• bacterial wilt;
  • phylotype;
  • population genetics;
  • sequevar


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

The genetic diversity of Ralstonia solanacearum causing bacterial wilt of tomato in Trinidad was assessed using the hierarchical phylotyping scheme and rep-PCR DNA fingerprinting. Seventy-one isolates were collected in 2003 on infected tomato crops in the four main vegetable cropping areas of Trinidad (North, Central, South-East and South). Two phylotypes were present, with phylotype II being much more prevalent (66%) than phylotype I (34%). Phylotype II strains consisted mainly of sequevar 7 in Central and South-East, and sequevar 35 in North, South-East and South. This is the first report of sequevar 7 outside south-eastern USA. In contrast, no ‘brown rot’ (phylotype IIB/1, race 3 biovar 2) or emerging strains of phylotype IIB/4NPB were identified. Rep-PCR data were used to assess population genetic structure. No significant clustering by geographical distance was found, suggesting regular gene flow among cropping areas (via waterways, plant or soil). However, the population from Central was significantly differentiated from the others, containing only phylotype II/seq 7 strains, with a high degree of clonality, suggesting a possible recent introduction from abroad. The South population was less aggressive and more genetically diverse, suggesting horizontal gene transfers within the population, even among isolates of different phylotypes. Phylotype I and phylotype II populations differed slightly in clonality levels, with indications of more frequent recombination events within phylotype I populations. Possible factors influencing genetic diversity and distribution within the island are discussed.


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

Ralstonia solanacearum is one of the most important phytopathogenic bacteria worldwide, both in terms of host range and destructive capability. This pathogen is responsible for lethal wilting diseases in over 200 species of plants in more than 50 families throughout the wet tropics and subtropical regions around the world (Hayward, 1994). The importance of R. solanacearum as a plant pathogen became evident in the 1890s when it was described by E.F. Smith as the causative agent of wilt disease of tomato and other solanaceous crops in southern USA (Kelman, 1954) and subsequently in the entire Americas, Asia, Africa and Australia (Hayward, 1994).

Members of R. solanacearum comprise a relatively diverse group of strains referred to as a species complex (Gillings & Fahy, 1994), consisting of four phylotypes of different origins (I: Asia; II: America; III: Africa and Indian Ocean; IV: Indonesia) (Fegan & Prior, 2005). Phylotypes can be further divided into ‘sequevars’ based on variations in sequences of the endoglucanase (egl) gene (Fegan & Prior, 2005; Wicker et al., 2007). Clonal lines within sequevars can then be determined using genome fingerprinting methods, such as rep-PCR (Fegan, 2005).

Trinidad has a very long history of bacterial wilt disease caused by R. solanacearum. Rorer (1911) showed that wilting of banana and plantain in the island was due to a bacterium he named Bacillus musae, but which was subsequently shown to be the bacterial wilt pathogen then known as Bacillus solanacearum. Briant (1932) reported widespread distribution of tomato bacterial wilt in Trinidad and also indicated the presence of the disease in other countries in the region including St. Lucia in 1904, St. Vincent in 1917 and British Guiana in 1925. During the period that followed his report, there have been several studies indicating the serious challenge bacterial wilt poses to tomato production in Trinidad. Despite the disease problems, tomato production has remained an important farming activity in the Republic of Trinidad and Tobago.

Strains from solanaceous crops have been described in Venezuela (Garcia et al., 1999a,b), Cuba (Stefanova, 1998) and the French West Indies (Prior & Steva, 1990; Wicker et al., 2007, 2009b). In Martinique, ‘emerging strains’ with a broad host range were identified. These strains with the capacity to overcome the main sources of resistance to tomato wilt were rapidly spreading throughout tomato cropping areas (Wicker et al., 2009a). These strains may be latently prevalent throughout the Caribbean and the Guyanas, and may constitute a real threat for all tomato farmers in the region.

