.f. bacterium. We began our evaluation by testing the specificity and sensitivity of two different sets of PCR primers which direct the Taq enzyme to amplify the X.f DNA. Results showed that both sets of PCR primers, which amplify different fragments of X.f DNA, worked equally well in detecting X.f. in grapevines. PCR detected the X.f bacterium in dormant canes in March, however the percentage of positive detections was low. The frequency of positive detections increased in diseased canes as the season progressed and the ability to consistently detect X.f. by PCR, ELISA and cultural methods directly correlated with the development of leaf symptoms. PCR consistently provided the most sensitive detection of X.f in diseased grapevines, however a low percentage of false-positive test results occasionally occurred when testing cane scrapings. PCR also detected X.f. in other woody plants, such as red willow, maple, elderberry and wild grape, woody plants that can be difficult to analyze by ELISA. The most surprising, and potentially significant, result of testing diseased grapevines over the growing season was that no X.f bacteria were detected in emerging green shoots until the middle of June. This result indicates that Xf colonizes growing tissues quite slowly and that infected grapevines are probably not a significant reservoir of the pathogen until late July or August. This observation also suggests, but doesn’t prove, that dormant buds on diseased vines probably contain few viable bacteria and, theoretically, the disease should be difficult to transmit from dormant buds collected in late winter. Experiments are now in progress to verify this hypothesis. It was also determined, using a DNA fingerprinting technique, that X.f. strains collected from diseased grapevines located throughout California were genetically quite similar to eachother and to wild grape infected with X.f. More importantly, all of these geographically diverse strains were consistently detected by both sets of PCR primers. We believe that PCR could provide more reliable pathogen detection if methods were developed to sample larger portions of the vine, without having to do multiple PCR tests on the same vine. We are now evaluating the usefulness of a technique called immunocapture PCR, which uses X.f. -specific antibodies to trap X.f. bacteria, to test large-size samples. The relative sensitivity and reliability of immunocapture PCR, standard PCR, ELISA and cultural methods will be evaluated during 1996.