Citrus huanglongbing in São Paulo State, Brazil: PCR detection of the ‘Candidatus’ Liberibacter species associated with the disease

Citrus huanglongbing in Sao Paulo State, Brazil: PCR detection of the

‘Candidatus’ Liberibacter species associated with the disease

Diva do Carmo Teixeiraa, Jean Luc Danetb, Sandrine Eveillardb, Elaine Cristina Martinsa,Waldir Cintra de Jesus Juniora, Pedro Takao Yamamotoa, Silvio Aparecido Lopesa,

Renato Beozzo Bassanezia, Antonio Juliano Ayresa, Colette Saillardb, Joseph Marie Boveb,*

aFundecitrus, Av. Dr. Adhemar Pereira de Barros, 201, CEP 14807-040 Araraquara, SP, BrazilbIBVM, Centre INRA, Institut National de la Recherche Agronomique and Universite de Bordeaux 2, Laboratoire de Biologie cellulaire et moleculaire,

71, Avenue Edouard Bordeaux, B.P. 81, 33883 Villenave d’Ornon Cedex, France

Received 16 September 2004; accepted for publication 12 November 2004

Abstract

Symptoms of huanglongbing (HLB), one of the most serious diseases of citrus in Asia and Africa, have been noticed in March 2004 in the

Araraquara region of Sao Paulo State, Brazil. HLB has not been reported previously from America. The causal HLB bacteria, Candidatus

Liberibacter africanus in Africa and Candidatus Liberibacter asiaticus in Asia, can be detected in symptomatic citrus leaves by PCR

amplification of their 16S rDNA with previously described primers. When this technique was applied to 43 symptomatic leaf samples from

the Araraquara region, all PCR reactions were negative. This suggested that a new pathogen, not detected by the above primers, could be

involved in HLB in the State of Sao Paulo. Indeed, by using universal primers for amplification of bacterial 16S rDNA, a new liberibacter

species, Candidatus Liberibacter americanus, has recently been identified. Specific primers for PCR amplification of the 16S rDNA of Ca. L.

americanus have been selected. Using these primers, the new liberibacter could be detected in 214 symptomatic leaf samples tested. The

leaves of two additional samples were infected with Candidatus Liberibacter asiaticus, and two further samples contained both Ca. L.

americanus and Ca. L. asiaticus. The samples came from 47 farms in 35 municipalities. The psyllid vector of Ca. L. asiaticus, Diaphorina

citri, is established in South, Central, and North America (Florida and Texas). Ca. L. americanus could be detected by PCR in several batches

of D. citri psyllids collected on symptomatic sweet orange trees infected with Ca. L. americanus, strongly suggesting that D. citri is the vector

of Ca. L. americanus. The results reported here confirm the presence of HLB in the State of Sao Paulo. Ca. L. americanus is the most widely

distributed pathogen.

q 2004 Elsevier Ltd. All rights reserved.

Keywords: Citrus; Huanglongbing; Greening; Liberibacter; 16S rDNA; Brazil

1. Introduction

Huanglongbing (HLB), previously called greening, is

one of the most serious diseases of citrus. The causal agent

is a non-cultured, sieve tube-restricted member of the a-

subdivision of the Proteobacteria: Candidatus Liberibacter

africanus in Africa and Candidatus Liberibacter asiaticus

in Asia [1]. HLB has not been reported previously from

America. However, the Asian psyllid vector of Ca. L.

0890-8508/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.mcp.2004.11.002

* Corresponding author. Tel.: C33 670 774 883; fax: C33 557 122 369.

E-mail address: [email protected] (J.M. Bove).

asiaticus, Diaphorina citri, is established in South, Central,

and North America (Florida and Texas). The insect reached

Brazil 60 years ago, but entered Texas only in 2001. In

March 2004, leaf and fruit symptoms resembling those of

HLB were observed in several sweet orange (Citrus sinensis

(L.) Osbeck) orchards in the Araraquara area of Sao Paulo

State. Leaf mottling or ‘blotchy mottle’ [2], a characteristic

feature of HLB, was the major foliar symptom. Fruits were

small and lopsided, and contained many aborted seeds [3].

