HLB-affected Valencia fruit together with healthy fruit, were harvested in the middle of the harvest season, and peel oil extracted from flavedo tissues using methylene chloride/pentane solvents. Volatile compounds were analyzed by GC-MS. All HLB-affected fruit produced a lower volume of peel oil compared to healthy fruit, however, only green fruit showed substantially different volatile profiles from healthy fruit. The green fruit contained low concentrations of most of sesquiterpene hydrocarbons and derivatives, such as, nootkatone, valencene, α-selinene, 7-epi-α-selinene and caryophyllene oxide, some monoterpene hydrocarbons and derivatives with orange/citrus characteristics, such as, limonene, myrcene, α-pinene, neral, geranial, and α-terpineol, and straight-chain aldehydes, such as, octanal, nonanal and decanal. Eight compounds were determined in the green fruit peel which were not present, or present in very minor quantities, in the healthy fruit. Their retention indices on a DB-5 column were RI 1137, 1144, 1149, 1189, 1320, 1334, 1346 and 1387. The preliminary identifications based on MS and the NIST database RI comparisons were -p-mentha-2,8-dien-1-ol, 1,8-p-menthadien-2-ol, 3-[-3- methyl-1-butenyl]-cyclohexene, 6-methyl-bicyclo[3.3.0]oct-2-en-7-one, p-mentha-2,8-diene-1-hydroperoxide, an unknown, p-menth-8-en-2-ol and p-mentha-[1,8]-diene 2-hydroperoxide, respectively. Other compounds in HLB-affected green fruit were cis-3-hexenol, cispinocarveol, carvone, and α-cubebene. The large scale surveys for detecting Asian citrus psyllids and HLB in Texas and California are on-going. We continue to analyze the spatial pattern of the diagnostic Ct-values from psyllid samples and our goal is to help focus the survey effort for HLB infected citrus trees. The realtime polymerase chain reaction diagnostic methods used to detect HLB are set to run for 40 cycles. The reaction must surpass a threshold of ≤ 32 prior to the completion of the run to be considered positive for HLB. Currently,growing hydroponically reactions that surpass the threshold at 33 or more cycles have proven impossible to acquire a confirmatory PCR bands and DNA sequence data and are considered inconclusive.
The question we are trying to answer is whether there is additional information in the spatial pattern of inconclusive psyllid samples for predicting locations with HLB infected trees. Based on analysis of the spatial pattern of psyllid Ct-values proximate to HLB positive tree detections in both Texas and California, there appears to be a biological process underlying indeterminate Ct-values in the range of 32 – 38. Analyses indicate significant clustering of indeterminate Ct-values of psyllid samples, and some clusters in both Texas and California were around known positive HLB trees. The San Gabriel area infections in California were found after more intensive survey around an inconclusive psyllid sample. Analyses have also revealed significant space-time scales of pattern in the California psyllid data, such that inconclusive samples tend to cluster within 1.5 miles and 15 months of each other. A secondary cluster was found for inconclusive samples > 4.5 miles and > 3 years from the first detection in Hacienda Heights. These analyses may serve as early warning indicators for HLB positive trees, as well as informing dynamic spread models in development. A long period of data about the occurrence of HLB in commercial citrus farms located in São Paulo state, Brazil, was used to evaluate the disease progress and its management. Approximately 260 Sweet orange plots from five farms , naturally affected by HLB and under strict disease management , were evaluated considering the proportion of HLB-affected trees, occurrence of Diaphorina citri, plot location, and presence of neighboring commercial and non-commercial citrus plants. The logistic model better explained, than Gompertz, the yearly proportion of HLB-affected trees. The cumulative disease incidence varied from <4.0% to 80% and highly correlated with area under disease progress curve, independently of plant age, but not with the disease rate of progress . Adults of D. citri were poorly detected by shoot evaluation conducted 2 to 5 times/month in 1% of trees in each plot. In contrast, higher frequency of D. citri identified in yellow stick cards occurred in the most HLB-affected plots.