Understanding local pathogen diversity is the foundation for a successful breeding and integrated management programme (Sanchez Perez et al., 2008). Despite the long history of the presence of R. solanacearum in Trinidad, there are no detailed reports available on the diversity and distribution of strains from this island. The first objective of this study was to assess the overall genetic diversity of R. solanacearum strains infecting tomato in Trinidad, compared with known populations of neighbouring islands. A second objective was to assess population genetic structure related to spatial and phylogenetic distance.

Materials and methods

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

Bacterial strains

Seventy-one isolates of R. solanacearum were isolated from infected tomato plants collected within a 1-week period in 2003 from fields situated in four major tomato growing regions in Trinidad: North (Macoya and Maloney), Central (Gasparillo), South-East (Rio Claro) and South (Barrackpore and Penal) (Fig. 1). A maximum of 10 tomato plants with symptoms were sampled per field (depending on field size) and single isolates were recovered from stems of individual wilted plants on tetrazolium chloride (TZC) agar medium (Kelman, 1954). After incubation at 30°C for 48 h, single colonies displaying the typical R. solanacearum morphology (white creamy, bird-eye shaped colonies with a pink centre) were subcultured on TZC medium until purified. Pure cultures were then stored in sterile water. All the isolates were confirmed as capable of wilting susceptible tomato variety Akash using the stem inoculation technique (Winstead & Kelman, 1952).


Figure 1.  Map of Trinidad showing origin of Ralstonia solanacearum isolates used in this study.

Download figure to PowerPoint

DNA extraction

DNA was isolated for phylotyping and rep-PCR DNA fingerprinting using a modified SDS lysis method (Sambrook et al., 1989). Briefly, one loopful of bacterial suspension was quadrant streaked onto TZC agar and incubated at 30°C for 48 h. Single colonies were then transferred to 5 mL of BG broth (Boucher et al., 1985) and incubated in a shaking incubator (250 rpm) at 30°C for 24 h. The BG broth cultures (1·5 mL per isolate) were pelleted by centrifuging at 12 000 g for 30 s in 2 mL centrifuge tubes and the pellet was then resuspended in 400 μL Tris-EDTA buffer (TE). The cells were then lysed at 37°C for 1 h after adding 50 μL of 10% SDS and 50 μL proteinase K (20 mg mL−1 in TE). Two extractions were carried out with 500 μL of phenol:chloroform followed by a single chloroform extraction. DNA was precipitated with 25 μL of 5 m NaCl and 1 mL of 95% ethanol. After centrifuging for 10 min at 12 000 g, the resulting pellet was resuspended in 100 μL TE. The DNA was further purified by adding 2·5 μL of 10 mg mL−1 RNase and incubating at 37°C for 30 min. DNA was then precipitated by adding 40 μL of 5 m NH4Ac and 250 μL isopropanol followed by incubation at room temperature for 5 min. The resulting precipitate was pelleted by centrifugation for 10 min at 12 000 g and then washed twice with 70% ethanol before drying and finally dissolving in 100 μL TE.

DNA typing

Two multiplex-PCR (mx-PCR) procedures were applied to characterize strains based on the hierarchical classification scheme described previously (Fegan & Prior, 2005; Prior & Fegan, 2005a). ‘Phylotype’ multiplex PCR (Pmx-PCR) (Fegan & Prior, 2005) was performed to identify the phylotype to which a strain belonged. A ‘Musa’ multiplex-PCR (Mmx-PCR) (Prior & Fegan, 2005a) was further performed on all phylotype II isolates to determine if strains belonged to sequevars 3, 4 or 6 (Moko disease-inducing strains). Primer sequences and PCR reactions were as previously described (Wicker et al., 2007). Taq DNA polymerase and dNTPs were from either Invitrogen Life Technologies or Promega France and primers used were either from Integrated DNA Technologies (IDT) Inc. or Eurogentec.

The PCR products were resolved through 2% agarose gels in 0·5 × TBE buffer and revealed by staining in 10 μg mL−1 ethidium bromide. Amplicon sizes were estimated by comparison to 1 kb or 100 bp (Invitrogen Life Technologies) DNA ladders. The following strains were used as positive controls for the phylotypes and sequevars: GMI1000, phylotype I; ANT307, phylotype IIB/sequevar 4NPB; CFBP3059, phylotype III; MAFF301558, phylotype IV; UW011, phylotype IIB/sequevar 3; UW21, phylotype IIB/sequevar 6; UW175, phylotype IIB/sequevar 4-SFR; UW162, phylotype IIB/sequevar 4A (Table 1).

