SEARCH

SEARCH BY CITATION

Keywords:

  • Candidatus Liberibacter americanus’;
  • Candidatus Liberibacter asiaticus’;
  • Citrus spp.;
  • Diaphorina citri;
  • inoculum sources;
  • Murraya exotica

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Two surveys (2005/2006 and 2009) were conducted in the state of São Paulo, Brazil, to investigate the incidence of ‘Candidatus Liberibacter asiaticus’ and ‘Ca. L. americanus’, two liberibacters associated with citrus huanglongbing (HLB) disease and both transmitted by Diaphorina citri, in orange jasmine (Murraya exotica), a widespread ornamental tree in cities and villages. The graft-transmissibility of the two species, and their DNA relatedness to citrus-associated liberibacters, were also investigated. Quantitative PCR was applied to PCR-positive orange jasmine and HLB-positive citrus growing in backyards and orchards to assess their inoculum source potentials. Liberibacters were detected in 91 of 786 sampled orange jasmine plants in 10 of 76 sampled locations. PCR-positive trees exhibited yellow shoots and/or dieback symptoms indistinguishable from those on PCR-negative trees. ‘Candidatus Liberibacter americanus’ was more common in 2005/2006 (96·6%) and ‘Ca. L. asiaticus’ in 2009 (84·8%). rplJ nucleotide sequences were identical within all populations of either species. Graft transmission succeeded only in homologous host combinations, including ‘Ca. L. americanus’ (2/10) from/to orange jasmine and ‘Ca. L. americanus’ (5/18) and ‘Ca. L. asiaticus’ (5/9) from/to citrus. Symptoms were mild and developed less rapidly in orange jasmine than in citrus, probably as a result of lower liberibacter multiplication rates. Respective titres of ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ in orange jasmine averaged 4·3 and 3·0 log cells g−1 tissue, compared with 5·5 and 7·3 in citrus. The results indicate that orange jasmine does not favour liberibacter multiplication as much as citrus. However, its importance in HLB epidemics should not be underestimated as it is a preferred host of D. citri and is not under any strict tree-eradication programme or measures for insect control.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Liberibacters are Gram-negative, phloem-limited, insect-transmitted α-Proteobacteria associated with huanglongbing (HLB), the most destructive citrus disease worldwide (Bové, 2006). Two ‘species’ are present in Brazil: ‘Candidatus Liberibacter asiaticus’ and the more recently described ‘Ca. L. americanus’ (Coletta-Filho et al., 2004; Teixeira et al., 2005b). Both are transmitted by the psyllid Diaphorina citri (Capoor et al., 1967; Yamamoto et al., 2006), an insect with a high reproductive capacity, especially when the environmental conditions favour the sprouting of young citrus leaves, where the females lay their eggs. On citrus, Husain & Nath (1927) recorded up to 807 eggs per female, Pande (1971) recorded 180–520 per female, and Huang (1990) reported a maximum fecundity of nearly 1900 eggs.

Orange jasmine (Murraya paniculata; Rutaceae) is a preferred host of D. citri (Halbert & Manjunah, 2004). The form present in Brazil, and widely grown throughout the world as a garden ornamental, resembles M. paniculata var. exotica as described and illustrated by Huang (1959) and M. exotica as described and illustrated by Huang (1997), but not M. paniculata as described and illustrated by Huang (1959, 1997). Its small relative size, abundant dark green leaves and aromatic white flowers have made it a popular ornamental tree in Brazil, where it is commonly found in streets, cemeteries and gardens. It is the form to which most host records for D. citri probably pertain.

Although genetically closely related, ‘Ca. L. asiaticus’ and ‘Ca. L. americanus’ differ considerably in two important aspects. First, ‘Ca. L. asiaticus’ tolerates higher temperatures than ‘Ca. L. americanus’ (Lopes et al., 2009b); multiplication of ‘Ca. L. americanus’ was limited when infected plants were maintained at 32°C for 6 h daily for 60 consecutive days, leading to permanent symptom remission; in contrast, a temperature of 38°C for the same duration only led to temporary symptom remission in ‘Ca. L. asiaticus’-affected trees. Secondly, ‘Ca. L. asiaticus’ is more adapted to citrus and more efficiently transmitted by grafting and D. citri than ‘Ca. L. americanus’ (Lopes et al., 2009a; P.T. Yamamoto, Fundecitrus, personal communication). Average bacterial titres, in log cells per gram of leaf midrib, were 6·4 for ‘Ca. L. asiaticus’ and 4·9 for ‘Ca. L. americanus’ in graft-inoculated experimental plants, and 6·7 for ‘Ca. L. asiaticus’ and 5·7 for ‘Ca. L. americanus’ in the naturally infected field trees used as the source of inoculum. Such variations may explain the contrasting competitive behaviour of these liberibacters in the field. A shift in liberibacter prevalence and an uneven spatial dynamics, both favouring ‘Ca. L. asiaticus’, have been observed in Brazil (Lopes et al., 2007, 2009a).

Conflicting information exists in the literature as to whether or not orange jasmine is an alternative host of liberibacters in other countries. All reported work refers to ‘Ca. L. asiaticus’ only. In Indonesia, orange jasmine seedlings exposed to D. citri obtained from field-grown citrus trees showed typical external and internal symptoms 10 months post-inoculation (Tirtawidjaja, 1981). In Thailand, however, no abnormalities were detected on orange jasmine or curry leaf (Bergera koenigii) plants 8 months after infectious psyllids fed on them (Koizumi et al., 1996). These studies should be interpreted with caution since no specific technique for liberibacter detection was used for plant evaluations. In Taiwan, plants graft-inoculated with HLB-positive sweet orange (Citrus × aurantium syn. C. sinensis) budwoods, and tested with a DNA probe, remained free from liberibacter for at least 1 year (Hung et al., 2000). A similar result was obtained in Japan for graft-inoculated plants assessed by PCR (Dai et al., 2005). In China, ‘Ca. L. asiaticus’ was successfully detected in city trees with the use of nested, but not standard, PCR (Li & Ke, 2002). In the USA, successful liberibacter transmission was reported from citrus to orange jasmine via psyllids (Damsteegt et al., 2007) and from orange jasmine to citrus by the parasitic plant, dodder (Zhou et al., 2007).

