Substantial progress in the detection of Candidatus Liberibacter asiaticus (CLas) in citrus tissues and psyllid vector using non radioactive digoxigenin labeled probes during the tenure of this project. We developed and used both digoxigenin-labeled Plabeled PCR probes as well as RNA probes. We produced digoxigenin-labeled PCR products using specific primers for the OMP, RNA polymerase beta subunit, DNA polymerase region, the r-DNA region, and the 23S and 16S ribosomal RNA intergenic regions. Citrus plants from the green house infected with HLB pathogen were used as the source plants for obtaining the tissue for the blots and to isolate total nucleic acids template necessary for the amplification of CLas specific DNA. Primers were designed based on the CLas sequence information and amplified specific amplicons from the DNA isolated from HLB infected plants, and amplification was not observed from the healthy control plants grown under similar conditions. In the initial tissue blot experiments we did not observe hybridization signals specific for HLB. The reason probably is the low titer of pathogen and/or the non uniform distribution of the pathogen in the infected tissues. It is also possible that the PCR probes failed to detect low titer of the pathogen. Therefore, the amplified regions of Las were also cloned in the transcription vector, and digoxigenin labeled strand specific RNA probes were generated by transcription and the probes generated substantially improved detection of Las in citrus tissues. In addition to tissue blots of the stem sections, we have used the midrib and petiole region and whole leaf and branch containing multiple leaves from healthy and infected citrus on nylon membranes. The hybridization observed with the midrib imprints and whole leaves showed clear signals compared to stem imprints. In other experiments we used imprints of the leaf midrib, petiole and stem from new flushes of infected citrus since the psyllids preferentially feed and acquire the pathogen from such tissues. We also imprinted on the membrane the inside surface of the bark which contains phloem tissue in which the HLB pathogen is located. This procedure, was observed to be to be very useful to the determine the distribution of CLas in infected tissue. The second area of our focus is on the detection of Las in psyllid vector by tissue blots (squash blot) on nylon membranes. The procedure for isolating the nucleic acid from single psyllid was optimized, and we have been able to amplify CLas specific amplicons from single infected psyllids using pairs of Las specific primers. Conditions of amplifications were optimized with different primer pairs and now we have been able to amplify HLB specific amplicons without non-specific bands in PCR. In initial studies of whole psyllid tissue blots, hybridization signal was also observed with healthy psyllids (psyllid colony from the healthy psyllid containment facility). However use of specific primer pair corresponding to the EFTU gene of CLas has been promising and we will use the probe generated for this gene in tissue blots of psyllids henceforth. Success of this work would benefit the citrus industry by providing simple, sensitive and rapid detection method for large-scale detection of HLB under field conditions. Tissue imprints, once on the membrane could be stored for later use, can be shipped easily and generally are stable. The modified form of tissue blot, the ‘squash-blot’, for psyllids developed will be useful to understand the epidemiology of HLB. Two manuscripts have been submitted and have come back for revisions (one in Molecular and Cellular Probes and the second in the Journal of Bioscience).
Research results are very similar to those reported in the September 2012 quarterly report. We ran additional experiments similar to those described in the previous quarterly report. Because of the season, seeds were mature and more difficult to homogenize completely with or without enzyme treatment. We continued investigation with two experimental approaches: 1) enzyme treatment without subsequent physical pulverization of the tissue and 2) enzyme treatment followed by physical pulverization of the tissue. Fractionation on Percoll gradients yielded results similar to previous. Percoll fractions with the lowest Ct values (most bacteria) were analyzed by Fluorescence In Situ Hybridization (FISH) microscopy. The most intense fluorescence was observed with spherical bodies and a few bacilliform shapes. Because of the manner in which the experiments have to be performed, contamination is possible though precautions are take to minimize this occurrence. As before, the numbers of cells calculated from qPCR values for the Percoll and preceding fractions indicated more cells should be present than were observed in the sample observed by microscopy. We know that the disruption of the vascular bundles is not complete since filtration prior to Percoll fractionation traps tissue fragments that contain Liberibacter DNA; and, most likely Liberibacter cells. We have expanded our isolation work to include adult psyllids and psyllid nymphs. Psyllids are softer bodied and preliminary experiments suggest they homogenize more efficiently than plant tissues. Real-time PCR analysis of fractions from Percoll gradients suggest Liberibacter cells band differently when extracted from psyllids. Currently, fractions are being analyzed by FISH.
