A transgenic test site at the USDA/ARS USHRL Picos Farm in Ft. Pierce supports HLB/ACP/Citrus Canker resistance screening for the citrus research community. There are numerous experiments in place at this site where HLB, ACP, and citrus canker are widespread. The first trees have been in place for over three years. Dr. Jude Grosser of UF has provided ~600 transgenic citrus plants expressing genes expected to provide HLB/canker resistance, which have been planted in the test site. Dr. Grosser planted an additional group of trees including preinoculated trees of sweet orange on a complex tetraploid rootstock that appeared to confer HLB resistance in an earlier test. Dr. Kim Bowman has planted several hundred rootstock genotypes, and Ed Stover 50 sweet oranges (400 trees due to replication) transformed with the antimicrobial peptide D4E1. Texas A&M Anti-ACP transgenics produced by Erik Mirkov and expressing the snow-drop Lectin (to suppress ACP) have been planted along with 150 sweet orange transgenics from USDA expressing the garlic lectin. Eliezer Louzada of Texas A&M has permission to plant his transgenics on this site, which have altered Ca metabolism to target canker, HLB and other diseases. More than 120 citranges, from a well-characterized mapping population, and other trifoliate hybrids (+ sweet orange standards) have been planted in a replicated trial in collaboration with Fred Gmitter of UF and Mikeal Roose of UCRiverside. Plants are being monitored for CLas development and HLB symptoms. Data from this trial should provide information on markers and perhaps genes associated with HLB resistance, for use in transgenic and conventional breeding. Dr. Roose has completed initial genotyping on a sample of the test material using a “genotyping by sequencing” approach. So far, the 1/16th poncirus hybrid nicknamed Gnarlyglo is growing extraordinarily well. It is being used aggressively as a parent in conventional breeding. In a project led by Richard Lee, an array of seedlings from the Germplasm Repository are in place, with half preinoculated with Liberibacter. Additional plantings are welcome from the research community.
This proposal aims to continue improvement to a novel psyllid trap and to use the trap to gather new information on the behavior, biology, population dynamics and biological control of ACP/Candidatus Liberibacter asiaticus. Lab and field testing was and continues to be conducted to increase trap efficiency by exploiting unique vector behaviors in response to traps and behaviorally active components. Obj. 1: We continue to conduct field and laboratory studies toward obtaining an understanding of ACP trap response behavior by manipulations of visual cues on and around the trap. We have a number of positive results from our latest bioassays and continue to tweak the trap structural components to increase trap efficiency through field testing. We have and continue to test the utility of various light types to increase psyllid response along with parameters of trap placement in and near trees. This work is ongoing and more experiments are being conducted but none of the lures we have tested have provided any increase in trap catch. This within Florida research component is continuing and we now added several new locations in Puerto Rico to exploit the significantly higher ACP populations there that make the research easier. Obj. 2: We have initiated the areawide psyllid sampling objective to detect and develop new biological controls for use against ACP. We have begun sampling in the northern most citrus populations in Alachua County and around Ft. Pierce with the intention of systematically sampling by working south from the northern region and outward in all direction from Ft. Pierce. The standard prototype trap is being used for this work and performs well enough to complete this effort, i.e., where ACP occur the trap captures and preserves them in proportion to their populations. So far we have not identified any new pathogens but we have also expanded research on this objective into Puerto Rico. Additionally, we have ontained cooperation with other USDA-ARS research personnel with expertise in the identification and rearing of entomopathogenic fungi.
cDNA and genomic DNA sequences from the three FT citrus constructs were aligned with GeneBank’s published Citrus Unshiu sequences to properly asses the identity of the constructs being used to transform tobacco and citrus. The evolutionary history was inferred using the Neighbor-Joining method with a total of 10 nucleotide sequences. The results display an optimal tree revealing that FT3 is the most different FT. Also, the analysis shows that ciFT1 and ciFT2 cluster together and display high sequence similarity, indicating that these genes might in fact be alleles and not two separate genes. The one year study of the in vivo tracking of FT1, FT2, and FT3 in various citrus trees differing in age and phenotype is currently being repeated with higher concentrations of cDNA to solidify data and four candidate genes in the flowering pathway (FLD, FLC, ELF5, and AP1) have been added to the study to determine their involvement in flowering time and what effect they might have in the induction of the three FT genes. Results and statistical analysis from the quantitative real time PCR for FT1 and FT2 are being analyzed to further prove the relationship between FT1 and FT2 as being allelic. Tobacco ciFT3 transgenic lines from T1 and T2 generations where studied in order to determine the segregation pattern for ciFT3. GUS histochemical testing was done on germinated seedlings from four lines of T1 plants and four lines of T2 plants. Results show a that the transgene is passed on to offspring in a stable manner and in accordance with expected segregation ratios given a single gene insertion assuming heterozygous parents or more than one genetic insertion of the transgene. One T2 line in particular appears to be homozygous for the GUS gene, indicating that all offspring should have the ciFT3 gene which induces early flowering. This segregation data will be included in the transgenic tobacco manuscript currently being written. The endogenous ciFT3 promoter was successfully cloned to be used in the transcription activator-like (TAL) effector system inducible by methoxyfenozide that will activate the naturally present FT3 gene in citrus.