Table 1. Ralstonia solanacearum strains from Trinidad and reference strains used in this study
StrainOriginaPhylotype/sequevar determined bybBOX-PCR ≥ 5 bands/straincReferencesdGenBank accession no.
  1. aRegion of origin for the Trinidad tomato strains (South = Penal and Barrackpore; South-East = Rio Claro; Central = Gasparillo; North = Maloney and Macoya), host of isolation and country for the reference strains.

  2. bThe phylotypes and sequevars determined by PCR were those determined by Pmx-PCR and Mmx-PCR (Fegan & Prior, 2005; Prior & Fegan, 2005a). The strains marked II gave no signal when tested by the Mmx-PCR and by the 630/631 primers (Fegan et al., 1998). The phylotype and sequevars determined by trees were determined by egl sequence analysis.

  3. cNT: not tested.

  4. dReference numbers are given with one exception. TS, this study.

  5. eCFBP: Collection Française de Bactéries Phytopathogènes, Angers, France; IBSBF: Coleção de Culturas de Fitobactérias do Instituto Biológico, São Paulo, Brazil.

Tomato strains from Trinidad (2003)
 RF8SouthIIIIA/35*TS EU726805
 RF9SouthIIIIA/35*TS EU726827
 RF10SouthII *TS 
 RF11SouthIIIIA/35*TS EU726799
 RF12SouthIIIIA/35*TS EU726800
 RF13SouthIIIIA/35 TS EU726828
 RF14SouthIIIIA/35*TS EU726801
 RF15SouthI *TS 
 RF16SouthII *TS 
 RF17SouthII *TS 
 RF18SouthII *TS 
 RF19SouthII *TS 
 RF20SouthIIIIA/7*TS EU726829
 RF50SouthI *TS 
 RF44SouthI *TS 
 RF45SouthI *TS 
 RF47SouthI *TS 
 RF48SouthI *TS 
 RF49SouthI *TS 
 RF7SouthIIIIA/35 TS EU726825
 RC8SouthIIIIA/38 TS EU726826
 RF13SouthIIIIA/35 TS EU726828
 RF46SouthI  TS 
 RF42SouthII  TS 
 RF43SouthI  TS 
 RF61SouthI  TS 
 RF51SouthI *TS 
 RF53SouthII/18*TS EU726842
 RF54SouthI *TS 
 RF55SouthI *TS 
 RF56SouthII/18*TS EU726833
 RF58SouthII/18*TS EU726821
 RF60SouthII/18*TS EU726834
 RF41SouthI *TS 
 RF57SouthII/18*TS EU726843
 RF21NorthIIIIA/35*TS EU726830
 RF22NorthII  TS 
 RF23NorthIIIIA/35 TS EU726802
 RF24NorthIIIIA/35*TS EU726803
 RF25NorthI *TS 
 RF26NorthII/14*TS EU726831
 RF29South-EastII *TS 
 RF30South-EastIIIIA/7*TS EU726807
 RF31South-EastIIIIA/7 TS EU726808
 RF32South-EastIIIIA/7*TS EU726810
 RF34South-EastI *TS 
 RF35South-EastIIIIA/7*TS EU726812
 RF27South-EastIIIIA/35*TS EU726804
 RF28South-EastI *TS 
 RF37South-EastIIIIA/35 TS EU726822
 RF38South-EastIIIIA/35*TS EU726832
 RF40South-EastII *TS 
 RF70CentralIIIIA/7*TS EU726814
 RF72CentralIIIIA/7*TS EU726815
 RF73CentralIIIIA/7*TS EU726816
 RF74CentralIIIIA/7*TS EU726839
 RF75CentralIIIIA/7 TS EU726840
 RF76CentralIIIIA/7*TS EU726817
 RF77CentralIIIIA/7*TS EU726841
 RF78CentralIIIIA/7*TS EU726818
 RF79CentralIIIIA/7*TS EU726826
 RF80CentralIIIIA/7*TS EU726819
 RF81CentralIIIIA/7*TS EU726809
 RF63CentralIIIIA/7 TS EU726835
 RF64CentralIIIIA/7 TS EU726836
 RF65CentralIIIIA/7 TS EU726837