A PCR method has been described previously, and

permits the detection of the two liberibacters in citrus leaves

by amplification of an 1160 bp fragment of their 16S rDNA

[4]. With both liberibacter species, the size of the amplicon is

Molecular and Cellular Probes 19 (2005) 173–179

www.elsevier.com/locate/ymcpr

D. do Carmo Teixeira et al. / Molecular and Cellular Probes 19 (2005) 173–179174

the same (1160 bp). However, the Ca. L asiaticus amplicon

contains one Xba1 restriction site, and yields two fragments

(640 and 520 bp) upon restriction, while Ca. L. africanus has

two such sites, and yields three fragments (520, 506 and

130 bp). It is thus easy to identify the liberibacter species

involved [4]. The PCR method has been assayed in many

Asian and African countries for the detection of the two HLB

liberibacters [see for instance 5–7]. Whenever leaves with

the classic blotchy mottle symptoms were used, positive PCR

reactions were always obtained, and yielded the character-

istic 1160 bp amplicon. In April 2004, it was decided to use

the PCR technique to confirm the presence of HLB in Sao

Paulo State, and identify the liberibacter involved: Ca. L.

asiaticus or Ca. L. africanus. Unexpectedly, none of the

symptomatic leaf samples from 43 affected trees, of which

many had severe fruit symptoms, yielded a positive PCR test,

under conditions where symptomatic control leaves infected

with Ca. L. asiaticus or Ca. L. africanus from the HLB

collection in Bordeaux, gave positive PCR reactions.

However, at the same time, and using the same PCR

technique, Ca. L. asiaticus was detected in two of 22 leaf

samples (M.A. Machado, Cordeiropolis, S.P., Brazil,

personal communication). In view of the many negative

PCR reactions witnessed, the presence of a new bacterial

pathogen in the symptomatic, blotchy mottle leaves from Sao

Paulo State was suspected and investigated. This work has

resulted in the discovery of a third liberibacter species, for

which we have proposed the name Candidatus Liberibacter

americanus [3]. The complete characterization of the new

liberibacter is in progress. In the work reported here, primers

for PCR amplification of the 16S rDNA of Ca. L. americanus

have been used to detect the new liberibacter in citrus leaves,

and evaluate its distribution, in comparison with that of

Ca. L. asiaticus and Ca. L. africanus, within the huanglongb-

ing-affected region of Sao Paulo State. Detection of the new

liberibacter in D. citri psyllids was also attempted.

2. Materials and methods

2.1. Plant material

Two-year-old ‘Hamlin’ sweet orange seedlings infected

with Ca. L. africanus or Ca. L. asiaticus were from the HLB

collection in Bordeaux, France, and served as positive

control material. Healthy and infected seedlings were

maintained in a greenhouse as previously described [4].

Symptomless citrus leaves and leaves with characteristic

blotchy mottle symptoms were collected in April, June and

August 2004, on trees from 47 citrus farms within the 35

municipalities of Sao Paulo State where symptoms of HLB

had been detected. Each leaf sample came from a single

tree. When symptomatic leaves were sampled on trees

bearing fruit, fruit symptoms of HLB always accompanied

the leaf symptoms. Leaves with symptoms other than HLB

blotchy mottle were also sampled. Each leaf sample

contained 10–20 leaves. Leaves were kept in plastic bags

at 4 8C before they were used for DNA extraction within

48 h [8]. In total, 271 leaf samples were collected. Most

symptomatic leaf samples came from sweet orange trees,

the major citrus cultivar grown in Sao Paulo State, but some

were also collected on affected Ponkan mandarin, Cravo

mandarin, and Murcott tangor trees.

2.2. Capture of D. citri psyllids

D. citri psyllids were captured with a motor-driven

aspirator on three fully infected, symptomatic ‘Pera’ sweet

orange trees, after PCR assays had shown the trees to be

infected with Ca. L. americanus only.