The HLB-associated bacteria ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ have been detected in orange jasmine trees in Brazil (Lopes et al., 2005, 2006a). The first suspicious tree was found in October 2004, at the citrus farm most affected by HLB at that time. The tree exhibited yellow leaves and shoot dieback throughout the canopy. Standard PCR revealed the presence of ‘Ca. L. americanus’ in the leaves with symptoms, and in seven lots of 10 adult psyllids captured from the same tree (Lopes et al., 2005). Gasparoto et al. (2010) recently reported transmission of ‘Ca. L. americanus’ from infected orange jasmine to sweet orange plants by D. citri. These findings, along with the widespread occurrence of orange jasmine in cities located in the main citrus-growing areas of the state of São Paulo, and the lack of existing information on orange jasmine as a host of HLB bacteria in Brazil, prompted the present study. The objectives were: (i) to determine the incidence and progress over time of infected orange jasmine trees in urban areas, (ii) to assess and compare bacterial populations present in naturally infected orange jasmine and citrus trees, (iii) to determine the efficiency of liberibacter transmission in graft-inoculation experiments, and (iv) to determine levels of DNA relatedness between orange jasmine- and citrus-associated liberibacters. Preliminary data from this study have been published (Lopes et al., 2005, 2006a,b).

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Orange jasmine surveys

Surveys were carried out in 2005/2006 in 10 cities and villages and in 2009 in 76 cities and villages located mainly in the central region of the state of São Paulo (Fig. 1). Whilst most locations were surveyed only in 2009, Américo Brasiliense, Araraquara, Cravinhos, Matão, Motuca, Rincão and Santa Lúcia were surveyed on both occasions. Citrus orchards affected by HLB surrounded all surveyed locations, although the incidence of affected citrus trees in the orchards, and their distances from urban areas, varied considerably. Samples were collected only from suspicious trees, i.e. those with shoots showing variable degrees of yellowing and/or shoot dieback. Samples from each tree consisted of 10–20 compound leaves from branches with symptoms. DNA was extracted from the samples using the CTAB method of Murray & Thompson (1980) as described by Teixeira et al. (2005a), then analysed by PCR for liberibacter detection and quantification as described below. The locations of infected surveyed trees in the state of São Paulo are shown in Figure 1. During the surveys, citrus trees with apparent symptoms of HLB were also sampled. These trees comprised known lime [Mexican (C. × aurantiifolia), Tahitian (C. × latifolia) and Rangpur (C. ×limon)] and mandarin (C.reticulata) cultivars, and unknown sweet oranges (C.aurantium) cultivars. Most of these trees were growing in backyards. The sampling and bacterial detection procedures were the same as those used for orange jasmine.

image

Figure 1.  Map of the state of Sao Paulo, Brazil, showing locations where orange jasmine (Murraya exotica) trees were sampled during 2005/2006 and 2009, with, in parentheses, the total number of trees sampled and the numbers infected by ‘Candidatus Liberibacter americanus’ (Lam) and/or ‘Ca. L. asiaticus’ (Las). Circled numbers are locations where infected trees were found. 1 Água Vermelha (5, 2 Lam) 2 Aguaí (5) 3 Agudos (3) 4 Américo Brasiliense (38, 4 Lam), 5 Analândia (16), 6 Anhembi (4), 7 Araraquara (475, 24 Lam, 20 Las), 8 Arealva (5), 9 Artemis (4), 10 Artur Nogueira (3), 11 Bariri (13), 12 Barra Bonita (6), 13 Barrinha (2), 14 Bocâina (10), 15 Borborema (1), 16 Botafogo (5), 17 Brotas (5), 18 Bueno de Andrada (28, 6 Lam), 19 Cachoeira de Emas (7), 20 Casa Branca (5), 21 Charqueada (2), 22 Córrego Rico (4), 23 Corumbataí (7), 24 Cravinhos (22, 5 Lam), 25 Dourado (12), 26 Gavião Peixoto (5), 27 Guarapiranga (12), 28 Guarapuâ (5), 29 Guariroba (2), 30 Holambra (4), 31 Iacanga (2), 32 Ibaté (15), 33 Ibiriçá (2), 34 Ibitinga (2), 35 Ibitiúva (2), 36 Igaraçú do Tietê (8), 37 Igaraí (2), 38 Iracemápolis (6), 39 Itajú (3), 40 Itirapina (6), 41 Jacuba (1), 42 Juquiratiba (1), 43 Lagoa Branca (2), 44 Iaras (2), 45 Luiz Antonio (9), 46 Lupércio (3), 47 Luzitânia (3), 48 Matão (130, 5 Lam, 5 Las), 49 Mococa (6), 50 Monte Alto (8), 51 Motuca (27, 7 Lam, 2 Las), 52 Nova Europa (13), 53 Pederneiras (5), 54 Pirambóia (1), 55 Pirangi (1), 56 Porto Ferreira (6), 57 Pratânia (1), 58 Ribeirão Bonito (11), 59 Rincão (25, 2 Las), 60 Santa Cruz da Estrela (4), 61 Santa Cruz das Palmeiras (7), 62 Santa Ernestina (4), 63 Santa Eudóxia (10), 64 Santa Lúcia (18, 3 Lam, 1 Las), 65 Santa Rosa do Viterbo (11), 66 São Manoel (5), 67 São Pedro (5), 68 São Simão (8), 69 Silvânia (18, 5 Lam), 70 Tabatinga (9), 71 Tambaú (8), 72 Taquaral (3), 73 Torrinha (4), 74 Trabijú (2), 75 Ubirajara (3), 76 Vista Alegre (1).

Download figure to PowerPoint

Graft inoculations and plant evaluations

Two distinct methods were used in the greenhouse to attempt to transmit liberibacters from orange jasmine to citrus and vice versa. The first method consisted of side-grafting a budstick to the stem of a healthy young plant, essentially as described by Lopes & Frare (2008). Briefly, 4-cm-long segments of liberibacter-positive shoots removed from field-grown trees were longitudinally sliced, inserted and fixed separately, with plastic bands, onto sliced segments of the stems of healthy young plants. The second method consisted of slicing 8- to 10-cm-long stem segments of young liberibacter-positive potted citrus plants and of healthy potted orange jasmine plants, then binding the sliced portions together with plastic bands for 2 months. The purpose of this procedure was to enhance connection of the vascular systems so as to permit exchange of vascular sap between the two plants growing in their own substrates. The first method was used in cross-inoculation experiments in which ‘Ca. L. asiaticus’- or ‘Ca. L. americanus’-infected budsticks were grafted onto plants in homologous (citrus–citrus) and heterologous (citrus–orange jasmine and vice versa) combinations. The second method was used in an experiment with only HLB-positive citrus as the source of liberibacters. All experimental plants were routinely evaluated for symptom expression and, 6 and 12 months after inoculation, 5–10 mature leaves were collected and processed for DNA extraction and detection and quantification of liberibacters. The PCR-positive orange jasmine trees were subjected to further standard and quantitative PCR (qPCR) testing.