This is a project to continue one of the most fruitful leads that accidentally resulted from our previously funded work. We have found that citrus becomes a source of Huanglongbing (HLB) inoculum for spreading the disease to other plants much earlier than previously thought. The working hypothesis is that the female psyllid finds an area of new flush to lay her eggs. As she is laying eggs, she probes the phloem to feed and transfers Candidatus Liberibacter asiaticus (Las) to the tree. As the eggs develop into nymphs, Las begins to multiply in that localized area of the plant, where the new nymphs then feed and acquire Las. Thus, infection of only a micro area of flush tissue where the nymphs develop is sufficient for the first generation of psyllids to become infected and to be vectors to spread the disease to other trees. Thus, the time-period after a tree becomes infested by infected psyllids until it is a donor for other trees could be as short as 15-30 days or less. The limitation is actually the time for the second generation of psyllids to develop. We are continuing experiments to find ways to quickly detect psyllid reproduction as a method to detect early infections.
Goals 1, 2 and 3 are partially completed and continuing. We have cloned three predicted “late gene” promoter regions with promoter activity from SC1. The first (promoter P0) is located between locus tags gp125 and gp130 and includes a portion of the region annotated as gp125. This region was cloned in both directions upstream of the lacZ reporter gene in E. coli. In the early gene direction, this promoter region gave a light blue color reaction that we regard as a moderately constitutive promoter activity. In the late gene direction, this promoter region gave a medium blue color reaction that we regard as a stronger constitutive promoter activity. We subsequently cloned two additional potential promoters from the late gene region of SC1, the first located between gp120 and gp125 (promoter P1) and the second between gp115 and gp120 (promoter P2). Both of the promoters were cloned in the late gene direction. Only the construct (forming pSZ81) of P2 fused with the lacZ gene showed a medium blue color reaction equivalent to that of P0. Promoter P1 was not active. We then cloned the SC1 gp125 (annotated as a phage C2-like repressor) into pUFJ5, forming pUFZ3-4, and transformed into the strain containing pSZ81. No change in color reaction was observed, indicating that either the pUFZ3-4 clone was not functional, or the gp125 locus didn’t function as a repressor of P0. We then searched the Las genome (psy62) and found that besides gp125, there was only one additional locus annotated as a repressor in the genome, CLIBASIA_01645. This protein homologue is also present in Lso (CKC_01785) and Lam (LAMF_00067) and all three have a predicted Peptidase S24 LexA-like protein domain at their its 3’ends. LexA is a repressor of genes involved in the cellular SOS response to DNA damage. We designed primers to clone this C1 repressor from Las strain UF506 based on the pys62 genome sequence. The PCR product sequence revealed that the corresponding ORF in UF506 was truncated and missing the Peptidase S24 LexA-like protein domain that is predicted in Las strain psy62, Lso and Lam. This truncated UF506 ORF was cloned into pUFR047, forming pSZ83, and transformed into the strain containing pSZ81. We also cloned the nontruncated Lam homolog (LAMF_00067) into pUFR047, forming pSZ84, and transformed this clone into the strain containing pSZ81. Again, no change in color reaction was observed, indicating that neither functioned to repress the late gene promoter regions P0 or P2. In an effort to determine if the promoter regions examined were really in the ‘late’ region, we used semi-quantitative, real time PCR assays to determine relative expression levels of several different genes on both SC1 and SC2. Relative expression levels of SC2-gp095 (peroxidase), SC2-gp100 (glutathione peroxidase), SC1-gp110 (‘holin’) although nominally placed in the ‘late’ gene regions of the phage, were much higher (>100X) than expression levels of SC1-gp025 (‘tail fiber’). These results indicated that the promoter regions P0, P1 and P2 that we examined as predicted ‘late gene’ promoters, are likely not late gene promoters, nor repressed. Since repressors are known to work at multiple promoter sites, we are cloning additional predicted promoter regions, this time from genes including SC1-gp025, now experimentally shown to be expressed as a late gene.