2.3. PCR detection of Ca. L. americanus

PCR primers specific for amplification of the 16S rDNA

of the new liberibacter were selected from the 16S rDNA

sequence (GenBank accession number: AY742824). Fig. 1

shows the sequence of forward primer (GB1) and reverse

primer (GB3) in comparison with corresponding sequences

of Ca. L. asiaticus and Ca. L. africanus 16S rDNAs. The use

of these primers leads to a 1027 bp amplicon. The PCR

reaction was performed in 40 ml of reaction mixture

containing 1 mM of each of the primers, 200 mM of each

of the four dNTP, 2 mM MgCl2, 20 mM Tris–HCl pH 8.4,

50 mM KCl, 1.5 U of Taq polymerase (Promega), and 1 ml

of DNA preparation. A mastercycle gradient thermocycler

(Eppendorf) with the following program was used for DNA

amplification: 35 cycles each at 94 8C for 45 s, 64 8C for

45 s, and 72 8C for 60 s. Following amplification, 10 ml

aliquots of each reaction mixture were analysed by

electrophoresis on 1.2% agarose gels.

2.4. PCR detection of Ca. L. africanus and Ca. L. asiaticus

PCR was performed according to [4] with 35 cycles each

at 92 8C for 40 s, and 72 8C for 90 s (annealing and primer

extension at same temperature). The reaction mixture was

the same than that for Ca. L. americanus. As shown on

Fig. 1, the sequence of reverse primer OI2c is the same for

both liberibacters. The sequences of forward primer OA1 for

Ca. L. africanus and OI1 for Ca. L. asiaticus are identical

except that GCA in OI1 is replaced by TTT in OA1. Both

forward primers were used in the reaction mixture to favor

amplification of either one of the two liberibacters [4].

Following amplification, the reaction mixture was analysed

on 1.2% agarose gels. The amplified DNA has a size of

1160 bp for both liberibacters. However, the 1160 bp

amplicon from Ca. L. asiaticus has one Xba1 restriction

site and yields two fragments upon digestion (640 and

520 bp), while Ca. L. africanus has two such sites, and gives

three fragments (520, 506, and 130 bp) [4]. The Xba1 test

was used to identify the liberibacter involved.

Fig. 1. Sequences of PCR primers for amplification of liberibacter 16S rDNA. GB1 and GB3: respectively, forward and reverse primers for Ca. L. americanus

16S rDNA amplification. OI1 and OI2c: respectively, forward and reverse primers for Ca. L. asiaticus 16S rDNA amplification. OA1 and OI2c: respectively,

forward and reverse primers for Ca. L. africanus 16S rDNA amplification. The ! indicates a mismatche between primers for Ca. L. americanus, and primers for

Ca. L. asiaticus and Ca. L. africanus.

D. do Carmo Teixeira et al. / Molecular and Cellular Probes 19 (2005) 173–179 175

3. Results

3.1. Attempts to detect Ca. L. africanus and Ca. L. asiaticus

by PCR in symptomatic leaf samples collected

in April and June, 2004

In April and June, 2004, soon after HLB had been

reported in Sao Paulo State, at a time when Ca. L.

americanus had not yet been discovered, 43 symptomatic

sweet orange leaf samples were collected in seven citrus

farms in the Araraquara region (Fig. 2). All samples gave

negative PCR reactions with the primers specific of Ca. L.

africanus and Ca. L.asiaticus 16S rDNA (Fig. 3) under

conditions where symptomatic control leaves infected with

Fig. 2. Map of Sao Paulo State showing the citrus area in green, and, in red, the 35 m

found. Position of Sao Paulo city and Araraquara city is indicated by a black dot

Ca. L. asiaticus (Fig. 3, AS) or Ca. L. africanus (Fig. 3, AF)

gave positive reactions. Fig. 3 represents the results from

only 31 of the 43 samples. The results from the other

samples were similarly negative (data not shown).

