Weeds as Alternative Hosts of the Citrus, Coffee, and Plum Strains of Xylella fastidiosa in Brazil

Abstract

In Brazil, Xylella fastidiosa is present in citrus (Citrus sinensis), coffee (Coffea arabica), and plum (Prunus sp.) crops, causing the citrus variegated chlorosis (CVC), coffee leaf scorch (CLS), and plum leaf scald (PLS). Also present in these crops and infesting weeds, which ultimately could serve as sources of inoculum for the cultivated trees, are diverse populations of xylem-feeding leafhopper vectors. In order to assess host range of X. fastidiosa among weeds and to better understand their role in epidemics, field surveys, mechanical inoculations, and insect transmission tests were conducted. Polymerase chain reaction (PCR) and culture plating were used to detect the pathogen from plant tissues. X. fastidiosa was detected in 10 out of 23 species of the weed plants sampled in two citrus groves affected by CVC. None of the weed plants showed external symptoms. In the greenhouse, the average percentages of infection on plants mechanically inoculated with the CVC, CLS, and PLS strains of X. fastidiosa were, respectively, 25, 10, 0 in Medicago sativa; 70, 45, 20 in Echinochloa crus-galli; 45, 30, 0 in Brachiaria decumbens; 72, 70, 40 in Brachiaria plantaginea; 13, 10, 0 in Digitaria horizontalis; 31, 30, 0 in Solanum americanum; and 17, 0, 0 in Bidens pilosa. Symptoms were observed only in S. americanum and citrus and only when inoculated with the CVC strain. In insect transmission tests, the grass leafhopper Ferrariana trivittata was first caged on citrus plants showing CVC symptoms and then on healthy citrus and on the four most common weeds. No plants tested positive by PCR or culture, or showed symptoms for at least 4 months after inoculation. The amount of X. fastidiosa cells that may accumulate in weeds inoculated by leafhoppers is probably under insect acquisition thresholds, a factor that would limit their importance to the CVC epidemics, as studies on spatial distribution of diseased citrus trees over time indicate.