 RF66CentralIIIIA/7 TS EU726838
 RF67CentralII  TS 
 RF68CentralIIIIA/7 TS EU726813
 RF69CentralII  TS 
Reference strains
 R292 Morus alba, China I/12NT(Prior & Fegan, 2005a) AF295255
 GMI1000Tomato, French Guiana I/18NT(Fegan & Prior, 2005) AF295251
 JT523Potato, Reunion Is. I/18NT(Prior & Fegan, 2005a) AF295252
 NCPPB3190Tomato, Malaysia I/18NT(Prior & Fegan, 2005a) AF295253
 ACH92Ginger, Australia I/16NT(Prior & Fegan, 2005a) AF295254
 JS759, NCPPB2198 Musa sp., Trinidad (1968) I/18NT(Poussier et al., 2000) EU726820
 CFBP6801e Heliconia caribea, Martinique IIA/NDNT(Wicker et al., 2007) EF371836
 CFBP6779e Canna indica, Martinique IIA/NDNT(Wicker et al., 2007) EF371835
 IBSBF1546e Fucsia sp., Brazil IIA/35NT(Wicker et al., 2007) EF371838
 IBSBF1900e Musa sp., Brazil IIA/24NT(Wicker et al., 2007) EF371839
 MOLK2 Musa sp., Philippines IIA/3NT(Wicker et al., 2007) EF371841
 UW9 Heliconia, Costa Rica IIA/3NT(Poussier et al., 2000) AF295257
 CFBP7016e Anthurium andreanum, Martinique IIA/39NT(Wicker et al., 2007) EF371828
 CFBP2958eTomato, Guadeloupe IIA/39NT(Prior & Fegan, 2005a) AF295266
 A3909 Heliconia, Hawaii IIA/6NT(Wicker et al., 2007) EF371812
 A3911 Heliconia, Hawaii IIA/6NT  
 A3907 Heliconia, Hawaii IIA/6NT  
 DAR64836 Musa sp., Australia IIA/6NT(Fegan & Prior, 2006) DQ011551
 RUN0582 Musa sp., Grenada (1984) IIA/6NT  
 RUN0394 Musa sp., Grenada (2007) IIA/6NT  GU295048
 ICMP7963Potato, Kenya IIA/7NT(Prior & Fegan, 2005a) AF295263
 K60Tomato, USA IIA/7NT(Prior & Fegan, 2005a) AF295262
 IPO1609Potato, Netherlands IIB/1NT(Prior & Fegan, 2005a) EF371814
 JT516Potato, Reunion Is. IIB/1NT(Prior & Fegan, 2005a) AF295258
 CFBP1183 Heliconia, Costa Rica IIB/3NT(Prior & Fegan, 2005a) EF371805
 CFBP6784 Anthurium sp., Martinique IIB/4NT(Wicker et al., 2007) EF371813
 CFBP6783 Heliconia caribea, Martinique IIB/4NT(Wicker et al., 2007) EF371817
 IBSBF1503 Cucumis sativus, Brazil IIB/4NT(Wicker et al., 2007) EF371840
 IBSBF1454 Cucurbita pepo, Brazil IIB/4NT(Wicker et al., 2007) EF371844
 UW70 Musa sp., Colombia IIB/4NT(Wicker et al., 2007) DQ011550
 UW129 Musa sp., Peru IIB/4NT(Prior & Fegan, 2005a) EF371811
 UW162 Musa sp., Peru IIB/4NT(Prior & Fegan, 2005b) AF295256
 CFBP7014 A. andreanum, Trinidad IIB/NDNT(Wicker et al., 2007) AF371831
 NCPPB3987Potato, Brazil IIB/NDNT(Poussier et al., 2000) AF295261
 UW477Potato, Peru IIB/NDNT(Poussier et al., 2000) AF295260
 CFBP3059Aubergine, Burkina Faso III/19-23NT(Fegan & Prior, 2005) AF295270
 CFBP734Potato, Madagascar III/19-23NT(Prior & Fegan, 2005a) AF295274
 NCPPB332Potato, Zimbabwe III/19-23NT(Prior & Fegan, 2005a) AF295276
 JT525 Pelargonium asperum, Reunion Is. III/19-23NT(Prior & Fegan, 2005a) AF295272
 MAFF301558Potato, Japan IV/10NT(Prior & Fegan, 2005a) DQ011558
 PSI7Tomato, Indonesia IV/10NT(Prior & Fegan, 2005a) EF371804