However, the majority of yellow cards presented no adults for most of the time. Spatial heterogeneity was observed for both HLB-tree incidence and D. citri detection . In all farms, a strong concentration of HLB-affected trees occurred in some plots located in their borders.In the majority of plots, especially those presenting higher disease incidences, >95% of HLB-trees occurred up to 200 meters from their borders. The disease incidence in each farm varied from 7.1% to 25.9%, though the similar disease management adopted for HLB . Recently, surveys around those farms have been started to identify neighboring citrus and Murraya sp. plants that could may act as HLB-sources of inoculum. Their location, represented by commercial and non-commercial citrus plants, strongly suggests that non-commercial trees are very effective sources of HLB, even when represented by a few plants. The Asian citrus psyllid Diaphorina citri Kuwayama is the insect vector of Candidatus Liberibacter asiaticus , the causative agent of the citrus greening disease. Understanding the interactions between the insect vector and the bacterial pathogen is essential for the development of efficient strategies aimed at limiting the spread of the infection and inhibition of bacterial growth. Furthermore, better knowledge of insect-pathogen interaction at the metabolite level will help develop in vitro culture of CLas, a much needed tool for studying its metabolism and designing means to combat it. In this study, we compared the primary metabolite and lipid profiles between CLasinfected and healthy psyllids using GC-TOF-MS and LC-TOF-MS, respectively. We found that both feeding on infected trees and the actual infection with CLas cause dramatic changes in monitored compounds’ accumulation. The results of metabolomic analyses and the possible implications for the pathogen-vector interactions will be presented and discussed. Huanglongbing is a devastating disease of citrus caused by the Candidatus Liberibacter asiaticus . The inability to sustainably culture and reinoculate Las into any host has prevented functional genomics and impeded research and development of disease control methods.
Phosphatidylcholine has recently been established as an important membrane component of those few bacterial species that associate with eukaryotic membranes, such as Legionella pneumophila, Agrobacterium tumefaciens and Rhizobium leguminosarum. For many of these species, PC is required for virulence. Las colonizes living citrus phloem cells, and infects psyllid hosts systemically, moving both intraand inter-cellularly, crossing through multiple membrane layers. We hypothesize that Las PC is similarly required for colonization of psyllids. Liberibacter crescens , the only cultured Liberibacter to date, is a useful and genetically tractable proxy for functional genomics of Las. It carries two predicted pathways for synthesis of PC: 1) import of exogenous choline and subsequent conversion of choline to PC utilizing phosphatidylcholine synthase , and 2) de novo synthesis of PC utilizing phospholipid N-methyltransferase . Notably, Las is missing the de novo pathway, and must import choline,flood and drain table which is abundantly present in psyllid hemolymph and phloem sap, to make PC. We created individual marker interruption mutations of both Pcs and Pmt in Lcr. Rescue of the Pmt mutant in Lcr required 0.1% choline supplementation in the medium. The standard culture medium for Lcr is BM7 which contains 6X10-6 % choline, an amount that appears to be insufficient for culturing Las. Other genes found in Lcr but not found in Las and that should significantly limit Las culturability are: glyoxylase I and the ATP/ADP translocase. Phage lytic cycle genes found in Las but missing from Lcr also likely significantly limit Las culturability. Despite the availability of multiple complete genomic DNA sequences of Candidatus Liberibacter asiaticus , inability to sustainably culture, manipulate or re-inoculate Las has seriously impeded the identification of virulence factors and subsequent development of disease control methods. Liberibacter crescens , is a culturable proxy for Las, but is not pathogenic. Given that the Las genome is even more reduced than Lcr and several Lcr genes are suspected of contributing to its culturability, one long term goal of our lab is to sequentially delete multiple Lcr genes not found in Las and to make its morphology and metabolism more comparable to Las, and make informed incremental adjustments in the growth medium to help compensate for the loss of the Lcr functions. To date, the best growth of Lcr has been obtained in the complex growth medium BM7, containing fetal bovine serum , ACES, α-ketoglutarate, and TNM-FH insect medium. We have found that replacing FBS with 1 mg/mL of methyl-β- cyclodextrin reduced growth, even though such replacement has been reported to improve culture of the fastidious pathogenicmicrobes Coxiella burnetii. In addition, both BSA and glucose, major components of FBS, inhibited growth of Lcr. Both MβCD and BSA are hypothesized to inhibit Lcr growth by sequestering fatty acids and other small hydrophobic molecules that are important to Lcr growth. Lcr growth was unaffected by the presence of cholesterol in the medium.
