Previous Article | Next Article ![]()
Applied and Environmental Microbiology, November 2005, p. 6473-6478, Vol. 71, No. 11
0099-2240/05/$08.00+0 doi:10.1128/AEM.71.11.6473-6478.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Laboratoire de Biologie Cellulaire et Moléculaire, UMR GDPP, INRA, Université Victor Segalen Bordeaux 2, 71 avenue Edouard Bourlaux BP81, 33883 Villenave D'Ornon Cedex, France
Received 17 December 2004/ Accepted 6 June 2005
|
|
|---|
|
|
|---|
To further characterize the liberibacters at the molecular level, additional DNA fragments have been isolated by the randomly amplified polymorphic DNA method (18). Sequence analyses of six such fragments showed that they were part of the "Ca. Liberibacter" genome and led to identification of the nusG, pgm, and omp genes, as well as a gene encoding a conserved hypothetical protein. The omp gene, encoding an outer membrane protein (Omp), was thought to be the most promising candidate for studying inter- and intraspecies variability. In bacteria, Omps are involved in various functions. They participate in exchanges with the external environment, and in some cases they may be involved in pathogenicity (23). Although the three-dimensional structures of Omps are conserved, their amino acid sequences are not. The omp nucleotide sequences show considerable variation between species and even between strains of the same species, and they have been shown to be helpful for studying bacterial biodiversity (4, 7, 29, 36). For example, omp has been used to differentiate Brucella species, which, like the liberibacters, belong to the
subdivision of the Proteobacteria (5).
In this study, we determined the nucleotide sequences of the omp genes of "Ca. Liberibacter africanus" and "Ca. Liberibacter asiaticus" isolates, and we found that each isolate can be characterized by a specific PCR-restriction fragment length polymorphism (RFLP) profile.
|
|
|---|
DNA isolation.
Total DNA of a Citrus plant was extracted from 0.5 g of leaf midribs using a Wizard DNA purification kit (Promega Biosciences Inc., San Luis Obispo, Calif.) by the method of Jagoueix et al. (20). Total DNA of periwinkle plants was extracted from 2 g of leaf midribs by the hexadecyltrimethylammonium bromide method of Murray and Thompson (26). Plant DNA preparations were treated with DNase-free RNase prior to further analyses.
Plasmids were prepared from Escherichia coli XL1-Blue transformants with the Wizard Plus Miniprep DNA purification system (Promega Biosciences Inc., San Luis Obispo, Calif.).
PCR amplification and RFLP analyses.
Amplification was carried out in a 25-µl reaction mixture containing 2 µl of "wizard extract", 0.1 µg of "hexadecyltrimethylammonium bromide-extracted" DNA, or 1 ng of purified plasmid as the target DNA and 2 U of Taq DNA polymerase (Promega Biosciences Inc., San Luis Obispo, Calif.) with the buffer recommended by the supplier. Amplification was carried out for 40 cycles, each consisting of 40 s at 92°C, 40 s at 55°C, and 3 min at 72°C. The annealing temperature was determined based on the melting temperatures of the primers used for amplification. Primer sequences are listed in Table 1. The positions of primers are shown in Fig. 1. PCR detection of "Ca. Liberibacter" in the plants was carried out with primers OA1 and OI2c or primers OI1 and OI2c as described previously (19, 20). For RFLP analyses, the entire omp gene of "Ca. Liberibacter" was amplified with primers HP1asinv and Lp1c. PCR products were digested with TacI, ApoI, SwaI, SacI, MssI, and SspI, and the DNA fragments were separated by 2% agarose gel electrophoresis using standard procedures.
|
View this table: [in a new window] |
TABLE 1. Primers used in this study
|
![]() View larger version (21K): [in a new window] |
FIG. 1. Gene organization of the omp gene regions of "Ca. Liberibacter africanus" (isolate "SouthAfrica-Nelspruit") and "Ca. Liberibacter asiaticus" (isolate "India-Poona"). The arrows indicate the positions of primers. The arrowheads indicate the directions of transcription.
|
Southern blot hybridization.