Liberibacter detection and quantification

All samples from field surveys and experimental plants were tested for the presence of liberibacter using standard duplex PCR, as described by Teixeira et al. (2009), using the primer pairs GB1/GB3 (Teixeira et al., 2005a) and A2/J5 (Hocquellet et al., 1999). All samples that tested positive were further analysed by qPCR for quantification of liberibacter DNA (Teixeira et al., 2008b), essentially as described by Lopes & Frare (2008), employing a standard curve with a single recombinant plasmid containing both ‘Ca. L. asiaticus’ and ‘Ca. L. americanus’ amplicons. To generate the recombinant plasmid, the primer pair f-rplJAm/r-rplJAm (Teixeira et al., 2008b) was used to amplify a 127-bp fragment from ‘Ca. L. americanus’. PCR was run for 40 cycles of 98°C for 10 s, 62°C for 15 s and 72°C for 15 s, preceded by a denaturation step of 98°C for 30 s and followed by an elongation step of 72°C for 5 min. Reaction mixtures of 50 μL contained 100 ng total DNA (DNA of periwinkle infected with ‘Ca. L. americanus’; Wulff et al., 2009), primers at 400 nm each, 200 μm dNTPs and 0·8 U Phusion polymerase in 1× buffer (Finnzymes). Amplified product of ‘Ca. L. americanus’ was purified using the Wizard SV Gel kit and PCR Clean-up System (Promega), phosphorylated with T4 polynucleotide kinase (Gibco BRL) and cloned into plasmid pBS, previously cut with SmaI with bacterial alkaline phosphatase (BAP)-phosphorylated ends. This plasmid (now named pBSQRT Am#1) was sequenced with primers m13 and T7 promoter and the sequences aligned with codoncode aligner (CodonCode Corporation). The ‘Ca. L. asiaticus’ PCR product was first cloned into pGEMT-Easy (Lopes et al., 2009a), excised with EcoRI, and inserted into the EcoRI restriction site of pBSQRT Am#1, after BAP treatment of vector ends. The resulting plasmid (hereafter named pBSQRT Am/As #10) was sequenced as described above. The ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ PCR products were inserted in opposite orientations into pBSQRTAm/As #10. Plasmid DNA was prepared with the Midi prep kit (Qiagen), linearized with SalI, cleaned with phenol and quantified with a NanoDrop1000 spectrophotometer (Thermo Fisher Scientific). Conversion of micrograms of dsDNA to pmoles was as described by Lopes et al. (2009a). qPCR amplifications of the orange jasmine and citrus samples were performed with Step One Plus (Applied Biosystems) using 1× Power SYBR green master mix (Applied Biosystems). Concentration and purity of the DNA samples were determined using the NanoDrop spectrophotometer. Prior to qPCR, DNA was diluted to 100 ng μL−1. Each sample was run at least twice through standard or qPCR, along with positive (DNA from HLB-positive citrus), healthy and water controls. A 10-fold serial dilution of the pBSQRT Am/As #10 plasmid, ranging from 2·6 × 106 to 2·6 × 101 DNA copies, was prepared and employed to generate the standard curve. The target genes of ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ used in this work are single-copy numbers in their respective genomes (Wulff et al., 2009). Liberibacter bacterial titres were estimated by determining the number of liberibacter DNA copies in the 500 ng total DNA added to the reaction mix and the total amount of DNA estimated per gram of leaf midribs.

Sequencing and analysis of the rplJ gene

The purpose of this part of the study was to examine potential variability among orange jasmine- and citrus-associated liberibacters based on sequence analysis of the rplJ gene. The study included 31 samples of orange jasmine (27 of ‘Ca. L. americanus’- and four of ‘Ca. L. asiaticus’-infected trees) from the city of Araraquara and 78 samples of sweet orange (52 of ‘Ca. L. americanus’- and 27 of ‘Ca. L. asiaticus’-infected trees) from farms located in the north, centre and south of São Paulo state. Primers were designed based on sequences of the rplA–rplJ intergenic region and of the rplL gene of the β operon to amplify the rplJ gene (Table 1). This gene was chosen because of its low similarity and identity values between the ‘Ca. L. americ-anus’ and ‘Ca. L. asiaticus’ associated with HLB in Brazil (Teixeira et al., 2008a). Standard PCR was carried out in a 50-μL reaction mixture containing 1 μL total DNA, 1× PCR buffer (Invitrogen), 2 mm MgCl2, dNTPs at 0·2 mm each, primers at 500 nm each and 1·5 U Taq DNA polymerase. With primers A4AmF/J7AmR, the programme consisted of 35 cycles of 95°C for 45 s, 58°C for 45 s and 72°C for 1 min 30 s, plus a final extension of 72°C for 10 min. The expected amplicon size was 777 bp. For primers AsiDF/AsiDR, the programme was the same except that the annealing temperature was 62°C and the expected amplicon size was 774 bp. Each PCR reaction was analysed by 1% agarose gel electrophoresis. PCR products were purified using the Wizard PCR clean-up system (Promega) and sequenced in an ABI 3730xl capillary sequencing machine (Applied Biosystems) using the BigDye® Terminator 3·1 Cycle Sequencing Kit (Applied Biosystems), following the manufacturer’s protocols. PCR products sequenced to a minimum of two independent sequences of the same DNA fragment were assembled using sequencher 3·1 (Gene Codes Corporation) and edited when necessary, based on visual inspection of the chromatograms. Consensus sequences from each strain were aligned and compared using the same software.

Table 1.   Primers used to determine genetic relationships among ‘Candidatus Liberibacter americanus’ and ‘Ca. L. asiaticus’ associated with citrus and orange jasmine (Murraya exotica) trees in Brazil
PrimerLocationSequence (5′–3′)Target
A4AmFrplA-rplJ intergenic regionTTCTAGGTATTTCCTTGCTTGGene rplJ of ‘Ca. L. americanus’
J7AmRrpl LTACCTAATTTCTCCTCCAGCT
AsiDFrplA-rplJ intergenic regionTTGTAAGGGATGCGTTAGGATGene rplJ of ‘Ca. L. asiaticus’
AsiDRrpl LCACCCCATTCCTTTTCTAATC

Statistical analysis

The sas system (SAS Institute) was used for statistical analysis of the data on bacterial populations in naturally infected citrus and orange jasmine trees. The F-test was used for analysis of variance and Tukey’s studentized range test (P = 0·05) was used to separate means.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Orange jasmine surveys

Orange jasmine trees infected with liberibacters were found in 10 of the 76 sampled locations (Fig. 1), within a radius of approximately 35 km in the geographic centre of São Paulo state. Across both surveys, ‘Ca. L. americanus’ was detected in 61 (7·8%) and ‘Ca. L. asiaticus’ in 30 (3·8%) of 786 sampled trees. Apart from one tree in the city of Araraquara, all the remaining trees were infected by only one species of liberibacter.