We aim in this project to genetically manipulate defense signaling networks to produce citrus cultivars with enhanced disease resistance. Defense signaling networks have been well elucidated in the model plant Arabidopsis but not yet in citrus. Salicylic acid (SA), jasmonic acid (JA) and ethylene (ET) are key hubs on the defense networks and are known to regulate broad-spectrum disease resistance. With a previous CRDF support, the PI’s laboratory has identified ten citrus genes with potential roles as positive SA regulators. Characterization of these genes indicate that Arabidopsis can be used not only as an excellent reference to guide the discovery of citrus defense genes and but also as a powerful tool to test function of citrus genes. This new project will significantly expand the scope of defense genes to be studied by examining the roles of negative SA regulators and genes affecting JA and ET-mediated pathways in regulating citrus defense. We have three specific objectives in this proposal: 1) identify SA negative regulators and genes affecting JA- and ET-mediated defense in citrus; 2) test function of citrus genes for their disease resistance by overexpression in Arabidopsis; and 3) produce and evaluate transgenic citrus with altered expression of defense genes for resistance to HLB and other diseases. With this support, currently we have cloned six full-length genes with potential roles regulating SA, ET, and/or JA pathways to the binary vector pBIN19plusARS and transferred the constructs to Agrobacteria. All six Agro strains were sent to co-PI Dr. Bowman’s lab to initiate citrus transformation. In the mean time, we started the process of transforming Arabidopsis to overexpress these genes and to test their defense function. T0 transformed seeds have been harvested for some constructs and will be screened for transgenic plants followed by disease resistance tests. We aim to clone at least 10 citrus genes for testing their effectiveness in conferring disease resistance to HLB and citrus canker diseases. Additional gene cloning is underway. In addition, we are continuing to characterize transgenic citrus plants expressing the SA positive regulators, as proposed in the previous project (#129), although the support of the project has already been terminated.
Obj 1A: We last reported that all components of the user-friendly web versions of the single (sTCW) and multi-(mTCW) psyllid transcriptome databases had been finalized in preparation for the public release which will allow researchers to quickly identify targets, not only based on expressional differences but in combination with predicted functions and tissue specificity (e.g., midgut and salivary glands). To date, the expression of 19 transcripts predicted to be important for nutrition, adhesion, immunity and defense has been validated in ACP by RT-PCR. Obj 1B: Yeast-2 hybrid studies were initiated to study protein-protein interactions important in psyllid-Liberibacter interactions. Previously it was reported that a total of 12 CLas candidate genes from a list of 19, had been moved into the Yeast 2 Hybrid (Y2H) mating experiments using the ACP gut and salivary gland libraries. To date 15 gut library matings and 13 salivary gland library matings have been performed. Data analysis has been completed for 26 of those experiments with the remaining being in various stages (PCR, cloning, sequencing, etc.) moving towards completion. Through those 26 experiments we have thus far discovered over 50 ACP gene products that have high levels of interest making them good candidates for RNAi and will be moved into that phase of the project if they fit the criteria specific for RNAi. Previously we reported the findings from the first 5 mating projects (4 midgut and 3 salivary glands) with the matings of one CLas bait highlighting the importance of tissue specificity when identifying candidate effectors. Since the last update, we have discovered more interesting candidate effectors that will be moved into the RNAi phase of the project many of which are predicted to be important in Liberibacter adherence to host tissues. One interesting candidate is a transporter-like protein that is known to be critical in host-pathogen interactions. The knockdown of this gene may inhibit CLas dissemination in the psyllid. An ACP gene product similar to an enzyme involved in apoptosis was also identified. The knockdown of this gene may interfere with psyllid tissue integrity, thus CLas’s ability to adhere properly which is likely required for successful transmission. To date, nine candidate ACP genes putatively involved in bacterial adhesion, nutrition, and defense have been