Total DNA (10 µg) extracted from periwinkle plants infected either with "Ca. Liberibacter africanus" (isolate "SouthAfrica-Nelspruit") or with "Ca. Liberibacter asiaticus" (isolates "India-Poona," "Thailand-NakhomPathom," and "Philippines-Lipacity") was digested with EcoRI. Restricted DNA was fractionated by 1% agarose gel electrophoresis, blotted onto positively charged nylon membranes by the alkali procedure, and hybridized with a digoxigenin-dUTP-labeled probe specific for the "Ca. Liberibacter africanus" omp gene. The 1,290-bp probe was obtained by PCR amplification with primers OMP1 and OMP2 in the presence of digoxigenin-dUTP (17). Hybridization signals were detected with anti-digoxigenin-alkaline phosphatase conjugate and 2-hydroxy-3-naphthoic acid-2'-phenylanilide phosphate as the substrate by following the supplier's instructions. Fluorescent signals were analyzed with a Fluor-S Multimager phosphorimager and the Quantity-One software (Bio-Rad Laboratories Inc., Hercules, Calif.).
Chromosome walking.
Chromosome walking was performed using a Universal Genome Walker kit (BD Biosciences Clontech, Palo Alto, Calif.) by following the manufacturer's instructions. DNA from periwinkle plants infected with "Ca. Liberibacter asiaticus" (isolate "India-Poona") was digested with restriction enzymes Bst1101, Eco105I, KspAI, SmaI, and StuI. For "Ca. Liberibacter africanus" (isolate "SouthAfrica-Nelspruit"), DNA was digested with restriction enzymes EcoRV, PvuII, ScaI, and StuI.
To construct genomic DNA libraries, each batch of digested DNA was ligated to an adapter. For "Ca. Liberibacter asiaticus," a first PCR amplification (seven cycles, each consisting of 25 s at 94°C and 3 min at 72°C; 32 cycles, each consisting of 25 s at 94°C and 3 min at 67°C; and a 7-min final extension at 67°C) using primers AP1 and OMP1asmc was followed by a nested PCR (five cycles, each consisting of 25 s at 94°C and 3 min at 72°C; 20 cycles, each consisting of 25 s at 94°C and 3 min at 67°C; and a 7-min final extension at 67°C) using internal primers AP2 and OMP2ascm. Primers AP1 and AP2 were designed from the adaptor sequence. In the case of "Ca. Liberibacter africanus," the primers used were primers OMP8 and AP1 for the PCR and primers OMP9 and AP2 for the nested PCR. After amplification, the larger PCR products were gel purified, cloned, and sequenced.
Sequence analyses.
DNA and protein sequence analyses were performed using the programs proposed by Infobiogen (http://www.infobiogen.fr/index.html). Searches for homologies in general databases (http://www.ncbi.nlm.nih.gov/BLAST) were carried out using the BLAST program (1). Multiple-sequence alignment was performed using the MULTALIN (6; http://www.toulouse.inra.fr/Igc/multalin/multalin.html) or CLUSTALW (32; http://www.infobiogen.fr/services/analyseq/cgi-bin/clustalw_in.pl) software. Phylogenetic trees were constructed by using the MEGA2 program (22; http://evolution.genetics.washington.edu/phylip.html). Six "Ca. Liberibacter" omp sequences were used in this work; one was from "Ca. Liberibacter africanus" isolate "SouthAfrica-Nelspruit" (accession no. AY642158), and five were from "Ca. Liberibacter asiaticus" isolates "India-Poona" (accession no. AY642159), "China-Behaï" (accession no. AY842429), "Thailand-NakhomPathom" (accession no. AY842432), "Nepal-Pokhara" (accession no. AY842430), and "Philippines-Lipacity" (accession no. AY842431).