The data analysis involved validation of LIBS spectral data with nutrient profiles. Each nutrient (N, P, K, C, B, Ca, Cu, Fe, Mg, Mn, Zn) was independently analyzed with the spectral data to verify whether the LIBS spectral data can be associated with nutrient concentration. Several statistical models such as Partial Least Square Regression (PLSR) and Support Vector Machine were tested on the processed LIBS spectral signals. Some of the LIBS data processing procedure which were performed were baseline correction, wavelet-based denoising for noise removal, resampling of LIBS spectra to reduce the number of spectral features, peak selection and alignment for extracting important peaks, etc. Although, the PLSR model showed a relationship between the LIBS spectra and nutrient concentrations showed good correlation between the actual and predicted values for various nutrients, the model could not be validated with an independent dataset. The major reason for this could be the low variation in the nutrient concentrations among the samples. For example, most nutrients, such as nitrogen, phosphorus, potassium, iron and manganese had nutrient concentrations in optimal or high ranges. Similarly, nutrients such as calcium and copper were either optimal or high/excess, respectively. A few other challenges that were identified in the process were: (i) several LIBS spectra replicates were collected per leaf sample, however, a minimum quantity of leaves (4-5 leaves) are required for chemical analysis, which made the comparison challenging; (ii) the LIBS spectra had to be averaged to compensate for limitations in chemical analysis results, which reduced the number of spectra to about 115 samples. A larger dataset may increase the robustness of the developed models; and (iii) the LIBS spectra have more than 6000 features. The tested feature selection methods were not be robust in representing the entire data; however using all spectral features will also not be feasible. Therefore, our future research will involve the following: (i) collecting a larger dataset, possibly including samples that represent different ranges (low, optimum and high) in order to increase the prediction/classification efficiency, (ii) collecting LIBS spectra from soil to verify if the system is more applicable for soil nutrient analysis, and (iii) improving the statistical analysis to develop a robust model with the right processing tools.
In this project we are developing a new method for rapid and efficient inoculation of plants with HLB based on a Pulse Micro Dose Injection System (PMDIS). In our preliminary experiments we have had some success in inoculation of Periwinkle plants using this technique, which suggested a feasibility of further adaptation of the PMDIS system for HLB inoculations. Our goal now is to optimize the protocol for PMDIS-based plant inoculation. We are conducting experiments in order to 1) identify what types of tissue within an infected citrus plant can serve as a good resource of the HLB bacteria for preparation of the inoculum by comparing extracts from stems, leaves and seed coats as inoculum sources; 2) examine whether HLB-infected psyllids can be utilized for preparation of the inoculum suspension; 3) optimize the composition of the extraction buffer used for preparation of the bacterial suspension and the extraction conditions, so they would support high efficiency of the PMDIS-mediated transmission of the pathogen; 4) optimize the parameters of injection. We are also evaluating how age of receptor plants, types of citrus varieties used as HLB bacterium donors as well as for plants being inoculated, types of flushes being inoculated affect efficiency of inoculation. Several sets of plants have been already injected using PMDIS. Those are being maintained in the greenhouse and monitored for the disease development.