Replacing TNM-FH with Schneider’s insect medium resulted in removal of sucrose and marginally improved growth of Lcr. Eliminating sucrose from the culture medium allows use of conditionally lethal DNA cloning vectors such as pUFR080 carrying sacB that can be forcibly and reliably evicted by adding sucrose to growth media. Such a system is important to make sequential mutations. We report here that Lcr is able to import isotopic sucrose in a time- and dose-dependent manner, enabling use of the sacB eviction system. We have successfully created 2 marker interruption mutations in Lcr that make Lcr more Las-like and that will likely inform culture media components. Other genes found in Lcr but not found in Las and that should significantly limit Las culturability are: glyoxylase I; ATP/ADP translocase and the sucrose transporter. We are currently attempting a series of conditional knockout mutations involving a sacB vector with genes suspected of enabling Lcr growth in media, each time evaluating different rescue strategies to keep the resulting mutant strains alive. Fruit from trees with severe symptoms of Huanglongbing produce juice that is described as less sweet, more sour with some bitterness and harshness together with typical off-flavor in comparison with juice made with fruit from healthy trees. Commercial juice processing includes separation and recovery of peel oil from the fruit, which is then used in flavor manufacturing including add-back to orange juice to standardize flavor. To this date, there is no published data on the aroma quality of peel oil from fruit affected by HLB. Hamlin and Valencia fruit from healthy and HLB-infected trees were harvested in December 2013 and April 2014, respectively. Fruit were washed and juiced using a JBT industrial extractor. Peel was separated and the oil was expressed using several cycles of manual cold pressing followed with centrifugation. Peel oil was then analyzed by gas chromatography mass spectrometry and olfactometry , as well as presented to a trained taste panel for aroma differences. Juice from the same fruit were also presented to the trained taste panel for sensory descriptive analysis. For peel oil from Hamlin, 14 compounds had higher peaks by GC-MS in healthy samples, including 1-octanol which was perceived by GC-O. Further, three unidentified peaks with mint , citrus and pungent odors were perceived only in the oil from healthy fruit by GC-O. For Valencia oil, differences by GC-O were for peaks with odors of -cadinene and metallic which were only perceived in HLB samples. Other differences were for RI=955 , α-phellandrene , and terpinolene , which had higher intensities in Valencia healthy samples. Overall, those differences detected by GC-O were not translated to aroma differences between oil samples by a trained panel. The same trained panel found differences when tasting the juice of Hamlin, with HLB juice having more ‘typical HLB flavor’, ’bitter’, ‘metallic’, ‘tingling’ and ‘astringent’, likely due to differences in nonvolatile compounds. There were no difference between HLB and healthy samples for Valencia juice. Some new hybrid root stocks grafted with commercial citrus scion cultivars produce trees with reduced huanglongbing symptoms and higher fruit production over a period of many years, following infection by Candidatus Liberibacter asiaticus , as compared with trees on other common commercial root stocks in the same trials. In one replicated trial with ‘Valencia’ scion severely affected by HLB, highest yields over five consecutive harvests were with ‘US-942’ and ‘US-1516’ root stocks. Other components of tree health, including fruit quality, tree growth, and tree survival, differed greatly among the root stocks, and provided additional evidence of improved field tolerance to HLB for some root stocks.