In 2005/2006, Água Vermelha and Cravinhos were the places with the highest liberibacter incidence (40·0%), followed by Motuca (35·0%), Silvânia (27·8%), Bueno de Andrada (21·4%), Américo Brasiliense (16·7%), Santa Lúcia (11·1%), Araraquara (9·0%) and Matão (6·1%) (Table 2). Rincão was the only place where no infected trees were found. ‘Candidatus Liberibacter americanus’ was detected in 56 trees distributed in nine locations, with 11·7% incidence, but ‘Ca. L. asiaticus’ in only two trees, both in Araraquara, with 0·4% incidence. ‘Candidatus Liberibacter americanus’ was present in 96·6% and ‘Ca. L. asiaticus’ in 3·4% of all PCR-positive orange jasmine trees.

Table 2.   Numbers (with percentages in parentheses) of orange jasmine (Murraya exotica) trees infected by ‘Candidatus Liberibacter americanus’ and ‘Ca. L. asiaticus’ in urban areas of São Paulo state during 2005/2006 and 2009
Location2005/20062009
Total treesInfected by ‘Ca. L. americanus’Infected by ‘Ca. L. asiaticus’Total treesInfected by ‘Ca. L. americanus’Infected by ‘Ca. L. asiaticus’ e
Água Vermelha52 (40·0)0
Américo Brasiliense244 (16·7)01400
Araraquara26822 (8·2)2 (0·8)2072 (1·0)18 (8·7)
Bueno de Andrada286 (21·4)0
Cravinhos104 (40·0)0121 (8·3)0
Matão825 (6·1)04805 (10·4)
Motuca207 (35·0)0702 (28·6)
Rincão13001202 (16·7)
Santa Lúcia91 (11·1)092 (22·2)1 (11·1)
Silvânia185 (27·8)0
Total7756 (11·7)2 (0·4)3095 (1·6)28 (9·1)

In 2009, Santa Lúcia was the most affected location, with 33·3% incidence, followed by Motuca (28·6%), Rincão (16·7%), Matão (10·4%), Araraquara (9·7%) and Cravinhos (8·3%). No infected trees were detected in Américo Brasiliense, nor in any of the remaining 69 surveyed locations. In the small villages of Água Vermelha, Bueno de Andrada and Silvânia, all existing trees with symptoms had already been sampled in 2005/2006. In 2009, ‘Ca. L. americanus’ was detected in five trees (1·6% incidence) in three locations, and ‘Ca. L. asiaticus’ in 28 trees (9·1% incidence) in five locations. In this year, ‘Ca. L. americanus’ was present in 15·2% and ‘Ca. L. asiaticus’ in 84·8% of all PCR-positive orange jasmine trees.

No obvious differences in size and symptom severity were observed between locations and surveys, at the time samples were collected, amongst ‘Ca. L. americanus’- infected orange jasmine trees, nor amongst ‘Ca. L. asiaticus’-infected ones. Also, no visible differences were detected between PCR-positive and PCR-negative trees. However, visible differences in symptom severity were found between ‘Ca. L. americanus’- and ‘Ca. L. asiaticus’-infected groups (Fig. 2). Trees infected with ‘Ca. L. americanus’ were, on average, 2·4 m high, with 27·7% showing just yellow leaves, 38·3% just shoot dieback, and 34·0% yellow leaves plus shoot dieback in one or more sectors of the tree canopy (Fig. 2a). Trees infected with ‘Ca. L. asiaticus’ were, on average, 2·2 m high, with 83·3% showing just yellow leaves, and 16·7% yellow leaves plus shoot dieback in one or more sectors (Fig. 2b); no tree in this group exhibited shoot dieback. PCR-negative trees were, on average, 2·5 m high, with 44·9% showing just yellow leaves, 21·6% just shoot dieback, and 33·5% yellow leaves plus shoot dieback in one or more sectors of the tree canopy. Mottled leaves and fruits with aborted seeds, both characteristic of HLB in citrus (da Graça, 1991; Bové, 2006) were not observed in the PCR-positive orange jasmine trees. The yellowing of orange jasmine leaves included a variety of chlorotic patterns, most frequently that resembling nitrogen deficiency.

image

Figure 2.  Orange jasmine (Murraya exotica) trees with huanglongbing symtoms that tested positive for ‘Candidatus Liberibacter americanus’ (a) and ‘Ca. L. asiaticus’ (b) during the 2005/2006 survey carried out in the city of Araraquara, SP, Brazil.

Download figure to PowerPoint

During the 2005/2006 survey, leaf samples were also collected from backyard trees of sweet orange and Ponkan mandarin exhibiting yellow shoots. In total, 54 trees were sampled in Araraquara, four in Bueno de Andrada and 21 in Silvânia. ‘Candidatus Liberibacter americanus’ was detected in nine trees (11·4%) (seven of unknown sweet orange cultivars and two of Ponkan mandarin) and ‘Ca. L. asiaticus’ in four (5·1%) sweet orange trees. As in 2005/2006, leaf samples also were collected from backyard citrus trees in 2009, all from the city of Araraquara. However, in contrast to the 2005/2006 survey, all sampled trees exhibited characteristic blotchy mottled leaves on the yellow shoots. In 2009, ‘Ca. L. asiaticus’ was detected in all trees, which included seven Rangpur limes, 14 sweet oranges, two Mexican limes, three Tahitian limes, and two Ponkan and two Cravo mandarins.