The overall goal of this project is to improve the efficacy of the copper loaded silica nanogel (CuSiNG) formulation for preventing citrus canker disease incidence. 2012 field trial data has been compiled. The data show that CuSiNG pH7 opaque formula demonstrated superior efficacy against preventing incidence of young canker lesions, scab and melanose. The formula exhibited no Cu toxicity. The overall efficacy of the CuSiNG pH7 formula was comparable and in some case better than a number of leading commercial products (at 1.0 lb/acre metallic Cu content). The efficacy was not compromised even at lower metallic Cu concentration of the formula when compared to other commercially available products. Field trial was also conducted on CuSiNG pH 4 which is a transparent water-soluble formula. The metallic Cu content was 0.2 lb/acre in CuSiNG pH 4 formula. Results indeed show that the formula was effective in preventing canker incidence. However, the overall efficacy was not as good as CuSiNG pH 7 formula. This suggests that Cu rate at 0.2 lb/acre was not adequate to fully protect against canker incidence. CuSiNG pH 4 formula did not exhibit any phytotoxicity. Future studies/trials will focus on transparent CuSiNG pH 4 formula to improve its efficacy without increasing metallic Cu content. We also plan on conducting trial in the upcoming season using mixed-valence CuSiNG formulations. Preliminary results were encouraging as improved efficacy was observed in laboratory conditions. Due to high surface area of the silica nanogel material, improved adherence to citrus plant is expected. We have conducted adherence studies using surface modified yellow-emitting CdS:Mn/ZnS quantum dots (a photostable semiconductor fluorescent nanocrystals of 3.5 nm size). Results indeed show that the silica nanogel material itself (in absence of Cu) strongly adhere to the plant surface. We are also conducting adherence studies using red-emitting (Cu insensitive) Qdots where Qdots are embedded in CuSiNG materials. Results from this study will directly demonstrate the adherence property of the CuSiNG nanoformulations.
Citrus huanglongbing (HLB) is associated with three species of Liberibacter’Candidatus Liberibacter asiaticus (Las), Ca. L. americanus, and Ca. L. africanus. The majority of the testing in Florida is focused on detection of Las, the only bacterium known to be associated with HLB in Florida to date. Over the past four years with funding from Citrus Research Board, we have conducted regular surveys of citrus and citrus relatives in Florida, from various germplasm collections, backyard plants, native and cultivated trees, testing for tolerance to HLB. We have focused on plants showing HLB symptoms but testing negative by standard qPCR tests. A small selected set of symptomatic, qPCR negative samples were analyzed for detection of other genomic regions of Liberibacters by conventional PCR (cPCR), cloning and sequencing. This study confirmed the presence of Liberibacter variants not detectable by standard assays. The purpose of this project is to conduct further research on variants of Liberibacters from citrus and citrus relatives and to develop rapid methods for detection of these variant populations. We also will study the biology of the variants under greenhouse conditions, determine the changes in Liberibacter populations within individual trees over time from analysis of DNA extractions we have made over the past 5 years, and determine if there are interactions among populations of Liberibacter variants which may ameliorate/enhance the symptoms of HLB. Understanding of HLB disease complex caused by all variants of Liberibacters will be useful for developing novel disease management strategies. Research on this project this third quarter has involved molecular characterization of samples collected in October in Southern and Central Florida. Research has included development of a unique methodology to selectively prepare DNA from Liberibacter and Liberibacter-like pathogens for genome sequencing. Observations are being made of symptoms produced on citrus in the field and greenhouse conditions.
The effectors of HLB engineered into Citrus Tristeza Virus (CTV) vector have been transferred into citrus as outlined in the previous research updates and are being screened for their response to HLB pressure.They include the following; CLIBASIA_05165, CLIBASIA_05605, CLIBASIA_01555, CLIBASIA_05195, CLIBASIA_05200, CLIBASIA_05620, CLIBASIA_05635, CLIBASIA_05665, CLIBASIA_05130, CLIBASIA_05155, CLIBASIA_05265, CLIBASIA_05560, CLIBASIA_05150, CLIBASIA_05180, CLIBASIA_05245, CLIBASIA_02250, CLIBASIA_03020, CLIBASIA_03025, CLIBASIA_02090 and CLIBASIA_02905. At this stage we are studying the interaction of the plants expressing HLB effectors and their response to HLB infection through psyllids. A potential useful outcome of this project is the finding that CTV-vector can be used as an RNAi vector. This has been demonstrated to silence citrus endogenous Phytoene desaturase (PDS), delta amino levulinic acid dehydratase (d ALA dehydaratase) genes, and has also be used to potentially silence endogenous genes of psyllids (vecotr of HLB) in co-operation with scientists at CREC and USDA, Fort Pierce. We have positive feedbacks from these groups on the successful silencing of psyllid endogenous genes. This line of research potentially holds great promise in mitigating the spread of HLB by psyllid vector. An oral presentation based on this investigation will be presented in the upcoming 2013 HLB conference in Orlnado. Shubash Hajeri, Choaa El-Mohtar, William O. Dawson and Siddarame Gowda. Citrus tristeza virus-based RNA-interference (RNAi) vector and its potential in combating citrus Huanglongbing (HLB).