3.2. PCR detection of Ca. L. americanus and Ca. L.

asiaticus in symptomatic leaves

By July 2004, specific primers for PCR amplification of

Ca. L. americanus 16S rDNA (Fig. 1, GB1/GB3) became

available [3]. These primers, as well as the primers specific

for the PCR detection of Ca. L. africanus and Ca. L.

asiaticus (Fig. 1, OI1C OA1/OI2c) were used for the

detection of the three liberibacters in each leaf sample

unicipaloties, where citrus trees infected with HLB liberibacters have been

and a white dot, respectively.

Fig. 3. Agarose gel electrophoresis of symptomatic leaf DNA amplified with 16S rDNA primers (OI1C OA1)/OI2c, specific for Ca. L. asiaticus and Ca. L.

africanus. 1–31: DNA from symptomatic leaf samples 1–31, collected in April and June, 2004. AS and AF: DNA from symptomatic Hamlin sweet orange

leaves, respectively, infected with Ca. L. asiaticus and Ca. L. africanus. H: DNA from healthy sweet orange leaves. Leaves for AS, AF, and H were from the

Bordeaux greenhouse. O: amplification in the absence of DNA. M; DNA size markers.

D. do Carmo Teixeira et al. / Molecular and Cellular Probes 19 (2005) 173–179176

collected. One aliquot of the DNA from a leaf sample was

used for the detection of Ca. L. americanus, and a second

aliquot served for the detection of Ca. L. africanus and/or

Ca. L. asiaticus in the same leaf sample.

3.2.1. Leaf samples collected in April and June, 2004

The DNAs of these samples had been saved, and were

used again in August 2004, when the Ca. L. americanus

primers GB1/GB3 had become available. As before, all 43

samples tested negative for Ca. L. africanus and Ca. L.

asiaticus, but they were all positive when assayed for Ca. L.

americanus. Fig. 4 shows the results obtained, respectively

with samples 1–5, and 11–18. The other April–June samples

(samples 6–10, and 19–43) gave identical results (data not

shown). As illustrated on Fig. 4, the PCR reactions were

negative with healthy leaves (H), and in the absence of DNA

(O). Primers GB1/GB3 (Am on Fig. 4) were specific for

Fig. 4. Agarose gel electrophoresis of symptomatic leaf DNA amplified with 16S

africanus, and 16S rDNA primers GB1/GB3, specific for Ca. L. americanus. Sympt

As: primers (OI1C OA1)/OI2c. Am: primers GB1/GB3. (a) DNA aliquots amplifie

Fig. 3. AM: DNA from symptomatic sweet orange leaves infected with Ca. L. am

Ca. L. americanus, as no amplification was obtained with

DNA from control leaves infected with Ca. L. africanus (AF

on Fig. 4) or Ca. L. asiaticus (AS on Fig. 4). Inversely,

primers (OA1C OI1)/OI2c (As on Fig. 4), specific for Ca.

L. africanus and Ca. L. asiaticus, gave no amplification with

Ca. L. americanus-infected citrus leaves (Fig. 4, lanes ‘b’).

Finally, as seen on the gels of Fig. 4–6 and 8, the 16S rDNA

amplicon from Ca. L. americanus, with a size of 1027 bp,

was easily distinguishable from the 1160 bp amplicon

characteristic of the other two liberibacters.

3.2.2. Leaf samples collected in August, 2004

Additional symptomatic citrus leaf samples were col-

lected in August 2004. Among these samples, 171 gave

the same results than the previous 43 samples: the PCR

reactions with the primers for Ca. L. americanus were

positive, but negative with the primers for Ca. L. africanus

rDNA primers (OI1C OA1)/OI2c, specific for Ca. L. asiaticus and Ca. L.

omatic citrus leaf samples 1–5 and 11–18 were collected in April/June 2004.

d with Am. (b) DNA aliquots amplified with As. AS, AF, H, M, and O: as in

ericanus.