The race 3/biovar 2 (R3bv2)-specific PCR primer pair 630/631 (Fegan et al., 1998) was used to identify R3bv2 strains, using strains CFBP4611 and CFBP4957 as positive controls, as previously described (Mahbou Somo Toukam et al., 2009).

DNA sequencing and analysis of partial endoglucanase (egl) gene

PCR amplification of a 750 bp region of the egl gene was performed using the primer pair Endo-F (5′-ATGCATGCCGCTGGTCGCCGC-3′) and Endo-R (5′-GCGTTGCCCGGCACGAACACC-3′) (Fegan & Prior, 2005). The reaction mixture and PCR conditions were as previously described (Wicker et al., 2007; Mahbou Somo Toukam et al., 2009).

The partial sequences of the megaplasmid virulence-related endoglucanase (egl) gene were analysed using the arb Software Environment (Ludwig et al., 2004). Sequences were manually aligned using the arb sequence editor. Phylogenetic trees were then constructed from the genetic distance data by three methods: (i) maximum likelihood (ML) method based on the nucleotide substitution model GTR + G4 (Rodríguez et al., 1990), as automatically determined by the phyml software implemented within rdp 3.5 (Martin et al., 2010) with 5000 bootstrap resamplings, (ii) neighbour-joining (NJ) method (Saitou & Nei, 1987) based on the substitution model of Jukes & Cantor (1969) with 5000 bootstrap resamplings, and (iii) the maximum parsimony (MP) method using the phylip dnapars package of the arb software. Sequences of egl reference strains from GenBank were retrieved and included in the analysis with those of the newly described strains from this study, which were also deposited in GenBank (Table 1).


The BOX A1R PCR primer (5′-CTACGGCAA GGCGACGCTGACG-3′) (Versalovic et al., 1994) was synthesized by Integrated DNA Technologies, Inc. The PCR amplification was performed using a Techne Touchgene Gradient Thermocycler. Each PCR reaction (25 μL) contained 2·5 μL PC2 Reaction Buffer (supplied by the manufacturer), 2·5 mm MgCl2, 10 mm dNTPs, 10 pmol of BOX A1R primer, 0·625 U KlenTaq DNA polymerase (DNA Polymerase Technology Inc.) and ∼10 ng of template DNA. Cycling conditions were: an initial denaturation for 7 min at 95°C; 30 cycles of 94°C/1 min, 53°C/1 min, 65°C/8 min; and a final extension at 65°C/16 min (Versalovic et al., 1994). Amplified PCR products (20 μL) were resolved by gel electrophoresis on a 1% agarose gel as described above, and bands were scored manually. One randomly selected isolate from the culture collection of the Department of Life Sciences, UWI, St. Augustine (RF62) was used as a reference in each gel for PCR quality control (band intensity) to ensure reproducibility of bands and accurate scoring.