To assess the potential of infected orange jasmine trees as reservoirs of liberibacters, qPCR was used to further analyse the DNA samples of all surveyed trees that tested positive by standard PCR (Fig. 3). DNA samples from HLB-positive citrus trees from commercial fields were also included for comparison. In orange jasmine, higher populations of bacteria were found in ‘Ca. L. americanus’- than in ‘Ca. L. asiaticus’-infected trees (Table 3). Average Ct values and estimated bacterial titres (log cells g−1 tissue were, respectively, 33·5 and 4·3 for ‘Ca. L. americanus’ and 35·4 and 3·0 for ‘Ca. L. asiaticus’. In contrast, higher populations were found in ‘Ca. L. asiaticus’- than in ‘Ca. L. americanus’-infected citrus trees, regardless of whether they were growing in backyards or in orchards. On backyard trees, average Ct and bacterial titres were, respectively, 26·7 and 6·2 for ‘Ca. L. americanus’ and 22·4 and 7·5 for ‘Ca. L. asiaticus’. For orchard trees, the respectively values were 30·7 and 4·9 for ‘Ca. L. americanus’, and 26·4 and 6·5 for ‘Ca. L. asiaticus’. Since the orange jasmine and the backyard citrus trees were growing in the same environment (urban areas), a further comparison was made between the average liberibacter populations of orange jasmine trees and the average liberibacter populations from combined data of all citrus trees. Significantly higher titres were observed in citrus for both liberibacters (6·2 for ‘Ca. L. americanus’ and 7·5 for ‘Ca. L. asiaticus’ in citrus, compared with 4·3 and 3·0, respectively, in orange jasmine).

image

Figure 3.  Agarose gel showing correspondence of band intensity and cycle threshold values obtained with standard and real-time quantitative PCR, respectively, for DNA samples of ‘Candidatus Liberibacter asiaticus’- or ‘Ca. L. americanus’-infected orange jasmine (Murraya exotica) and sweet orange trees.

Download figure to PowerPoint

Table 3.   Quantitative PCR cycle threshold values and estimated populations of ‘Candidatus Liberibacter americanus’ and ‘Ca. L. asiaticus’ in naturally infected orange jasmine (Murraya exotica) and citrus trees growing in urban areas and commercial fields in the central region of São Paulo state
LocationLiberibacter hostLiberibacter speciesNumber of treesAverage Ct values ± standard erroraAverage titres ± standard errora
  1. aAverage Ct values and titres (log cells g−1 tissue) with different letters are significantly different (< 0·05). Comparisons were made between liberibacter species within host.

Urban areasOrange jasmineCa. L. americanus’4833·5 ± 2·21 b4·3 ± 0·68 a
Ca. L. asiaticus’2735·4 ± 2·24 a3·0 ± 0·92 b
Sweet orange‘Ca. L. americanus’726·0 ± 1·21 a6·5 ± 0·31 b
(unknown cultivars)Ca. L. asiaticus’1822·4 ± 1·95 b7·6 ± 0·54 a
Ponkan mandarinCa. L. americanus’229·2 ± 0·18 a5·6 ± 0·08 b
Ca. L. asiaticus’220·1 ± 1·05 b8·0 ± 0·33 a
Commercial fieldsValencia sweet orangeCa. L. americanus’628·6 ± 3·59 a5·5 ± 0·94 b
Ca. L. asiaticus’725·4 ± 2·57 a6·8 ± 0·86 a
Valencia Americana sweet orangeCa. L. americanus’832·3 ± 2·91 a4·4 ± 0·90 b
Ca. L. asiaticus’1027·2 ± 2·88 b6·3 ± 0·85 a

For individual citrus categories/cultivars infected by ‘Ca. L. americanus’, average overall titres were 6·5 for sweet oranges and 5·6 for Ponkan mandarin growing in backyards, and 4·4 for Valencia Americana and 5·6 for Valencia sweet orange growing in orchards. For those infected by ‘Ca. L. asiaticus’, average titres were 7·6 for sweet oranges, 8·0 for Ponkan mandarin, 8·0 for Cravo mandarin, 7·0 for Cravo Rangpur lime, 7·3 for Mexican lime and 7·6 for Tahitian lime growing in backyards, and 6·8 for Valencia and 6·3 for Valencia Americana sweet oranges growing in orchards. Standard PCR and qPCR results from five samples each of ‘Ca. L. asiaticus’- or ‘Ca. L. americanus’-infected orange jasmine and citrus trees assessed in this study are illustrated in Figure 3. In general, the higher the band intensity on the agarose gel, the lower the cycle threshold value and the higher the bacterial titre.

Graft-inoculation experiments

Low budstick survival was observed in heterologous plant combinations (Table 4) when the grafted budsticks were unwrapped 2 months after inoculation. On orange jasmine, survival percentages were 6·7% (2/30) for budsticks from infected citrus and 60% (18/30) for those from infected orange jasmine shoots. On citrus, survival percentages were 3·3% (1/30) for budsticks from infected orange jasmine and 90% (27/30) for those from infected citrus shoots.

Table 4.   Graft-transmission experiments involving orange jasmine (Murraya exotica) and sweet orange trees infected by ‘Candidatus Liberibacter americanus’ and ‘Ca. L. asiaticus’ in homologous and heterologous host combinations
Method of inoculationInoculated hostInoculum sourceTotal inoculated treesTrees on which the inoculum source survivedTotal (and percentage) PCR-positive trees
Side-grafted budstickOrange jasmineOrange jasmine, ‘Ca. L. americanus’-infected shoot20102 (20·0%)
Orange jasmine, ‘Ca. L. asiaticus’-infected shoot1080
Citrus, ‘Ca. L. americanus’-infected shoot200
Citrus, ‘Ca. L. asiaticus’-infected shoot1020
Sweet orangeOrange jasmine, ‘Ca. L. americanus’-infected shoot2010
Orange jasmine, Ca. L. asiaticus-infected shoot100
Citrus, ‘Ca. L. americanus’-infected shoot20185 (27·8%)
Citrus, ‘Ca. L. asiaticus’-infected shoot1095 (55·6%)
Side-grafted young treeOrange jasmineCitrus, ‘Ca. L. americanus’-infected plant22100
Citrus, ‘Ca. L. asiaticus’-infected plant10100

Standard PCR was applied to samples from all inoculated plants 6 and 12 months after inoculation. In orange jasmine, liberibacter was detected in two of eight ‘Ca. L. americanus’-inoculated plants exhibiting general leaf chlorosis (Fig. 4a) 6 months post-inoculation. A faint band was also detected from samples of two symptomless plants: one grafted with a budstick from ‘Ca. L. asiaticus’-infected citrus and another grafted with a budstick from ‘Ca. L. asiaticus’-infected orange jasmine. One year after inoculation, however, only the two ‘Ca. L. americanus’-inoculated plants continued to test positive by standard PCR. These plants were transferred to larger pots, fertilized regularly and pruned for use in D. citri transmission experiments. Almost 5 years post-inoculation, most branches had dark green leaves, except for a few showing symptoms similar to those induced by manganese and zinc deficiencies in citrus trees (Fig. 4b). qPCR conducted on DNA extracted from these plants resulted in average Ct values and bacterial titres of 29·5 (±0·39 SE) and 5·5, respectively for the first, and 31·0 (±0·19 SE) and 5·0, respectively for the second plant, 56 months post-inoculation.

image

Figure 4.  Orange jasmine (Murraya exotica) plants graft-inoculated with budsticks removed from shoots (showing huanglongbing symptoms) of trees growing in urban areas and testing positive for ‘Candidatus Liberibacter americanus’, photographed in 2005 (a) and in 2009 (b). (a) On the left is a symptomless, PCR-negative plant and on the right, a PCR-positive plant with symptoms (general leaf chlorosis), 12 months after inoculation. (b) The same PCR-positive tree 4 years later, exhibiting mineral deficiency-like symptoms at the top of a few shoots.