To date, 793 medium formulations have been tested for their ability to support growth of Candidatus Liberibacter asiaticus. Growth has been observed over 2-3 low level passages to fresh media for periods of over two months. Attempts are now aimed at obtaining vigorous growth. Insect tissue culture media have been used as the source of the basic nutritional requirements of the bacterium. DS2 tissue culture medium (Mediatech) has appeared advantageous; however, the formulation of this medium is proprietary. So attempts are presently being made to find a suitable replacement for DS2 medium that has a freely known formulation. The effects of complex reagents, such as Bacto Proteose Peptone No.3, Phytone Peptone, and Brain Heart Infusion Broth, are also being evaluated. Presently, a medium formulation containing Graces tissue culture medium instead of DS2 medium, and supplemented with, among other ingredients, a mixture of lipids, sterol, and pluronic acid, has produced good early results and is being evaluated for continuous growth support.
PCR test results for the 90 leaf samples collected in Aug. 2012 were received and compared with starch measurement results. The cycle threshold (CT value) which is the number of necessary cycles for the fluorescent signal to cross the threshold was used to decide the samples’ infection status. The CT value below 33 showed HLB infection of the samples. The PCR results were used as a reference for defining training and validation sets. The starch concentration for healthy samples should not exceed 5 ‘g/mm2, however, there were some healthy samples with starch density of more than 5 ‘g/mm2 which means that the reason of starch accumulation was not HLB infection. There was also one HLB infected sample with less than 5 ‘g/mm2 of starch density which could be considered as an experimental error. A pixel based analysis was performed to determine the HLB detection ability in each imaging condition. Two clusters were defined for a K-means algorithm, and the clustered pixels were evaluated to determine whether the pixels belonging to the same class (with prior knowledge) were clustered in the same cluster as well. The investigation showed that only the 591 nm (Min) images contained very useful information and classes were distinctive, however, for some samples, the general average intensity of the image was higher and led to a misclassification of some healthy pixels in other classes, and still the healthy and not-healthy (including HLB infected and zinc deficient) areas were distinguishable, when the clustering was performed for individual samples separately. Then, the same textural features (as reported previously) were extracted only from 591 nm (Min) images and a step by step classification method was implemented, in which at the first step zinc deficiency was detected and HLB infection was identified at the second step. A best set of textural features was determined separately for each classification step using Bhattacharyya distance feature selection method. Also performances of seven classifiers (as reported previously) were evaluated to determine the best classification approach for each step. From the analyses results, zinc deficient samples were detected with an accuracy of 96.7%. HLB infection was also detected with average accuracies of 90% and 96.5% for zinc deficient and non-zinc deficient classes, respectively. However, the main purpose in this study was to detect HLB infection and not the zinc deficiency. Therefore, when only the HLB detection was considered, an average classification accuracy of 93.1% was obtained, including 97.9% and 88.4% accuracies for healthy and HLB infected leaves, respectively. A manuscript is currently being prepared based on the above results to be published in the Transactions of the ASABE journal. An abstract was submitted for the 2013 ASABE annual meeting in July 2013. In Dec. 2012, a new dataset of 96 samples containing 20 healthy, 20 magnesium deficient, 20 zinc deficient, 20 HLB infected, and 16 HLB infected & zinc deficient samples in a Valencia variety were collected from the CREC grove to be evaluated and compared with the previous dataset from a Hamlin variety. Starch measurement for these samples were completed and the samples were sent for a PCR test.