Fig. 5. Agarose gel electrophoresis of symptomatic leaf DNA amplified with 16S rDNA primers (OI1C OA1)/OI2c, specific for Ca. L. asiaticus and Ca. L.

africanus, and 16S rDNA primers GB1/GB3, specific for for Ca. L. americanus. Symptomatic citrus leaf samples 50, 64–67, and 69–71 were collected in July,

2004. DNA aliquots a and b, AM, Am and As: as in Fig. 4. AS, M, and O: as in Fig. 3.

D. do Carmo Teixeira et al. / Molecular and Cellular Probes 19 (2005) 173–179 177

and Ca. L. asiaticus. Fig. 5 shows the results for some of these

samples. Eventually however, four samples gave a positive

PCR reaction with the primers for Ca. L. africanus and Ca. L.

asiaticus, and the Xba1 test identified the liberibacter as Ca.

L. asiaticus. Two of these samples, sample 51, from a

Murcott tangor orchard, and sample 121, from a ‘Hamlin’

sweet orange orchard, were infected with Ca. L. asiaticus

only, but sample 34, from a backyard ‘Lima’ sweet orange

tree, and sample 322, from a Pera sweet orange tree, were

positive for both Ca. L. asiaticus and Ca. L. americanus. The

results from samples 34 and 51 are illustrated on Fig. 6. The

results from samples 34, 51 and 121, were confirmed a first

time with additional leaves of these samples, which had been

left over and kept at 4 8C, and a second time with a new batch

of leaves collected on the same trees.

In total, Ca. L. americanus was detected in 216 citrus leaf

samples, of which five were from Ponkan mandarin trees, one

from a Murcott tangor tree, two from Cravo mandarin trees,

and 208 from sweet orange trees. All sweet orange varieties,

including ‘Chamout’, ‘Hamlin’, ‘Lima’, ‘Natal’, ‘Pera’,

‘Valencia’ and ‘Westin’, were found to be infected. The 216

samples were from 47 farms in 35 municipalities (Fig. 2).

Only four samples were found to be infected with

Ca. L. asiaticus, of which two were also infected with

Ca. L. americanus. The four farms in which Ca. L. asiaticus

Fig. 6. Agarose gel electrophoresis of symptomatic leaf DNA amplified with 16S

africanus, and 16S rDNA primers GB1/GB3, specific for for Ca. L. americanus. S

from samples 34 and 51, amplified with Am. 34b and 51b: DNA aliquots from sam

and O: as in Fig. 3.

was detected, had also trees infected with Ca. L. americanus.

No evidence was obtained for the presence of Ca. L.

africanus.

3.3. Search for Ca. L. americanus, Ca. L. africanus and Ca.

L. asiaticus in symptomless leaves by PCR

Samples of symptomless leaves were collected on three

types of trees: (i) on symptomatic trees, opposite the

affected sector, (ii) on symptomless trees adjacent to

symptomatic trees, and (iii) on symptomless trees from a

region not affected by HLB. As expected from previous

experience, all symptomless leaves gave negative PCR

reactions (data not shown).

3.4. Search for Ca. L. americanus, Ca. L. africanus and

Ca. L. asiaticus in D. citri psyllids by PCR

D. citri psyllids were collected on three Pera sweet

orange trees with severe symptoms of HLB and shown by

PCR to be infected with Ca. L. americanus only.

In the experiment of Fig. 7, 22 batches of psyllids (10

insects per batch) were tested by PCR for the presence of

liberibacters. The 22 batches gave negative PCR reactions

with the primers specific for Ca. L. africanus and

rDNA primers (OI1C OA1)/OI2c, specific for Ca. L. asiaticus and Ca. L.

amples 34 and 51 were collected in July 2004. 34a and 51a: DNA aliquots

ples 34 and 51, amplified with As. AM, Am and As: as in Fig. 4. AS, H, M,

Fig. 7. Agarose gel electrophoresis of Diaphorina citri psyllid DNA amplified with 16S rDNA primers (OI1C OA1)/OI2c, specific for Ca. L. asiaticus and Ca.