In addition to the "Ca. Liberibacter" sequences, the omp sequences of the following bacteria were included in this study: Sinorhizobium meliloti (accession no. NP 385608; coding for a putative outer membrane protein), Agrobacterium tumefaciens (accession no. NP 354388; coding for AGR_C_2554p, an Omp1 precursor), Brucella suis (accession no. NP 698159; coding for a bacterial surface antigen), Brucella melitensis (accession no. NP 539747; coding for an outer membrane protein), Bartonella henselae (accession no. AAL 66374; coding for Omp89), Mesorhizobium loti (accession no. NP_102404; coding for an outer membrane protein), Rhodospirillum rubrum (accession no. ZP_00015130; coding for an outer membrane protein), Bradyrhizobium japonicum (accession no. NP_770352; coding for a probable outer membrane protein), and E. coli (accession no. NP_285871; yaeT gene coding for a hypothetical protein).
|
|
|---|
Similar experiments were carried out with "Ca. Liberibacter asiaticus" isolate "India-Poona", which was also shown to possess omp sequences by Southern blot hybridization with the omp probe (data not shown).
To determine the entire sequence of the "Ca. Liberibacter asiaticus" isolate "India-Poona" omp gene, the 5' part was first amplified with primers HP1 and OMP8inv, designed from the sequence of "Ca. Liberibacter africanus" (Table 1 and Fig. 1), and the nucleotide sequence of the 1.2-kbp fragment was determined. Then the nucleotide sequence of the 3' part of the gene was determined by the chromosome walking technique.
The complete sequence of the "Ca. Liberibacter africanus" omp gene was obtained by combining the nucleotide sequences of the 1,285-bp OMP1-OMP2 and the 1,500-bp OMP8-Lp3c fragments (primer Lp3c was designed from the "Ca. Liberibacter asiaticus" sequence). Additional sequences flanking the omp gene were determined by the chromosome walking method.
The sequenced regions of "Ca. Liberibacter africanus" and "Ca. Liberibacter asiaticus" were 3,694 and 3,701 bp long, respectively, and were found to have the same gene organization (Fig. 1). The omp region comprises three putative coding sequences, the 3' end of an incomplete open reading frame (ORF), a short 71-bp ("Ca. Liberibacter africanus") or 66-bp ("Ca. Liberibacter asiaticus") intergenic region, the omp gene followed by a short 47-bp ("Ca. Liberibacter africanus") or 29-bp ("Ca. Liberibacter asiaticus") intergenic region, and the 5' end of an incomplete ORF. The incomplete ORFs flanking omp were found to be highly conserved (more than 57% identity for the ORF upstream of omp and 68% identity for the ORF downstream of omp) in the two "Ca. Liberibacter" species. The ORF upstream of omp putatively encodes a conserved hypothetical polypeptide exhibiting 50% similarity with a hypothetical zinc metallopeptidase of A. tumefaciens str. C58 (best hit for "Ca. Liberibacter africanus") and 51% similarity with a putative integral membrane protein of Neisseria meningitidis Z2491 (best hit for "Ca. Liberibacter asiaticus"). The incomplete ORF downstream of omp encodes a putative 146-amino-acid polypeptide in the case of "Ca. Liberibacter africanus" and a putative 271-amino-acid polypeptide in the case of "Ca. Liberibacter asiaticus," which exhibit 63% and 70% similarity, respectively, with a probable UDP-3-O-3-hydroxymyristoyl glucosamine N-acyltransferase of S. meliloti that is encoded by the lpxD gene and is involved in lipopolysaccharide biosynthesis (28). The gene organization of the omp region of "Ca. Liberibacter" species was similar to the organization found in the
-proteobacteria S. meliloti, A. tumefaciens, and B. henselae, as well as in the
-proteobacteria E. coli and Salmonella enterica serovar Typhi (15). In these two organisms, however, an additional gene encoding a hypothetical protein was located between the Omp-encoding gene yaeT and lpxD.
Sequence analyses of "Ca. Liberibacter africanus" and "Ca. Liberibacter asiaticus" omp genes.