mated against the CLas prey library. Several interacting gene products (‘prey’ inserts) have been identified with many still being analyzed. One ACP candidate with putative functions in endocytosis and an adhesion-related transcript interacted with two different CLas proteins, both of which are associated with extracellular matrices. Another ACP bait with putative peptidase activity interacted with a CLas pilus-associated protein. Due to the potential impact on virulence of the identified CLas prey, the knockdown of the ACP interacting partners is being investigated. Obj 2: RNA interference (RNAi) studies are underway to functionally validate candidate effectors. To date, good quality dsRNA has been synthesized for six psyllid genes predicted to be involved in cytoskeleton formation, defense response, vesicle transport, or transcytosis and nutrition. Previously we reported on the impact of the knockdown of two of the cytoskeleton-related genes on Liberibacter transmission using both oral delivery and microinjection methods. The preliminary data suggested that the method of dsRNA delivery is crucial, and should be optimized accordingly. To date, dsRNA for 6 genes have been oral delivered and/or microinjected into psyllids. Confirmation of knockdown by qPCR has been confirmed for two, one of which was tested in the transmission bioassay and showed a reduction in transmission by 18%. Confirmation of knockdown, followed by transmission bioassays of the remaining genes, is ongoing.
In this proposal our objective is to find citrus versions for the two proteins that make up the functional components of a chimeric antimicrobial protein (CAP) previously described by us (Dandekar et al., 2012 PNAS 109(10): 3721-3725). We have successfully identified a suitable replacement for the first component, the human neutrophil elastase (HNE) that also serves as the surface binding component of CAP. Since HNE is a serine protease with elastase activity whose 3D structure has been determined we used the PDB database to find a suitable plant protein with the same 3D structure. Using the active site geometry of HNE is a consistent structural feature we focused on a set of 288 non-redundant plant derived proteins extracted from the PDB database to narrow our search criteria. The key feature of our search involved using CLASP to search for a match using the electrostatic properties and structural geometry of the three amino acids that make up the active site of HNE. We obtained a close match with the tomato PR14a protein. Using the tomato amino acid sequences we then searched for a similar citrus protein by searching through citrus genome information in Phytosome (http://www.phytozome.net). This was successful and we have identified a single protein that has the identical amino acid sequence in both Citrus sinensis (Cs) and Citrus clementina (Cc) genomes. We have focused the 165 amino acid P14a protein from Cs which we refer to as CsP14a. We have analyzed this sequence and have determined that it is a secreted protein and contains what appears to be a 25 amino acid signal sequence. We have utilized the 137 aa mature protein and successfully constructed two synthetic genes that encode this protein, one that just contains the coding region of CsP14a and the other is a chimeric version that contains the CsP14a coding region linked to the CecropinB (CecB) protein. We have included a signal peptide (22aa) that we have used before and know works really well at secreting proteins to the plant apoplast and xylem. This signal peptide has been added at the N-terminal of this protein and we have added a Flag Tag also at the N-terminal so that the protein can be easily detected and purified. The Flag tag will remain a part of the secreted protein after cleavage of the signal peptide. We are constructing two CaMV35S expression cassettes to express both these synthetic genes. We are using CLASP to identify a citrus replacement component for CecB. Since CB has no enzymatic activity, we could not use a well-constrained motif like an active site. We chose instead the structural motif Lys10, Lys11, Lys16, and Lys29 a unique feature of CecB. Our analysis has proved fruitful and we have identified a good plant candidate that has the same shape and that is highly conserved in plants.
Our accomplishments are: 1) Various young and mature citrus plants and also citrus seeds that are used for this project were purchased, planted and maintained in a greenhouse; 2) Sterile culture of citrus plant materials were established; 3) Construction of the proposed genes that should enhance shoot regeneration and embryogenesis has been started and is well underway.