L. africanus (A), and 16S rDNA primers GB1/GB3, specific for for Ca. L. americanus (B). 1–22: amplified DNA from 22 psyllid batches (10 psyllids per

batch). Arrows indicate lanes with a DNA band amplified.from psyllid DNA. AF, AS, H, M, and O: as in Fig. 3. AM: as in Fig. 4.

D. do Carmo Teixeira et al. / Molecular and Cellular Probes 19 (2005) 173–179178

Ca. L. asiaticus (Fig. 7A), as expected, since the insects were

collected on trees that were not infected with these

liberibacters. However, with primers specific for Ca. L.

americanus, several batches of insects gave positve PCR

signals (Fig. 7B, batches 3, 5, 7, 10, 17 and 18). Fig. 8 shows

that the amplicons from batches 3, 5, 7, 10 and 18 (lanes 3–7)

have the same size and electrophoretic mobility than the

1027 bp amplicon characteristic of Ca. L. americanus (Fig. 8,

lanes 2 and 8), and confirms that at least one psyllid in each

positive batch was infected with Ca. L. americanus.

Fig. 8. Electrophoretic characterization of the DNA amplified from psyllid

batches 3, 5, 7, 10, and 18 (see Fig. 7) with 16S rDNA primers GB1/GB3,

specific for Ca. L. americanus. M1 and M2: DNA size markers. Lanes 1 and

9: amplified 16S rDNA of Ca. L. asiaticus. Lanes 2 and 8: amplified 16S

rDNA of Ca. L. americanus. Lanes 3–7; DNA amplified with primers

GB1/GB3 from psyllid DNA of batches 3, 5, 7, 10, and 18, respectively.

4. Discussion and conclusion

With primers GB1 and GB3, it has been possible to

specifically amplify the 16S rDNA of Ca. L. americanus

from symptomatic citrus leaves, and the amplicon obtained

had the expected size of 1027 bp. With the same primers, no

amplification was obtained with DNA from Ca. L. africanus

or Ca. L. asiaticus. As shown on Fig. 1, the sequences of

GB1 and GB3 were chosen in such a way that annealing

with 16S rDNA of Ca. L. africanus and Ca. L. asiaticus

cannot occur. In particular, GB1 has three mismatches,

indicated by ‘!’ on Fig. 1, and a one-base insertion (G),

while GB3 has a deletion of 17 bases. Inversely, the primers

specific of Ca. L. africanus and Ca. L. asiaticus, (OA1COI1)/OI2c, do not permit amplification of Ca. L. americanus

16S rDNA, and this fact explains why so many symptomatic

leaf samples, known today to be infected with Ca. L.

americanus, gave negative PCR reactions when tested with

primers (OA1C OI1)/OI2c. When these ‘negative’ samples

were later analysed with primers GB1/GB3, they all turned

positive.

Before Ca. L. americanus was discovered, the first

liberibacter detected in Sao Paulo State was Ca. L. asiaticus

(M. A. Machado, personal communication).

Since primers GB1/GB3 and (OA1C OI1)/OI2c are

specific of their respective liberibacters, it has been possible

to analyse the same symptomatic leaf sample for the

presence of Ca. L. americanus by a PCR reaction with

GB1/GB3, as well as of Ca. L. africanus and/or Ca. L.

asiaticus by a PCR reaction with (OA1C OI1)/OI2c. In this

way, 218 symptomatic leaf samples could be analysed. Ca.

L. americanus was detected in 214, Ca. L. asiaticus in 2, Ca.

L. africanus in none, and two samples were infected with

both Ca. L. americanus and Ca. L. asiaticus. The proportion

D. do Carmo Teixeira et al. / Molecular and Cellular Probes 19 (2005) 173–179 179

of Ca. L. asiaticus to Ca. L. americanus is thus 4 to 216, or

1.9%. Another way to get a rough estimate of this

percentage is as follows. Ca. L. americanus has been

detected in as many as 35 municipalities, and Ca. L.

asiaticus in only 3. In these three municipalities, the number

of samples in which, respectively, Ca. L. asiaticus and Ca.