The omp genes of "Ca. Liberibacter africanus" (isolate "SouthAfrica-Nelspruit") and "Ca. Liberibacter asiaticus" (isolate "India-Poona") were 2,340 and 2,346 bp long, respectively, beginning with a GTG start codon preceded by a ribosome binding site and ending with a TAG stop codon. The nucleotide sequences exhibited 72.2% identity, whereas the encoded polypeptides had 58% identical and 86.5% similar amino acids. Protein alignment also showed that blocks of identical amino acids were distributed all along the protein sequence with the exception of two short regions, one at the N-terminal end (positions 1 to 39) and the other in the central part of the protein (positions 508 to 565).
A BLAST search for homologies revealed that the Omp of "Ca. Liberibacter africanus" exhibited 37% identity and 55% similarity over 743 amino acids with the group 1 outer membrane protein precursor Omp1 of A. tumefaciens (accession no. NP_354388.1) and 37% identity and 56% similarity with a putative outer membrane protein of S. meliloti, whereas the Omp of "Ca. Liberibacter asiaticus" exhibited 40% identity and 62% similarity over 744 amino acids and 38% identity and 61% similarity over 743 amino acids with the outer membrane proteins of S. meliloti and A. tumefaciens, respectively. A search for conserved domains showed that the "Ca. Liberibacter" Omp protein belongs to the COG4775 family, which includes the outer membrane protective antigen Oma87 from Pasteurella multocida (23) and Omp1 of A. tumefaciens (16). Interestingly, structure predictions revealed that Omp of "Ca. Liberibacter" had an organization similar to that of A. tumefaciens, S. meliloti, B. melitensis, and B. henselae. In particular, it possessed a single hydrophobic transmembrane segment (positions 19 to 41), while the major part of the polypeptide was amphiphilic, suggesting that it is exposed at the cell surface. Most of the Omps are known to have a phenylalanine (F) residue as the carboxyl-terminal amino acid (30). In the "Ca. Liberibacter" species, however, the carboxyl-terminal amino acids were arginine (R) for "Ca. Liberibacter africanus" and methionine (M) for "Ca. Liberibacter asiaticus." A similar situation was found in E. coli, in which the yaeT-encoded polypeptide ends with a tryptophan (W) residue (accession no. NP_414719) (2).
Phylogenetic analyses.
To further characterize the "Ca. Liberibacter" Omp, the nucleotide sequences of the omp genes from five isolates of "Ca. Liberibacter asiaticus" (isolates "India-Poona," "Thailand-NakhomPathom," "Philippines-Lipacity," "China-Behaï," and "Nepal-Pokhara") were determined. These sequences were aligned with the omp sequences from "Ca. Liberibacter africanus" isolate "SouthAfrica-Nelspruit" and nine other bacteria, including eight
-proteobacteria and the
-proteobacterium E. coli. A phylogenetic tree was constructed using the maximum-parsimony method (Fig. 2). The finding that all "Ca. Liberibacter" isolates grouped together was consistent with phylogenetic studies based on the 16S rRNA gene sequences showing that "Ca. Liberibacter" species are members of a new lineage in the
subdivision of the Proteobacteria (19). Interestingly, the group was divided into two subgroups corresponding to the two "Ca. Liberibacter" species, one containing the "Ca. Liberibacter africanus" "SouthAfrica-Nelspruit" isolate and the other comprising all five "Ca. Liberibacter asiaticus" isolates. Whereas the levels of identity between the Omp amino acid sequences of "Ca. Liberibacter africanus" isolate "SouthAfrica-Nelspruit" and "Ca. Liberibacter asiaticus" ranged from 58.5 to 59%, the levels of identity were 99.5 to 99.9% for the various isolates of "Ca. Liberibacter asiaticus." As shown by the omp phylogenetic tree, the liberibacters are more closely related to proteobacteria belonging to the
-2 subgroup.