A gene obtained from Dr. Mou that confers tolerance to canker has been transformed into mature Hamlin, Valencia, Pineapple, Ray Ruby, Carrizo, and Swingle, and shoots are regenerating. Unfortunately a number of PCR positive Pineapple shoots that were micro-grafted onto Carrizo rootstock and maintained in liquid media were lost due to an error in media preparation. This error has been corrected. Six shoots of micro-grafted Pineapple survived from this batch and an additional 33 Pineapple shoots have been micro-grafted onto Carrizo rootstock. Shoots regenerated from Carrizo and Swingle explants are being elongated in tissue culture prior to rooting. GUS assays for the reporter gene and additional molecular analyses will be conducted once the shoots are larger. Two constructs obtained from Dr. Wang were also used in transformation experiments of mature Valencia. Shoots were micro-grafted onto Carrizo rootstock and are still growing in liquid media prior to secondary grafting or have already been transferred to the soil. Molecular analyses will continue once the plants are larger. Molecular analyses of Hamlin, Valencia, Pineapple, and grapefruit scion transformed with marker genes are underway. Thus far, nine out of ten GUS or GFP positive plants have tested positive for the expression of the nptII transgene using the nptII immunostrip assay. The remaining ~40 transgenics will be tested using nptII immunostrips, nptII ELISAs. and Southern blotting. The first flower buds have formed on a Valencia transgenic event originally generated on 5/30/12, so it has been ~19 months for flowering to occur which agrees with Dr. Pena’s protocol. One of the limitations of the mature scion transformation protocol is the relatively slow process of bud break of the scion following grafting and the slow growth of the scion for explants. The double budding procedure and daily hormone applications to induce early bud break have significantly increased productively of the growth room. Buds now break one week after the grafting tape has been removed. We have observed significant differences in bud break of the scion on different rootstocks following hormone application. The photoperiod has been extended to 19 hours of light and 5 hours of dark to further increase the rate of vegetative growth and productivity of the growth room. A number of scientists were contacted to provide additional constructs and three scientists indicated they will provide constructs in the near future.
Cytoplasmic (CiLV-C and CiLV-C2) and nuclear (CiLV-N) citrus leprosis virus cause citrus leprosis disease in North and South America. All types of the leprosis viruses are transmitted by Brevipalpus mites. We continued mite transmission experiments at the USDA, ARS, Foreign Disease and Weed Science Research Unit, Ft. Detrick, MD with endemic healthy Brevivalpus yothersii (syn. phoenicis) mites from Florida. We again did mite transmission experiments with the citrus leprosis affected samples from Mexico (CiLV-N) & Colombia (CiLV-C2). As reported previously six weeks after completion of the transmission experiments none of the citrus seedlings showed leprosis symptoms. For confirmation of the negative test results leaf tissue from the experiments were analyzed by reverse transcription polymerase chain reaction (RT-PCR) using CiLV type-specific primers. Recently, nuclear CiLV was reported from Mexico but no prior sequence information was available. We successfully determined the entire genome sequence of nuclear CiLV and and published a manuscript on this sequence in the journal Genome Announcement (‘Genome assembly of citrus leprosis virus nuclear type reveals a close association with orchid fleck virus’. Our collaborator in Mexico has shipped us further shipment of infected nuclear leprosis samples to continue transmission experiments. Results are pending on these transmissions. Using newly developed PCR primers we are hoping to determine the viruliferous status of Brevipalpus mites in acquisition of the nuclear citrus leprosis virus. Based on these results we will determine if Florida endemic healthy Brevivalpus Florida mites are able to acquire the various citrus leprosis viruses. In addition mites (preserved in alcohol) from Mexico have been sent to USDA cooperators to continue to compare their taxonomic status with those that do transmit in Colombia and elsewhere. In Colombia our cooperator, Guillermo Leon, continued work on the transmission, viability and interactions of the of CiLV-C2 and the vector, Brevipalpus phoenicis (Geijskes). The acquisition of the virus from citrus, then feeding on non-citrus host plants and returning to citrus plants for virus transmission was evaluated. Results were that B. phoenicis mite was able to transmit the virus to Valencia orange plants (Citrus sinensis L.) at a transmission rate of about 80%, after having fed for periods of two to twenty days in any of the six alternate mite host plants. The alternative host plants included Dieffenbachia sp., Hibiscus rosacinensis, Codiaeum variegatum, Swinglea glutinosa, Sida acuta or Stachytarpheta cayennensis. The appearance of the first leprosis symptoms in the Valencia orange leaves, when the mite previously fed on S. glutinosa, Dieffenbachia sp. and C. variegatum appeared the earliest followed by S. acuta and then followed by H. rosacinensis. All of these results have been submitted for publication.