2012 LAS experiments have been completed and data analysis is being conducted. These experiments were similar in design to 2011 experiments, with the treatments being 1/3 King’s B (K) medium and commercial grapefruit juice (G) medium. However, the focus was shifted to analyses of the biofilm-like substance formed at the air-liquid interface of culture flasks. 2011 experiments showed that increased formation of this substance was found in juice-containing media, which correlated with prolonged LAS viability. In 2012 experiments, slides with the biofilm-like substance formed on them were stained with Calcofluor (stains exopolysaccharides white) and LIVE/DEAD BacLight (stains live cells green and dead cells red). Micrographs of stained slides showed that exopolysaccharides were present, indicating that the substance was a true biofilm. Results also showed that the majority of cells in biofilm from medium K were dead after several weeks, while the majority of cells were alive in medium G. Analyses are currently underway to determine the percentages of viable and nonviable LAS cells in this biofilm, as well as determine the presence of other potential “helper” bacteria. Analysis of Illumina MiSeq high-throughput sequencing data of 16S amplicons from experimental samples will soon be completed. Data from the initial citrus-seed inoculum showed that LAS was the largest bacterial component of the inoculum, but significant populations of several other bacterial endophyte species were also present in the seed coats. These other populations, if they also grow in the cultures, could have effects (either positive or negative) on the culture of LAS. The 16S amplicon data from biofilm samples will soon be analyzed. This will indicate if any of these other endophytes were able to grow in the LAS cultures. A draft of the publication for these data has been written. When the rest of the 2011 and 2012 biofilm bacterial community data is analyzed (qPCR and Illumina data) it will be added to the publication. The data will be presented at the 3rd International Conference on Huanglongbing to obtain any additional insights/comments from other researchers in the field. Subsequently, the publication should be ready for submission.
The objective of this project is to characterize the hypI (renamed as hyvI) gene and determine its effects on insect transmission and/or virulence in host plants. Transient expression with alternative expression systems and RT-qPCR, etc., will be used to elucidate the function of the hypI (hyvII) gene of Las and shed light on the molecular mechanism of this “phase variation” phenomenon; thereby developing a novel control strategy for citrus HLB. In addition, antibodies and probes along with standardized protocols developed during this project can be applied for better detection and differentiation of the HLB bacteria. The hyvI and hyvII within two Las prophages were further characterized and some of the results were published in Applied Environmental Microbiology 77:6663-6673, 2011. “Diversity and Plasticity of the Intracellular Plant Pathogen and Insect Symbiont “Candidatus Liberibacter asiaticus” as Revealed by Hypervariable Prophage Genes with Intragenic Tandem Repeats”. We have developed an improved real-time PCR using SYBR Green 1 (LJ900fr) and TaqMan’ (LJ900fpr) protocols with primers and probe targeting the nearly identical tandem repeats of 100bp hyvI and hyvII. The results were published in Molecular and Cellular probes, 26:90-98, 2012. Working with Mesa Tech International, Inc., we have developed a rapid and simple HLB detection kit for field testing, which only requires less than 30 minutes. Monoclonal antibodies against the partial HyvI protein (only one repeat) were generated, and their sensitivity and specificity were evaluated for the detection of HyvI protein expressed in E. coli and HLB-infected citrus and psyllids. All antibodies were able to recognize the E. coli expressed HyvI antigen, but were not able to detect the HyvI antigen from HLB-infected plants and psyllids. We have characterized two novel autotransporter proteins of ‘Candidatus Liberibacter asiaticus’ (Las), and redesignated them as LasAI and LasAII in lieu of the previous names HyvI and HyvII. Bioinformatic analyses revealed that LasAI and LasAII share the structural features of an autotransporter family containing large repeats of a passenger domain and a unique C-terminal translocator domain. When fused to the GFP gene and expressed in E. coli, the LasAI C-terminus and the full length LasAII were localized to the bacterial poles, similar to other members of autotransporter family. Despite the absence of the signal peptide, LasAI was found to localize at the cell surface by immuno-dot blot using a monoclonal antibody against the partial LasAI protein. Its surface localization was also confirmed by the removal of the LasAI antigen using a proteinase K treatment of the intact bacterial cells. When co-inoculated with a P19 gene silencing suppressor and transiently expressed in tobacco leaves, the GFP-LasAI translocator targeted to the mitochondria. This is the first report that Las encodes novel autotransporters that target to mitochondria. These findings may lead to a better understanding of the pathogenesis of this intracellular ‘energy parasitic’ bacterium and to characterizing new molecular targets for HLB control. We are currently conducting functional analysis and determination of the LasAI and LasAII functional domain(s) via mutagenesis.