L. americanus were detected, amounts to 4 and 99, thus

giving a proportion of 4 to 99, or 4%. These percentages

indicate that Ca. L. americanus is probably today the most

widely distributed HLB pathogen in Sao Paulo State.

Even though, HLB was reported only in March of 2004,

the disease is probably present since many years in the

Araraquara area, where some citrus orchards with a high

proportion of affected trees have been identified recently. In

one of these orchards, according to the owner, the symptoms

of HLB, as seen now, were already present some 9 years

ago. This would explain the relatively wide spatial

extension of the disease, as observed today, only 5 months

after the first report. However, except for some orchards in

the Araraquara area, the incidence of the disease in most

farms is still low, with approximately 0.2% of affected trees.

However, surveys have to be carried out in order to estimate

the incidence of the disease more precisely.

In the frame of this work, all symptomatic leaf samples

showing blotchy mottle gave positive PCR reactions, and

Ca. L. americanus was the liberibacter detected most

frequently. However, even though blotchy mottle is

characteristic, it is not specific of HLB, and certain leaf

patterns might resemble HLB symptoms without any

liberibacters being involved. In such cases, the PCR tests

described here are most valuable for HLB identification.

Also, control of HLB will require removal of infected trees

in order to reduce inoculum pressure. In orchards affected

by a number of diseases, such as blight, leprosis, citrus

variegated chlorosis, and sudden death, the PCR detection

method will undoubtedly help identify the trees to be

removed.

D. citri, the Asian psyllid vector of Ca. L. asiaticus, is

present in Sao Paulo State since 60 years. It is very likely

that this insect is also responsible for the spread of Ca. L.

americanus. In support of this hypothesis, the presence of

the liberibacter in psyllids could be demonstrated.

References

[1] Jagoueix S, Bove JM, Garnier M. The phloem-limited bacterium of

greening disease of citrus is a member of the a subdivision of the

proteobacteria. Int J Syst Bacteriol 1994;44:397–486.

[2] McClean APD, Schwarz RE. Greening or blotchy-mottle disease of

citrus. Phytophylactica 1970;2:177–94.

[3] Texeira DC, Ayres J, Danet JL, Jagoueix-Eveillard S, Saillard C,

Bove JM. First report of a huanglongbing-like disease of citrus in Sao

Paulo State Brazil, and association of a new liberibacter species,

Candidatus Liberibacter americanus, with the disease. Plant Dis Note,

January 2005.

[4] Jagoueix S, Bove JM, Garnier M. PCR detection of the two

‘Candidatus’ liberobacter species associated with greening disease of

citrus. Mol Cell Probes 1996;10:43–50.

[5] Garnier M, Bove JM. Distribution of the huanglongbing (greening)

liberobacter species in fifteen African and Asian countries. In: da

Graca JV, Moreno P, Yokomi RK, editors. Proceedings of 13th

international conference on organization citrus virologists. Riverside,

California: Department of Plant Pathology; 1996. p. 388–91.

[6] Garnier M, Jagoueix S, Toorawa P, Grisoni M, Mallessard R,

Dookun A, et al. Both huanglongbing (greening) liberobacter species

are present in Mauritius and Reunion. In: da Graca JV, Moreno P,

Yokomi RK et al, editors. Proceedings of 13th International

Conference on organization citrus virologists. Riverside, California:

Department of Plant Pathology; 1996. p. 392–4.

[7] Bove JM, Erti Dwiastuti M, Triviratno A, Supriyanto A, Nasli E,

Becu P, et al. Incidence of huanglongbing and citrus rehabilitation in

North Bali, Indonesia. In: Proceedings of 14th international conference

on organization citrus virologists. Riverside, California: Department of

Plant Pathology; 2000. p. 200–6.

[8] Maixner M, Ahrens U, Seemuller E. Detection of German grapevine

yellows (Vergilbungskrankheit) MLO in grapevine, alternative hosts

and a vector by a specific PCR procedure. Eur J Plant Pathol 1995;101:

241–50.