The goal of this experiment is twofold, first to determine the effects of plant growth regulators on addressing vascular degeneration and fruit drop, and second to determine the effects of HLB and ACPS citriculture on drought tolerance. A field experiment was installed in April 2013 to test the efficacy of the synthetic auxin 2,4-D and a micro-emulsion ‘based surfactant to reduce HLB symptom severity in a mature ‘Hamlin’ orange block. The HLB incidence in the block is currently more than 50% and consequently the fruit yield losses due to pre-harvest fruit drop from symptomatic trees were devastating in the 2012/13 season. The experimental design is a 4×4 Latin Square with four replications and four factorial foliar spray treatments consisting of 2,4-D, Eco-Agra’ surfactant, 2,4-D + Eco-Agra’, and untreated control. Each whole plot is split into two sub-plots containing Swingle and Carrizo rootstocks. A basal foliar nutrient spray treatment applied to the whole experiment consists of a comprehensive, balanced fertilization program of micronutrients, macronutrients and potassium phosphite products timed to coincide with the major leaf flushes. The basal ground-applied fertilizer program consists of a dry granular bulk-blended N-P-K +Ca +Mg + Fe + Mn +Zn +B +S product applied four times in the growing season. The automated micro-sprinkler irrigation system is used to apply water to the trees according to seasonal evapotranspiration demand as needed, up to twice per day. We have been monitoring preharvest fruit drop in this 4-acre Hamlin block in Lake Alfred since 8/19/2013 every two weeks by raking out the dropped fruit from 4 trees per block, 2 trees on Swingle and 2 trees on Carrizo to determine the efficacy of the 2,4-D/EcoAgra treatments. The control and 2,4-D+EcoAgra treatments appear to be having the most significant fruit drop, with an average of 200-215 pieces of fruit dropped per tree. The EcoAgra or 2,4-D treatments on their own appear to be reducing drop, but only by about 35%. These data are preliminary and should not be used as a recommendation for growers at this point. Within the next two weeks the fruit from the experimental trees will be picked and we can then determine the percentage of fruit dropped prior to harvest for individual trees. This will be extremely important due to the high variability of symptoms in this block. For example, several trees entered the experiment with moderate to severe HLB symptoms and over the past 4 months have completely defoliated and dropped all of their fruit. Meanwhile, other trees remain vigorous with mild HLB symptoms and a good crop load.
Understanding the transmission of CLas within the citrus tree remains one of the principal obstacles in the global efforts to undermine the pathogenicity of HLB (citrus greening). The movement of CLas has been assumed to follow the photoassimilate stream through the phloem. However, many observations based on our knowledge of the bacteria and general phloem anatomy have exposed inconsistencies with the accepted beliefs. The brevity of available information on the ultrastructural properties of citrus phloem sieve elements has hindered efforts to understand the spread of the disease within a tree. For example, lateral movement of CLas around an infected stem appears improbable given the size of cytoplasmic plasmodesmata connections between adjacent sieve elements and the isolated nature of phloem cells. Furthermore, spreading of CLas from the roots to uninfected aerial tree parts through the phloem seems highly unlikely given the direction of phloem sap. To date we lack a thorough investigation into the ultrastructure of citrus phloem and the surrounding tissue, the potential pathways that CLas could utilize to move long distances through citrus trees, and the location of CLas habitat within different citrus tissue. Using a variety of grafting and girdling experiments, SEM, TEM, confocal, high resolution computed tomography, and PCR tissue analysis we aim to gain a better understanding of the anatomical traits that facilitate the spread of CLas through citrus. These data will allow us to develop new screening tools that breeders can use to select for resistant scion/rootstock combinations to confer resistance or tolerance to HLB. As of this progress report the Valencia/Swingle trees have had HLB+ tissue grafted onto them. Approximately 80% of the grafted tissue has produced new flush and we are waiting for the remainder to produce new tissue. The girdling experiments have been completed and we are actively monitoring the wounds and the healing process to make sure that the redifferentiated tissue does not act as a pathway for CLas. We have hired a part-time employee to begin learning the microscopy techniques so they are well positioned to start the anatomical analysis once the trees are ready. We expect to begin tree dissection within the next quarter, followed by another 6 months to allow the remaining trees to develop HLB symptoms.