The objective of this project is to characterize the hypI (renamed as hyvI) gene and determine its effects on insect transmission and/or virulence in host plants. Transient expression with alternative expression systems and RT-qPCR, etc., will be used to elucidate the function of the hypI (hyvII) gene of Las and shed light on the molecular mechanism of this “phase variation” phenomenon; thereby developing a novel control strategy for citrus HLB. In addition, antibodies and probes along with standardized protocols developed during this project can be applied for better detection and differentiation of the HLB bacteria. The hyvI and hyvII within two Las prophages were further characterized and some of the results were published in Applied Environmental Microbiology 77:6663-6673, 2011. “Diversity and Plasticity of the Intracellular Plant Pathogen and Insect Symbiont “Candidatus Liberibacter asiaticus” as Revealed by Hypervariable Prophage Genes with Intragenic Tandem Repeats”. We have developed an improved real-time PCR using SYBR Green 1 (LJ900fr) and TaqMan’ (LJ900fpr) protocols with primers and probe targeting the nearly identical tandem repeats of 100bp hyvI and hyvII. The results were published in Molecular and Cellular probes, 26:90-98, 2012. Working with Mesa Tech International, Inc., we have developed a rapid and simple HLB detection kit for field testing, which only requires less than 30 minutes. Monoclonal antibodies against the partial HyvI protein (only one repeat) were generated, and their sensitivity and specificity were evaluated for the detection of HyvI protein expressed in E. coli and HLB-infected citrus and psyllids. All antibodies were able to recognize the E. coli expressed HyvI antigen, but were not able to detect the HyvI antigen from HLB-infected plants and psyllids. We have characterized two novel autotransporter proteins of ‘Candidatus Liberibacter asiaticus’ (Las), and redesignated them as LasAI and LasAII in lieu of the previous names HyvI and HyvII. Bioinformatic analyses revealed that LasAI and LasAII share the structural features of an autotransporter family containing large repeats of a passenger domain and a unique C-terminal translocator domain. When fused to the GFP gene and expressed in E. coli, the LasAI C-terminus and the full length LasAII were localized to the bacterial poles, similar to other members of autotransporter family. Despite the absence of the signal peptide, LasAI was found to localize at the cell surface by immuno-dot blot using a monoclonal antibody against the partial LasAI protein. Its surface localization was also confirmed by the removal of the LasAI antigen using a proteinase K treatment of the intact bacterial cells. When co-inoculated with a P19 gene silencing suppressor and transiently expressed in tobacco leaves, the GFP-LasAI translocator targeted to the mitochondria. This is the first report that Las encodes novel autotransporters that target to mitochondria. These findings may lead to a better understanding of the pathogenesis of this intracellular ‘energy parasitic’ bacterium and to characterizing new molecular targets for HLB control. We are currently conducting functional analysis and determination of the LasAI and LasAII functional domain(s) via mutagenesis.
During summer, seeds generally are immature, relatively soft and easily homogenized. We tested two experimental approaches: 1) enzyme treatment without subsequent physical pulverization of the tissue and 2) enzyme treatment followed by physical pulverization of the tissue. For immature seeds, enzyme treatment followed by only extensive vortexing could cause extensive degradation of the tissues and release bacteria for further purification by filtration and centrifugation on a Percoll gradient. Enzyme treatment followed by physical disruption of tissue by treatment with a Polytron apparently gave at least equivalent tissue disruption if not slightly better based upon the calculated number of bacterial cells determined to be in the different portions of the Percoll gradient. Percoll fractions that were indicated to contain the highest numbers of cells as determined by qPCR were assayed with Fluorescence In Situ Hybridization (FISH) microscopy. The most intense fluorescence was observed with spherical bodies (not seen in negative control preparations); the bacilliform shapes seen in fresh tissues disrupted and chemically fixed in place (paraformaldehyde) were infrequent, suggesting a change in the shape of the bacteria occurs during extraction from the tissue. Also, the numbers of cells calculated from qPCR values indicated more cells should be present than were observed the microscope. We do not know the basis of this discrepancy.
(863) 956-8817
(863) 956-5894
700 Experiment Station Road
Lake Alfred, FL 33850
Email Us