Download figure to PowerPoint

In sweet orange, liberibacters were detected in 27·8% (5/18) of plants inoculated with ‘Ca. L. americanus’-infected budsticks and in 55·6% (5/9) of plants inoculated with ‘Ca. L. asiaticus’-infected budsticks, 12 months after inoculation. The PCR-negative plants exhibited normal growth and green leaves. The PCR-positive plants were smaller and exhibited characteristic HLB symptoms (Lopes & Frare, 2008), including blotchy mottled and curved leaves plus symptoms similar to those induced by mineral deficiencies.

Side grafting of the stems of young trees growing in their own pots was an attempt to increase survival rates of the sources of inoculum. Percentage survival increased to 45·5% (10/22) for ‘Ca. L. americanus’-, and to 100% for ‘Ca. L. asiaticus’-infected citrus trees (Table 4). The orange jasmine and citrus stems subsequently remained attached for more than 2 years, during which time only the citrus trees exhibited leaf symptoms and tested positive by standard and qPCR conducted at approximately 6-month intervals.

Sequencing and analysis of the rplJ gene

In order to determine potential variability within the orange jasmine- and citrus-associated liberibacters, the rplJ gene was sequenced and analysed. Sequence alignments of 548 bp of ‘Ca. L. americanus’ or of 495 bp of ‘Ca. L. asiaticus’ revealed 100% nucleotide identity among strains within each species, regardless of host origin, and between the consensus sequences obtained for each species and sequences available in GenBank (accession EF122254.1 of ‘Ca. L. americanus’ and accession AB490691.1 of ‘Ca. L. asiaticus’).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The presence of ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ in orange jasmine and citrus trees in urban areas of the main citrus-growing region of São Paulo was demonstrated. DNA sequence comparisons revealed no variation in the rplJ gene within each species, suggesting that the orange jasmine- and citrus-associated liberibacters are probably the same, and that they have been repeatedly transmitted from one to the other host by D. citri.

Although both citrus and orange jasmine were found to host liberibacters in the cities, their responses to infection seemed to differ considerably. A few trees found infected in 2005/2006 were observed for subsequent symptom development in the city of Araraquara. The damage to orange jasmine was much less severe than that on citrus. Although the extent of yellow shoots on ‘Ca. L. americanus’- or ‘Ca. L. asiaticus’-infected trees generally increased over time, the rate of symptom progress was much lower on orange jasmine than on citrus (results not shown). By 2009, the backyard citrus trees found infected during the 2005/2006 survey had already been removed or were showing extensive shoot dieback on most of the canopy.

The apparently less severe effects of ‘Ca. L. americanus’ infections in naturally- infected adult orange jasmine trees than in citrus were confirmed experimentally in graft-inoculated young plants. The mild symptoms observed 6 months after inoculation almost disappeared after transferring the infected plants to larger pots and applying fertilizers frequently. Almost 5 years post-inoculation, the symptoms in those plants were restricted to just the tops of a few branches. This less severe symptom progress might be related to a lower capability of liberibacters to multiply in this host than in citrus. In naturally infected trees in the cities, average titres found for ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ were log 1·9 and log 4·5 times lower, respectively, in orange jasmine than in citrus.

The results of this work also suggest that ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ interact with orange jasmine as opportunistic pathogens. The failure to find any symptom that could be specifically associated with liberibacter infection on the city-grown, naturally infected trees or on the pot-grown, graft-inoculated plants was further evidence of this. Also, most sampled trees with symptoms were PCR-negative for the pathogens despite the overall similarity of symptoms expressed by PCR-positive trees.

The high mortality of the infected citrus budwood on stems of grafted orange jasmine plants and vice versa was reduced with the young-plant, side-grafting method. However, even with apparent fusion of bark tissues between the plants, no increase in liberibacter transmission was observed for ‘Ca. L. americanus’ and no transmission was achieved with ‘Ca. L. asiaticus’, possibly because of tissue incompatibility, as observed in other studies (Hung et al., 2000), hampering effective connections between the vascular system of the plants and, consequently, movement of bacteria from one host to the other.

In the graft-inoculation experiments with homologous orange jasmine combinations, only ‘Ca. L. americanus’ was transmitted successfully, despite the higher percentage of the ‘Ca. L. asiaticus’-infected budsticks that survived. In contrast, in homologous citrus combinations, higher transmission efficiencies were observed for ‘Ca. L. asiaticus’. These differences, also observed in citrus in previous work (Lopes et al., 2009a), may have been the result of the high bacterial titres in naturally infected orange jasmine than citrus trees for ‘Ca. L. americanus’, and vice versa for ‘Ca. L. asiaticus’, or irregular distribution of liberibacters in the infected shoots used as sources of inoculum.

The combined data from both surveys showed a disproportionate increase in ‘Ca. L. asiaticus’ compared with ‘Ca. L. americanus’ over time in the cities, a pattern similar to that observed in commercial citrus orchards in São Paulo state (Lopes et al., 2009a). This may have been related to higher titres of ‘Ca. L. asiaticus’ in citrus trees, and to higher incidence of trees infected by this species in orchards and backyards, favouring its acquisition by D. citri from infected citrus and subsequent transmission to healthy citrus and orange jasmine trees.

Conflicting information exists in the literature from other countries as to whether orange jasmine is an alternative host of ‘Ca. L. asiaticus’. Variation in liberibacter populations may lead to variation in host ranges (Halbert & Manjunah, 2004), as observed in Taiwan by Tsai et al. (2008). Also, variation between forms of orange jasmine, as observed in China and India (But et al., 1986; Kong et al., 1986, 1987; Li et al., 1988; Huang, 1997; Ranade et al., 2006; Verma et al., 2009), may lead to variation in susceptibility to liberibacter infection.