This portion of the project investigated the effect of HLB on Hamlin and Valencia orange peel oil quality by analyzing the volatile and olfactory profiles. In one study, peel oil samples were evaluated by gas chromatography-olfactometry (GC-O), where a person sniffs the volatiles as they come off a GC column, by GC-mass spectroscopy (MS) for peak identification, and by a sensory descriptive panel evaluating oil on filter paper. Analysis by GC-O revealed 57 odor-active peaks of which 33 were identified by GC-MS, and 22 confirmed by smelling chemical standards by GC-O. All identified compounds have been preciously reported in citrus and orange essential oil. There were 7 aliphatic aldehydes (hexanal, octanal, nonanal, decanal, undecanal, decadienal, dodecanal), 4 monoterpene aldehydes (citronellal, neral, geranial, perilla aldehyde), 3 aliphatic alcohols (hexanol, octanol, nonanol), 7 monoterpene alcohols (sabinene hydrate, linalool, 4-terpineol, citronellol, nerol, geraniol, cis-carveol), 8 monoterpene hydrocarbons (.-pinene, sabinene, myrcene, .-phellandrene, .-3-carene, limonene, .-phellandrene, terpinolene), 8 sesquiterpene hydrocarbons (.-copaene, .-cubebene, .-caryophyllene, .-copaene, .-humulene, germacrene D, valencene, .-cadinene), 1 monoterpene oxide (limonene oxide), 1 acid (methyl-octanoate) and 1 monoterpene ketone (carvone). Furthermore, 24 unknown compounds were detected by smell. Paired comparisons looked at Hamlin healthy/Hamlin HLB, and Valencia healthy/Valencia HLB. Multivariate statistics (PCA) found no differences in peak intensities between Hamlin healthy and HLB samples. Likewise, panelists could not distinguish between healthy and HLB samples for Hamlin oil. For Valencia oil, LRI 1185, .-cadinene and LRI 1645 were only perceived in HLB samples. Other differences were for LRI 955, .-phellandrene, and terpinolene, which had higher intensities in Valencia healthy. For GC-MS analyses, 10 compounds were significantly different between Valencia healthy and HLB, but contrary to Hamlin oils, they presented higher peak areas in Valencia HLB than in Valencia healthy, except for hexanal and cis-p-mentha-2,8-dien-1-ol. No significant difference was perceived between healthy and HLB Valencia samples by panelists during the difference test. Likewise, there were no significant differences between Valencia healthy and Valencia HLB juice made from the same oranges as for the peel oil extract. Therefore, the small differences detected by GC-O between healthy and HLB peel oil were not perceived by panelists smelling of whole oil, or drinking orange juice. In conclusion, this study showed little difference between samples due to disease, either by GC-O or sensory evaluation. Only a few volatiles were perceived with greater intensity in the oils from healthy fruit in both Hamlin and Valencia, and three volatiles were only perceived in Valencia samples from HLB fruit. These differences were small and not important enough to be perceived in the oil by a sensory panel. However, in another study, cold pressed peel oil samples from Valencia fruit (26), each obtained from healthy, severely infected (HLBs) or mildly infected trees (HLBm), showed more, albeit similar differences. A total of 57 volatile compounds were identified by GC-MS in peel oil samples, including 9 monoterpenes, 16 sesquiterpenes, 12 alcohols, 13 aldehydes, 1 alkane, 2 ketones, 2 esters, and 2 terpene oxides. Of those, 14 compounds were found to be significantly different among healthy, HLBs and HLBm samples. Hexanal, (E,E)-2,4-decadienal, .-cadinene and .-copaene were significantly lower in HLBs samples than in the healthy samples, while sabinene, (E)-p-mentha-2,8-dien-1-ol, .-terpineneol, 3,7-dimethyl-6-octen-1-ol, (Z)-3,7-dimethyl-2,6-octadien-1-ol, carvone, cyclodecane, .-cubebene, (E)-.-farnesene, .-humulene and .-farnesene were significantly higher in HLBs samples. The contents of those volatiles in HLBm were in between. In conclusion for this study, HLB altered Valencia peel oil volatile profiles in that many terpenes were higher in HLB samples, probably due to disease stress upregulation of these compounds,while some aldehydes were surpressed, which may negatively impact the peel oil quality. However, sensory analysis was not done on these samples to confirm detectability.
