Author
VIDAL, E - Valencian Institute For Agricultural Research | |
Yokomi, Raymond - Ray | |
MORENO, A - Instituto De Agricultura | |
BERTOLINI, E - Valencian Institute For Agricultural Research | |
CAMBRA, M - Valencian Institute For Agricultural Research |
Submitted to: Phytopathology
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 8/16/2011 Publication Date: 2/12/2012 Citation: Vidal, E., Yokomi, R.K., Moreno, A., Bertolini, E., Cambra, M. 2012. Calculation of diagnostic parameters of advanced serological and molecular tissue-print methods for detection of Citrus tristeza virus. A model for other plant pathogens. Phytopathology. 102:114-121. Interpretive Summary: Citrus tristeza virus (CTV) is a graft-transmissible pathogen of citrus and is transmitted in nature by aphids. Quarantine and certification programs are typically enforced in registered citrus nurseries to maintain zero tolerance for this pathogen. Such regulations also pertain to severe strains of the pathogen which may be introduced from abroad or from another region with budwood or trees moved to a new area. These programs require sensitive, specific and reliable pathogen-detection methods. The purpose of this research was to evaluate a tissue-print (TP) Taqman-based real-time reverse transcription (RT)-PCR assay recently developed for CTV diagnosis in comparison with TP-enzyme-linked immunosorbent assay (ELISA) for efficacy of CTV detection. The latter procedure was used as the gold standard as it is a certified and accepted technique for detection of CTV. The initial tests analyzed 1,395 samples from healthy and CTV-infected plants from a nursery and mature trees in commercial groves. Kappa statistics was used to 1) measure the independence between techniques to make yes/no decision about infection and 2) to quantify the level of agreement between methods. The agreement between techniques was significant (P <0.05) with Cohen’s kappa index of 0.77 ± 0.03. Since kappa is not “chance corrected”, a second set of data generated from a nursery plot of 658 Citrus macrophylla plants were evaluated to measure sensitivity and specificity and the likelihood ratios of measuring the true health status of a tree. The likelihood ratio is the probability of a specific test to be correct and was used to calculate post-test odds from pre-test odds to assess risk of abnormality and to adjust for various prior probabilities due to chance. For TP-ELISA, a sensitivity of 0.8015 with a specificity of 0.9962 and a positive to negative likelihood ratio of 216.42 to 0.199 was calculated; whereas TP-real-time RT-PCR had an observed sensitivity of 0.982 with a specificity of 0.8519 and a positive to negative likelihood ratio of 6.63 to 0.021, respectively. These results showed TP real-time-RT-PCR was more sensitive than TP-ELISA but less specific, which can lead to more false positives or negatives. Nevertheless, these data clearly show that TP-real time RT-PCR can be used reliably for detection of CTV especially when the pathogen is known to be present and consequences of a false result are less critical than in a situation where the disease agent is thought to be absent (as in the maintenance of an eradication program). Technical Abstract: Citrus tristeza virus (CTV) is one of the most important virus diseases which affect citrus. Control of CTV in Spain and central California is achieved by planting virus-free citrus on CTV-tolerant or -resistant rootstocks. Quarantine and certification programs remain essential to avoid importation and propagation of severe strains of CTV. Citrus nurseries in Spain and U.S. maintain a zero tolerance for CTV. These programs require sensitive, specific and reliable pathogen-detection methods. A tissue-print (TP) real-time reverse transcription (RT)-PCR assay has been recently developed for CTV diagnosis and its effectiveness was compared with TP-enzyme-linked immunosorbent assay (ELISA), a certified technique for detection of CTV. The initial test analyzed 1,395 samples from healthy and CTV-infected plants in nursery and mature trees in commercial groves. The total agreement between techniques was significant (P <0.05) with Cohen’s kappa index of 0.77 ± 0.03. The reliability of each technique which included sensitivity, specificity and likelihood ratios of tree health status was tested in a second test involving a nursery plot of 658 Citrus macrophylla plants. For TP-ELISA, a sensitivity of 0.8015 with a specificity of 0.9962 and a positive to negative likelihood ratio of 216.42 to 0.199 was calculated; whereas TP-real-time RT-PCR had an observed sensitivity of 0.982 with a specificity of 0.8519 and a positive to negative likelihood ratio of 6.63 to 0.021, respectively. These results showed TP real-time-RT-PCR was more sensitive than TP-ELISA and validates it for use to detect CTV for diagnostic purposes. |