Although the results indicate that orange jasmine, in this instance the widely cultivated ornamental form (M. exotica), is less conducive to liberibacter multiplication than citrus, its importance in the HLB epidemics should not be underestimated. It is a preferred feeding and rearing host of D. citri (Halbert & Manjunah, 2004) and its presence in regions where HLB and D. citri occur increases the probability of transmission of the HLB-associated liberibacters from infected to healthy hosts. Indeed, ‘Ca. L. americanus’ transmission from orange jasmine to citrus was experimentally demonstrated recently (Gasparoto et al., 2010). Also, the orange jasmine trees in the cities are not under an insect-control programme and the suggested tree-eradication/-substitution programme is in effect only in a few cities. Therefore, the orange jasmine trees in Brazilian cities may have been continually serving as sources of liberibacters and D. citri.

The surveyed area of orange jasmine included 76 urban locations distributed in the main citrus-growing regions of the state of São Paulo. Liberibacter-affected urban localities were relatively close to each other and coincident with an area of high incidence of liberibacters in commercial orchards (Anonymous, 2009), strongly suggesting that both ‘Ca. L. asiaticus’ and ‘Ca. L. americanus’ are being transmitted from one to the other host by D. citri and indicating the importance of implementing regulations to reduce this risk. After approximately 5 years of attempts to combat HLB spread in commercial orchards, it is clear that, to succeed, any strategy must address the reduction of inoculum pressure on a regional scale. Removal of infected orange jasmine and citrus trees in urban areas should be part of any HLB control programme.