Chemical and sensory analyses of fruit harvested from Huanglongbing (HLB) or greening-infected trees in winter (December Hamlin), 2012 (actually prior to start of project), winter (January, Hamlin) and 2013 spring (March and April, Valencia) were conducted. Trained sensory panels were completed and analyzed, while difference-from-control tests are still ongoing. Chemical analyses of sugars, acids, aroma volatiles, vitamin C, limonoids and flavonoids are completed, but the data not yet completely analyzed for aroma volatiles. The electronic tongue (etongue) analysis and molecular (qPCR) analysis of the titer of the pathogen (Liberibacter asiaticus) DNA in the juice (patent submitted) has been completed and analyzed. So far physical fruit measurements show that fruit from HLB-infected trees are smaller and more green than fruit from healthy trees, regardless of nutritional treatments of which 3 were investigated so far. Sugars, acids, ratio and total ascorbic acid were generally lower in HLB juice, although acids were sometimes higher in juice from symptomatic fruit, regardless of treatment, except for one nutritional treatment for Valencia in March, 2013. Bitter limonoids along with many other flavonoids were higher in HLB juice, regardless of nutritional treatment and especially in symptomatic fruit or fruit from severely infected trees. So far, the nutritional treatments have not shown a consistent effect, but there are sporadic positive effects on flavor chemicals. The nutritional treatments did not show an effect on reduction of Liberbacter titer in the juice as evidenced by qPCR analysis. The electronic tongue (etongue) and nose could discriminate between juices from healthy, asymptomatic-HLB and symptomatic-HLB fruit, with the etongue being much more effective. The etongue could also discriminate the different nutritional treatments within a harvest, but was confounded by seasonal changes across harvests. The etongue was more effective for Hamlin than Valencia, reflecting the more severe HLB-induced flavor effects for Hamlin. Trained panel showed differences in perception of orange and grapefruit, fruity, green and stale flavors and sweet, sour, bitter, metallic, tingling, astingent and umami (salty) tastes. The differences were minimized by nutritional treatments for Hamlin in December, 2012 and January, 2013, and sometimes nutritional treatments generally increased perception of sweetness. For Valencia, March and April, 2013 the nutritional treatments had no effect or enhanced sweetness but did not mitgate other off-flavors, respectively.
Fruit harvested from nutritionally treated or conventionally treated healthy or Huanglongbing (HLB) or greening-infected trees in January (Hamlin), 2014 and early and late April 2014 (Valencia with 3 new nutritional treatments added for a total of 9 that were replicated in the field). All analyses have been completed for the first year including trained and consumer sensory panels, chemical analyses of sugars, acids, aroma volatiles, vitamin C, limonoids and flavonoids and electronic tongue and the data analyzed. Juice samples were also analyzed by qPCR for Ct values to determin Liberbacter titer in the juice. So far physical fruit measurements show that fruit from HLB-infected trees are smaller and more green than fruit from healthy trees, regardless of nutritional treatments of which 3 were investigated so far for the earlier harvests. Sugars, acids and ratio were generally lower in HLB juice, and acids were sometimes higher in juice from symptomatic fruit, regardless of treatment, except for one nutritional treatment for Valencia in March, 2013. Bitter limonoids along with many other flavonoids were higher in HLB juice, regardless of nutritional treatment and especially in symptomatic fruit or fruit from severely infected trees, although this is more the case in early season harvests. So far, the nutritional treatments have not shown a consistent effect, but there are sporadic positive effects on flavor chemicals. The nutritional treatments did not show an effect on reduction of Liberbacter titer in the juice as evidenced by qPCR analysis so far for the first year. Several times, however one nutritional treatment has shown the ability to make the juice taste sweeter. The electronic tongue (etongue) and nose could discriminate between juices from healthy, asymptomatic-HLB and symptomatic-HLB fruit, with the etongue being much more effective. The etongue could also discriminate the different nutritional treatments within a harvest, but was confounded by seasonal changes across harvests. The etongue was more effective for Hamlin than Valencia, reflecting the more severe HLB-induced flavor effects for Hamlin in earlier samples, but the spring, 2013, Valencia samples did show separation indicating that the disease is becoming more severe for Valencia. Trained panel showed differences in perception of orange and grapefruit, fruity, green and stale flavors and sweet, sour, bitter, metallic, tingling, astingent and umami (salty) tastes. Consumer difference-from-control panels showed that the panelists detect differences that correlate to lower ratio and/or higher limonoids in HLB compared to healthy juice. The differences is greatest when there is both low ratio and high limonoids. Aroma volatile analysis through 2013 show that some tope notes, especially esters are lower in HLB juice generally, but more analysis is needed to determine nutritional treatment effects (unclear at this time). The general flavor differences were minimized by nutritional treatments for Hamlin in December, 2012 and January, 2013, and one nutritional treatment sometimes increased perception of sweetness. Samples were taken in recent harvests for analysis of peel oil and were processed but the samples have not all been run on the gas chromatograph nor the data analyzed. Mineral analysis on the first year juice showed elevated calcium in the HLB samples compared to healthy. So far, there are not consistent pattern for differences in ascorbic acid (vitamin C) due to HLB infection.
