The control of tristeza quick decline (QD) of citrus is based on the use of rootstocks that are tolerant or resistant to the Citrus tristeza virus (CTV), but some of them show bio-agronomic limits. The application of cross-protection (CP) has been insufficiently explored. The present study examined the possibility of QD control by cross-protection (CP) following reports showing the dependence of the CP strategy on the close genetic relationships between the protective and challenging CTV isolates. Taking advantage of deep sequencing technologies, we located six naturally infected trees harboring no-seedling yellow (no-SY) and no QD decline (mild) VT isolates and used these for challenge inoculation with three QD VT isolates. Symptom monitoring showed that all six Sicilian mild no-SY isolates, based on their genomic relatedness and mild symptoms reactions, provide effective protection against the three severe local VT isolates. The differences between the six mild and three severe isolates were confined to just a few nucleotide variations conserved in eight positions of three CTV genes (p23, p33, and Orf1a). These results confirm that the superinfection exclusion (SIE mechanism) depends on close genetic relatedness between the protective and challenging severe VT strain isolates. Ten years of investigation suggest that CP could turn into an efficient strategy to contain CTV QD infections of sweet orange trees on SO rootstock.

The Citrus tristeza virus (CTV), a member of the Closterovirus genus, is considered a serious threat to citrus trees grafted onto sour orange (SO) rootstock. In the Mediterranean area, the most prevalent CTV strains are VT and T30. The VT strain includes both mild and severe isolates, some of them associated with seedling yellows (SY) syndrome. Mild CTV-VT isolates that do not induce SY symptoms (no-SY) show minor variations in their Orf1a, p23, and p33 genes, with a single nucleotide polymorphism at position 161 of the p23 gene. These isolates can repress superinfection with homologous severe isolates. The aim of this study was to investigate the mechanism of cross-protection by means of biological indexing, real-time RT-PCR high-resolution melting (HRM), and p23 gene amplicon sequencing. Four no-SY CTV-VT isolates were inoculated onto SO seedlings and Hamlin sweet orange trees grafted on SO. These plants were later challenged with two homologous CTV-VT SY isolates and remained asymptomatic. The biological evaluation of the infection process in superinfected plants was investigated via inoculation of the bark on SO seedlings that were also asymptomatic. A parallel HRM analysis of midvein RNA extracts revealed that the melting temperature (Tm) of the no-SY isolates was statistically lower than that of the SY isolates. The Tm values of RNAs extracts from superinfected plants were not statistically different from those of the no-SY isolates. This suggests that the SY isolates failed to establish infection or replicate in plants pre-inoculated with no-SY isolates. This blockage of replication resembles superinfection exclusion, with attractive perspectives to prevent SY damage in field applications.