Aggressiveness on tomato

The aggressiveness of 19 randomly selected isolates was determined on six commercial tomato varieties (Akash, Calypso, Gempack, Gempride, Heatmaster and Hybrid 6) using a modified seedling inoculation technique described by Jaunet & Wang (1999). Tomato seedlings (∼28 days old) grown in a commercial potting medium (Premier® Promix®) in ∼300 mL styrofoam containers were inoculated each with 108 cells of R. solanacearum isolates by soil drenching. The inoculated seedlings were placed in a greenhouse under natural conditions (temperature range 24–32°C) and number of seedlings exhibiting signs of wilting recorded every 2 days for up to 20 days after inoculation. The seedlings were irrigated as necessary and fertilized weekly with 20-20-20 NPK fertilizer (Nutrex®). The experiment was conducted using a split-plot design with two replications. Isolates were assigned as main plots and varieties as subplots, which contained 12 seedlings per variety.

Data analysis

As the samples were haploid, each BOX fragment was scored as one putative locus with two alleles, one indicating presence and the other indicating absence of the fragment. The number of bands scored ranged between one and 20 per isolate and the band sizes ranged from approximately 500 to 3050 bp. Fifteen loci were considered and 55 isolates of known origin giving at least five fragments were retained for further analysis. Population genetic analysis was carried out to determine gene diversity, genetic distance, clonality and genetic differentiation. To test for linkage disequilibrium in the R. solanacearum 55 strains-data set, the software multilocus 1.3 was used ( to estimate the index of association, Ia (Maynard Smith et al., 1993; Burt et al., 1996) and rd (Agapow & Burt, 2001), as well as the standardized index of association Isa by using lian 3.5 (Haubold et al., 1998; Haubold & Hudson, 2000). The association between the scored alleles was estimated by comparing the variance of the genetic distances of the actual data set to the mean variance of 1000 artificial resampled data sets. Values of Ia and inline image differing significantly from 0 reject the null hypothesis of random mating, whereas panmixis and sexual recombination are expected to result in Ia and inline image values close to zero (Burt et al., 1996; Haubold et al., 1998). The proportion of phylogenetically compatible pairs of loci (PrCL) was used to test for linkage disequilibrium in the data set. The null hypothesis of free recombination could be rejected if there were fewer than two locus pairs with all four allele combinations than expected under panmixis (Bennett et al., 2005). The significance was estimated from 1000 randomizations calculated with the multilocus 1.3 software.

The Weir & Cockerham’s (1984) theta values (θ) were calculated using multilocus 1.3 in order to estimate the geographic population differentiation of R. solanacearum between regions and between fields. This method estimates the difference in allele frequency distributions, independent of the number of populations and number of individuals sampled in each population (Weir & Cockerham, 1984). The null hypothesis of no differentiation was tested from 1000 randomizations. Nei’s analysis of genetic identity and distance was also applied (Nei, 1973), using popgene 1.32 ( to reveal hierarchical subdivisions within the sampling range. To correct for dominance, popgene uses a modification of the method of Clark & Lanigan (Chong et al., 1994).

To indirectly estimate the amount of dispersion, the isolation by distance model formulated by Rousset (1997) was used, which correlates θ/1−θ with spatial distances (one-dimension gene flow). The significance of this correlation was determined by the Mantel test (Mantel, 1967) using the zt software (Bonnet & Van De Peer, 2002). This test computes the linear correlation between two proximity matrices to reveal whether environmental variables are inter-correlated among themselves. The statistical significance of the resulting correlation coefficient was tested by performing 1000 random permutations of the data set and calculating the proportion yielding values that were equal to, or greater than, the observed coefficient of correlation. The correlation was tested at P = 0·05 according to Pearson’s coefficient. Analysis of variance (GenStat Discovery Edition 3) was used to test differences in aggressiveness among isolates on the different varieties of tomato.