![]() View larger version (18K): [in a new window] |
FIG. 2. Maximum-parsimony phylogenetic tree generated from alignment of omp sequences from "Ca. Liberibacter" and other -proteobacteria. E. coli was added as an outgroup. Bootstrap values are shown at the nodes.
|
![]() View larger version (80K): [in a new window] |
FIG. 3. TaqI restriction profile analyses of PCR-amplified DNAs from Citrus infected with various isolates of "Ca. Liberibacter asiaticus." Amplification was performed with primers HP1asinv and Lp1c. Lane M, 1-kb DNA ladder; lane 1, "India-Poona"; lane 2, "Vietnam-Bentre"; lane 3, "Thailand-NakhomPathom"; lane 4, "Philippines-Lipacity"; lane 5, "China-Behaï"; lane 6, "Nepal-Pokhara"; lane 7, "Taiwan-Taiwan12"; lane 8, "China-Yunnan"; lane 9, "India-Chetally"; lane 10, "India-Kodali."
|
|
View this table: [in a new window] |
TABLE 2. RFLP profiles of isolates of "Ca. Liberibacter asiaticus"
|
|
|
|---|
In this work, we studied the genetic variability of "Ca. Liberibacter" isolates by looking at the putative sequence variations of the omp gene. The omp regions of "Ca. Liberibacter africanus" and "Ca. Liberibacter asiaticus" were found to have the same gene organization as the omp regions of other
-proteobacteria, and the genes upstream and downstream of omp were homologous to E. coli yaeL and lpxD, respectively. Although the Omps are conserved among gram-negative bacteria, they exhibit sequence variations, which have been used to study biodiversity in various pathogenic bacteria (4, 5, 7). In "Ca. Liberibacter," the levels of amino acid identity ranged from 58.5% between "Ca. Liberibacter africanus" isolate "SouthAfrica-Nelspruit" and "Ca. Liberibacter asiaticus" isolate "India-Poona" to 99.9% between isolates "Philippines-Lipacity" and "Thailand-NakhomPathom." The phylogenetic tree constructed from omp DNA sequence alignments matched exactly the tree based on the 16S rRNA gene sequences, suggesting that the omp gene is part of the core genome and was not acquired by horizontal transfer from unrelated bacteria. As expected, all "Ca. Liberibacter asiaticus" isolates clustered in a branch distinct from the branch containing "Ca. Liberibacter africanus." The phylogenetic tree also showed that "Ca. Liberibacter asiaticus" isolates were distributed in two groups depending on their geographical origins; one group contained isolates from India and Nepal, and the other contained isolates from Thailand, The Philippines, and China. However, due to the high levels of identity, resulting in low bootstrap values, and the relatively low numbers of isolates available, the true occurrence of these two groups should be investigated further. Interestingly, however, based on the omp nucleotide sequence, each "Ca. Liberibacter asiaticus" isolate could be identified by a specific PCR-RFLP pattern, in spite of the high levels of identity. In particular, isolates from similar geographic areas, such as "China-Yunnan" and "China-Behaï" or "India-Poona" and "India-Kodali," did not have similar RFLP patterns. Isolates "India-Kodali" and "India-Chetally," which had identical RFLP profiles, probably represent a single isolate, as they were collected at the same time from the same citrus orchard (J. M. Bové and M. Garnier, unpublished data). The genetic diversity of "Ca. Liberibacter asiaticus" isolates as revealed by omp-based PCR-RFLP analysis is in good agreement with serological studies showing the occurrence of seven distinct serogroups (9). However, in contrast to RFLP analysis, serological methods did not necessarily differentiate isolates from different countries (9). Thus, RFLP analysis seems to be more discriminating. As a whole, these results indicate that "Ca. Liberibacter asiaticus," even within a given region, includes several different variants, which can be distinguished by a PCR-RFLP method based on the omp gene sequence. This technique might also be useful for studying "Ca. Liberibacter" diversity on the basis of phenotypic traits, such as virulence and insect transmissibility. Huanglongbing has recently been reported from São Paulo State, Brazil, and a new "Ca. Liberibacter" species, "Ca. Liberibacter americanus," was found to be associated with the disease (31).
|
|
|---|
subdivision of the Proteobacteria. Int. J. Syst. Bacteriol. 44:379-386.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»