This portion of the project investigated the effect of HLB on Hamlin and Valencia orange peel oil quality by analyzing the volatile and olfactory profiles. In one study, peel oil samples were evaluated by gas chromatography-olfactometry (GC-O), where a person sniffs the volatiles as they come off a GC column, by GC-mass spectroscopy (MS) for peak identification, and by a sensory descriptive panel evaluating oil on filter paper. Analysis by GC-O revealed 57 odor-active peaks of which 33 were identified by GC-MS, and 22 confirmed by smelling chemical standards by GC-O. All identified compounds have been preciously reported in citrus and orange essential oil. There were 7 aliphatic aldehydes (hexanal, octanal, nonanal, decanal, undecanal, decadienal, dodecanal), 4 monoterpene aldehydes (citronellal, neral, geranial, perilla aldehyde), 3 aliphatic alcohols (hexanol, octanol, nonanol), 7 monoterpene alcohols (sabinene hydrate, linalool, 4-terpineol, citronellol, nerol, geraniol, cis-carveol), 8 monoterpene hydrocarbons (.-pinene, sabinene, myrcene, .-phellandrene, .-3-carene, limonene, .-phellandrene, terpinolene), 8 sesquiterpene hydrocarbons (.-copaene, .-cubebene, .-caryophyllene, .-copaene, .-humulene, germacrene D, valencene, .-cadinene), 1 monoterpene oxide (limonene oxide), 1 acid (methyl-octanoate) and 1 monoterpene ketone (carvone). Furthermore, 24 unknown compounds were detected by smell. Paired comparisons looked at Hamlin healthy/Hamlin HLB, and Valencia healthy/Valencia HLB. Multivariate statistics (PCA) found no differences in peak intensities between Hamlin healthy and HLB samples. Likewise, panelists could not distinguish between healthy and HLB samples for Hamlin oil. For Valencia oil, LRI 1185, .-cadinene and LRI 1645 were only perceived in HLB samples. Other differences were for LRI 955, .-phellandrene, and terpinolene, which had higher intensities in Valencia healthy. For GC-MS analyses, 10 compounds were significantly different between Valencia healthy and HLB, but contrary to Hamlin oils, they presented higher peak areas in Valencia HLB than in Valencia healthy, except for hexanal and cis-p-mentha-2,8-dien-1-ol. No significant difference was perceived between healthy and HLB Valencia samples by panelists during the difference test. Likewise, there were no significant differences between Valencia healthy and Valencia HLB juice made from the same oranges as for the peel oil extract. Therefore, the small differences detected by GC-O between healthy and HLB peel oil were not perceived by panelists smelling of whole oil, or drinking orange juice. In conclusion, this study showed little difference between samples due to disease, either by GC-O or sensory evaluation. Only a few volatiles were perceived with greater intensity in the oils from healthy fruit in both Hamlin and Valencia, and three volatiles were only perceived in Valencia samples from HLB fruit. These differences were small and not important enough to be perceived in the oil by a sensory panel. However, in another study, cold pressed peel oil samples from Valencia fruit (26), each obtained from healthy, severely infected (HLBs) or mildly infected trees (HLBm), showed more, albeit similar differences. A total of 57 volatile compounds were identified by GC-MS in peel oil samples, including 9 monoterpenes, 16 sesquiterpenes, 12 alcohols, 13 aldehydes, 1 alkane, 2 ketones, 2 esters, and 2 terpene oxides. Of those, 14 compounds were found to be significantly different among healthy, HLBs and HLBm samples. Hexanal, (E,E)-2,4-decadienal, .-cadinene and .-copaene were significantly lower in HLBs samples than in the healthy samples, while sabinene, (E)-p-mentha-2,8-dien-1-ol, .-terpineneol, 3,7-dimethyl-6-octen-1-ol, (Z)-3,7-dimethyl-2,6-octadien-1-ol, carvone, cyclodecane, .-cubebene, (E)-.-farnesene, .-humulene and .-farnesene were significantly higher in HLBs samples. The contents of those volatiles in HLBm were in between. In conclusion for this study, HLB altered Valencia peel oil volatile profiles in that many terpenes were higher in HLB samples, probably due to disease stress upregulation of these compounds,while some aldehydes were surpressed, which may negatively impact the peel oil quality. However, sensory analysis was not done on these samples to confirm detectability.
(863) 956-8817
(863) 956-5894
700 Experiment Station Road
Lake Alfred, FL 33850
Email Us
