Comparative spatial spread overtime of Zucchini Yellow Mosaic Virus (ZYMV) and Watermelon Mosaic Virus (WMV) in fields of transgenic squash expressing the coat protein genes of ZYMV and WMV, and in fields of nontransgenic squash.

The spatial and temporal patterns of aphid-vectored spread of Zucchini Yellow Mosaic Virus (ZYMV) and Watermelon Mosaic Virus (WMV) were monitored over two consecutive years in plantings of nontransgenic and transgenic squash ZW-20H (commercial cv. Freedom II) and ZW-20B, both expressing the coat protein genes of ZYMV and WMV. All test plants were surrounded by nontransgenic plants that were mechanically inoculated with ZYMV or WMV, and served as primary virus source. Across all trials, none of the transgenic plants exhibited systemic symptoms upon infection by ZYMV and WMV but a few of them developed localized chlorotic dots and/or blotches, and had low mixed infection rates [4% (6 of 139) of ZW-20H and 9% (13 of 139) of ZW-20B], as shown by ELISA. Geostatistical analysis of ELISA positive transgenic plants indicated, (i) a lack of spatial relationship on spread of ZYMV and WMV for ZW-20H with flat omnidirectional experimental semivariograms that fitted poorly theoretical models, and (ii) some extent of spatial dependence on ZYMV spread for ZW-20B with a well structured experimental semivariogram that fitted poorly theoretical models during the first but not the second growing season. In contrast, a strong spatial dependence on spread of ZYMV and WMV was found for nontransgenic plants, which developed severe systemic symptoms, had prevalent mixed infection rates (62%, 86 of 139), and well-defined omnidirectional experimental semivariograms that fitted a spherical model. Geostatistical data were sustained by virus transmission experiments with Myzus persicae in screenhouses, showing that commercial transgenic squash ZW-20H alter the dynamics of ZYMV and WMV epidemics by preventing secondary plant-to-plant spread.