In Brazil, orange jasmine is usually sold in flower retail centres and in native and ornamental tree nurseries, where it is propagated from seed in unscreened environments. Usually, the seed sources are trees growing in urban areas. Although PCR and qPCR assessments of hundreds of young seedlings originating from seeds of trees with symptoms tested negative for both ‘Ca. L. americanus’ and ‘Ca. L. asiaticus’ (S.A. Lopes, unpublished results), given the fact that orange jasmine is a preferred host and attractive to D. citri (Halbert & Manjunah, 2004; Wenninger et al., 2009), its multiplication in unprotected areas and the lack of regulation in its production and commercialization favour the wide dissemination of these bacteria. This is probably how ‘Ca. L. asiaticus’ was quickly dispersed from south to north in Florida (Manjunath et al., 2008) and should alert the authorities to initiate action plans to prevent HLB spread to areas still free from this destructive citrus disease.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors thank Anelise G. Mariano and Elaine C. Martins for technical assistance in DNA cloning and qPCR analysis, and FAPESP (grants 2005/00718-2 and 2007/55013-9) for financial resources.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Anonymous, 2009. Incidência de greening aumenta 30% no estado de São Paulo. Revista do Fundecitrus 151, 89.
  • Bové JM, 2006. Huanglongbing: a destructive, newly-emerging, century-old disease of citrus. Journal of Plant Pathology 88, 737.
  • But PPH, Kong YC, Ng KH et al. , 1986. A chemotaxonomic study of Murraya (Rutaceae) in China. Acta Phytotaxonomica Sinica 24, 18692.
  • Capoor SP, Rao DG, Viswanath SM, 1967. Diaphorina citri Kuway., a vector of greening disease of citrus in India. Indian Journal of Agricultural Science 37, 5726.
  • Coletta-Filho HD, Targon MLPN, Takita MA et al. , 2004. First report of the causal agent of huanglongbing (‘Candidatus Liberibacter asiaticus’) in Brazil. Plant Disease 88, 1382.
  • Dai K, Ikeshiro T, Matsuura T et al. , 2005. Investigation of host range of Candidatus Liberobacter asiaticum – is Murraya paniculata a host plant of Candidatus L. asiaticum? Research Bulletin of the Plant Protection Service Japan 41, 537.
  • Damsteegt V, Schneider W, Stone A, 2007. Murraya paniculata (orange jasmine), a host and possible inoculum reservoir for Candidatus Liberibacter asiaticus, causal agent of Huanglongbing. Phytopathology 97, S26.
  • Gasparoto MCG, Bassanezi RB, Amorim L et al. , 2010. First report of ‘Candidatus Liberibacter americanus’ transmission from Murraya paniculata to sweet orange by Diaphorina citri. Journal of Plant Pathology 92, 546.
  • Da Graça JV, 1991. Citrus greening disease. Annual Review of Phytopathology 29, 10936.
  • Halbert SE, Manjunah KL, 2004. Asian citrus psyllids (Sternorrhyncha: Psyllidae) and greening disease of citrus: a literature review and assessment of risk in Florida. Florida Entomologist 87, 33053.
  • Hocquellet A, Torawa P, Bové JM, Garnier M, 1999. Detection and identification of the two Candidatus Liberobacter species associated with citrus huanglongbing by PCR amplification of ribosomal protein genes of the β operon. Molecular and Cellular Probes 13, 3739.
  • Huang CC, 1959. Preliminary study on Chinese Rutaceae. Acta Phytotaxonomica Sinica 8, 69124.
  • Huang B, 1990. Citrus psyllid. In: WuFZ, TanJJ, LiCY, ChenCM, ZhangSM, CaoCY, eds. Chinese Agricultural Encyclopedia, Insects Volume Insects. Beijing, China: Agriculture Press, 11920.
  • Huang CC, 1997. Rutaceae. Beijing, China: Science Press: Flora Reipublicae Popularis Sinicae vol. 43.
  • Hung TH, Wu ML, Su HJ, 2000. Identification of alternative hosts of the fastidious bacterium causing citrus greening disease. Journal of Phytopathology 148, 3216.
  • Husain MA, Nath D, 1927. The citrus psylla (Diaphorina citri, Kuw.) [Psyllidae: Homoptera]. Memoirs of the Department of Agriculture in India, Entomological Series 10, 127.
  • Koizumi M, Prommintara M, Ohtsu Y, 1996. Wood apple, Limonia acidissima: a new host for the huanglongbing (greening) vector, Diaphorina citri. In: Da GraçaJV, MorenoP, YokomiRK, eds. Proceedings of the Thirteenth Conference of the International Organization of Citrus Virologists, 1996. Riverside, CA, USA: IOCV, 2715.
  • Kong YC, Cheng KF, Ng KH et al. , 1986. A chemotaxonomic division of Murraya based on the distribution of the alkaloids yuehchukene and girinimbine. Biochemical Systematics and Ecology 14, 4917.
  • Kong YC, Ng KH, But PPH, Cheng KF, Waterman PG, 1987. Aurantiamide acetate in the stem bark of Murraya exotica. Planta Medica 53, 393.
  • Li T, Ke C, 2002. Detection of the bearing rate of Liberobacter asiaticum in citrus psylla and its host plant Murraya paniculata by nested PCR. Acta Phytophylacica Sinica 29, 315.
  • Li Q, Zhu LF, But PPH, Kong YC, Chang HT, Waterman PG, 1988. Monoterpene and sesquiterpene rich oils from the leaves of Murraya species: chemotaxonomic significance. Biochemical Systematics and Ecology 16, 4914.
  • Lopes SA, Frare GF, 2008. Graft transmission and cultivar reaction of citrus to ‘Candidatus Liberibacter americanus’. Plant Disease 92, 214.
  • Lopes SA, Martins EC, Frare GF, 2005. Detecção de Candidatus Liberibacter americanus em Murraya paniculata. Summa Phytopathologyca 31, 489.
  • Lopes SA, Martins EC, Frare GF, 2006a. Detecção de Candidatus Liberibacter asiaticus em Murraya paniculata. Fitopatologia Brasileira 31(Suppl.), 303.
  • Lopes SA, Martins EC, Frare GF, 2006b. Plantas hospedeiras de liberibacter no Brasil. In: Proceedings of Huanglongbing – Greening International Workshop, 2006. Ribeirão Preto, Brazil: Fundecitrus, 27.
  • Lopes SA, Frare GF, Wulff NA, Fernandes NG, 2007. Epidemia de huanglongbing (greening) no Estado de São Paulo: provável influência da temperatura e das diferenças em transmissibilidade dos patógenos na evolução da doença. In: Proceedings do Workshop sobre Epidemiologia de Doenças de Plantas, Campos do Jordão, Brazil. 6976.
  • Lopes SA, Bertolini E, Frare GF et al. , 2009a. Graft transmission efficiencies and multiplication of ‘Candidatus Liberibacter americanus’ and ‘Ca. Liberibacter asiaticus’ in citrus plants. Phytopathology 99, 3016.
  • Lopes SA, Frare GF, Bertolini E et al. , 2009b. Liberibacters associated with citrus huanglongbing in Brazil: ‘Candidatus Liberibacter asiaticus’ is heat tolerant, ‘Ca. L. americanus’ is heat sensitive. Plant Disease 93, 25762.
  • Manjunath KL, Halbert SE, Ramadugu C, Webb S, Lee RF, 2008. Detection of ‘Candidatus Liberibacter asiaticus’ in Diaphorina citri and its importance in the management of citrus huanglongbing in Florida. Phytopathology 98, 38796.
  • Murray MG, Thompson WF, 1980. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Research 8, 43215.
  • Pande YD, 1971. Biology of citrus psylla, Diaphorina citri Kuw. (Hemiptera: Psyllidae). Israel Journal of Entomology 6, 30711.
  • Ranade SA, Rana TS, Srivastava AP, Nair KN, 2006. Molecular differentiation in Murraya Koenig ex L. species in India inferred through ITS, RAPD and DAMD analysis. Current Science India 90, 12538.
  • Teixeira DC, Danet JL, Eveillard S et al. , 2005a. Citrus huanglongbing in São Paulo State, Brazil: PCR detection of the ‘Candidatus’ Liberibacter species associated with the disease. Molecular and Cellular Probes 19, 1739.
  • Teixeira DC, Saillard C, Eveillard S, Danet JL, Ayres AJ, Bové JM, 2005b. Candidatus Liberibacter americanus’ associated with citrus huanglongbing (greening disease) in São Paulo State, Brazil. International Journal of Systematic and Evolutionary Microbiology 55, 185762.
  • Teixeira DC, Eveillard S, Sirand-Pugnet P et al. , 2008a. The tufB–secE–nusG–rplKAJL–rpoB gene cluster of the liberibacters: sequence comparisons, phylogeny and speciation. International Journal of Systematic and Evolutionary Microbiology 58, 141421.
  • Teixeira DC, Saillard C, Couture C et al. , 2008b. Distribution and quantification of Candidatus Liberibacter americanus, agent of huanglongbing disease of citrus in São Paulo State, Brazil, in leaves of an affected sweet orange tree as determined by PCR. Molecular and Cellular Probes 22, 13950.
  • Teixeira DC, Wulff NA, Leite APR, Martins EC, Ayres AJ, Bové JM, 2009. Identification, PCR detection and occurrence in São Paulo state, Brasil, of citrus huanglongbing-associated agents: Candidatus Liberibacter americanus, Ca. L. asiaticus, and the 16Sr group IX phytoplasma. In: XLII Congresso Brasileiro de Fitopato-logia, 2009, Rio de Janeiro. Tropical Plant Pathology 34, S7.
  • Tirtawidjaja S, 1981. Insect, dodder and seed transmissions of citrus vein phloem degeneration (CVPD). In: Proceedings of the Fourth International Society of Citriculture Congress. Tokyo, Japan, 1981. International Society of Citriculture, 46971.
  • Tsai CH, Hung TH, Su HJ, 2008. Strain identification and distribution of citrus Huanglongbing bacteria in Taiwan. Botanical Studies 49, 4956.
  • Verma S, Rana TS, Ranade SA, 2009. Genetic variation and clustering in Murraya paniculata complex as revealed by single primer amplification reaction methods. Current Science 96, 12106.
  • Wenninger EJ, Stelinski LL, Hall DG, 2009. Role of olfactory cues, visual cues, and mating status in orientation of Diaphorina citri Kuwayama (Hemiptera: Psyllidae) to four different host plants. Environmental Entomology 38, 22534.
  • Wulff NA, Eveillard S, Foissac X, Ayres AJ, Bové JM, 2009. Ribosomal RNA operons and genome size of ‘Candidatus Liberibacter americanus’, a bacterium associated with citrus huanglongbing in Brazil. International Journal of Systematic and Evolutionary Microbiology 59, 198491.
  • Yamamoto PT, Felippe MR, Garbim LF et al. , 2006. Diaphorina citri (Kuwayama) (Hemiptera: Psyllydae): vector of the bacterium Candidatus Liberibacter americanus. In: Proceedings of Huanglongbing – Greening International Workshop, 2006. Ribeirão Preto, Brazil: Fundecitrus, 256.
  • Zhou LJ, Gabriel DW, Duan YP, Halbert SE, Dixon WN, 2007. First report of dodder transmission of huanglongbing from naturally infected Murraya paniculata to citrus. Plant Disease 91, 227.