Vol. 125 (2012): Proceedings of the Florida State Horticultural Society
Citrus

GC-MS Analysis of Secondary Metabolites in Leaves from Orange Trees Infected with HLB: A 9-Month Course Study

Shelley E Jones
University of Florida, IFAS, Citrus Research and Education Center, Lake Alfred, FL 33850
Faraj Hijaz
University of Florida, IFAS, Citrus Research and Education Center, Lake Alfred, FL 33850
Craig L Davis
University of Florida, IFAS, Citrus Research and Education Center, Lake Alfred, FL 33850
Svetlana Y Folimonova
University of Florida, IFAS, Citrus Research and Education Center, Lake Alfred, FL 33850
John A Manthey
USDA Horticultural Research Laboratory, Fort Pierce, FL 34945
Jose I Reyes-De-Corcuera
University of Florida, IFAS, Citrus Research and Education Center, Lake Alfred, FL 33850

Published 2012-12-01

Keywords

  • biomarker,
  • Citrus sinensis,
  • HLB,
  • leaf metabolites,
  • principal component analysis

Abstract

Citrus Huanglongbing (HLB) disease is considered the most serious threat to the citrus industry. We previously characterized non-targeted metabolite profiles of healthy, zinc deficient, and HLB-infected orange leaves from mature ‘Valencia’ trees located in commercial citrus groves. To identify potential biomarkers for HLB and to better understand the movement of HLB in planta, the current greenhouse study was performed using trees newly grafted with budwood from symptomatic HLB field trees. Trees of the same variety and age were used as controls and were inoculated with budwood from HLB-free trees. After extraction, derivatization, and GC-MS analysis of metabolites, principal component analysis (PCA) was performed on the normalized percentage peak area of each compound in the chromatogram. PCR data was compared with PCA results. Eighty-five percent of the young ‘Hamlin’ trees and 95% of ‘Valencia’ trees were infected by week 40 (confirmed by PCR). For ‘Valencia’, differences in GC-MS metabolite profiles could be detected beginning at week 16 (20 weeks post-leaf graft). These differences disappeared, however, as there was not a consistent trend throughout the study for any compound. In addition, 5 of 9 trees were already HLB positive when tested at week 10. For ‘Hamlin’, 3 of 9 trees tested PCR+ in week 10, but strong PCA clustering was not observed until week 32. Strong PCA clustering correlated well with visible symptom development. Of the 44 compounds identified by GC-MS as leaf metabolites, only malic acid, pyrollidine, and quinic acid were consistently found in both ‘Hamlin’ and ‘Valencia’ in sufficient quantity to be a potential biomarker.