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

Distribution of phylotypes

All 71 isolates gave a ∼280 bp band as expected for 759/760 primers (Opina et al., 1997) and were therefore confirmed as R. solanacearum. Almost two-thirds of the isolates (47/71) gave the phylotype II banding pattern while the remaining isolates (24/71) displayed the phylotype I pattern (Table 1). No phylotype II isolate gave a band with the 630/631 primers (Fegan et al., 1998) or the ‘Musa’ multiplex-PCR (Prior & Fegan, 2005a). Thus, all phylotype II isolates were non-sequevar 1 or 2 (race 3/biovar 2) and non-sequevar 3, 4, 6 or 24 (Moko strains). All regions except Central were found composed of a mixture of phylotypes I and II. However, phylotype distributions across locations were not homogenous (Fig. 2) as there was significant association of phylotype with region (chi-square test, P < 0·001). Based on the entire data set (71 strains), phylotypes I and II frequencies within South and Central regions were significantly different from their expected values (chi-square test, P = 0·006 and 0·002, respectively). The South population comprised a majority of phylotype I (20/36), but by contrast, the Central population (18 isolates) was comprised solely of phylotype II isolates. The number of isolates obtained from North and South-East was small, but both phylotypes were present in these regions (North: 2 phylotype I/4 phylotype II; South-East: 2 phylotype I/9 phylotype II).


Figure 2.  Phylotype distributions of Ralstonia solanacearum within each location. RS1: phylotype I and RS2: phylotype II. P: values of the chi-square goodness-of fit test; NS: not significant (P ≥ 0·05); *significant (P < 0·05); **highly significant (P < 0·01); ***very highly significant (P < 0·001).

Download figure to PowerPoint

Phylogenetic analysis

Analysis of the partial egl gene sequences of 45 representative strains from Trinidad and 40 reference strains (Fegan, 2005; Prior & Fegan, 2005a; Wicker et al., 2007) covering all four phylotypes showed that trees constructed with maximum likelihood, parsimony and distance-neighbour joining phylogenetic approaches were congruent (Figs S1 and S2). Thus only the ML phylogenetic tree is presented (Fig. 3), built by using the substitution model GTR + G4 (general-time reversible model with four gamma-shaped rate parameters; Rodríguez et al., 1990). All phylotype II strains from Trinidad clustered with the sequevars 7, 35 or 38 reference strains in the phylotype IIA branch, a subcluster containing strains isolated from diverse hosts (Fegan, 2005). None of the isolated tomato strains from this study was found in the subcluster IIB, which contains the Moko-inducing (sequevars 3, 4, 6), potato brown rot-inducing (sequevars 1 and 2) and the emerging strains (sequevar 4NPB) described in Martinique (Wicker et al., 2007). All phylotype IIA strains from Gasparillo (Central) were grouped with the sequevar 7 reference strain K60. By contrast, all the phylotype II strains from Maloney and Macoya (North) and all except one from Barrackpore-Penal (South) belonged to sequevar 35. Phylotype II strains from Rio Claro (South-East) grouped in both sequevars 7 and 35. All phylotype I isolates from Barrackpore-Penal (South) included in the phylogenetic analysis were sequevar 18 while the sole isolate from Maloney (North) was sequevar 14.


Figure 3.  Phylogenetic maximum likelihood tree based on partial endoglucanase (egl) gene sequences of strains from Trinidad and reference strains for the Ralstonia solanacearum species complex (nucleotide substitution model: GTR + G4). Strains recovered from tomato and originating from Trinidad are written in bold type. The number at each node is the bootstrap value (5000 resamplings). The scale bar represents one nucleotide substitution per 20 nucleotides.

Download figure to PowerPoint

Genotypic and gene diversity

Among the 55 isolates considered for population genetic analysis, phylotype II (62%) was predominant compared to phylotype I (38%). Forty-seven multilocus genotypes were observed within this collection. The largest number of genotypes was found in the South, followed by South-East, Central and North (26, 10, seven and four genotypes, respectively; Table 2). Nei’s gene diversity inline image (Nei, 1973) was lowest within Central region (inline image = 0·201, 53·33% polymorphic loci), followed by North (inline image = 0·245, 66·67% polymorphic loci), and South-East (inline image = 0·436, 100% polymorphic loci). The highest diversity was observed within the South region (inline image = 0·467, 100% polymorphic loci).

Table 2.   Gene diversity within South, South-East, Central and North populations of Ralstonia solanacearum from Trinidad
Populations n a g obs a inline image b %Polb
  1. a n: number of isolates; gobs: number of genotypes observed.

  2. b inline image: Nei’s (Nei, 1973) gene diversity; %Pol: percentage of polymorphic loci.

Maloney (North)540·24566·67
Gasparillo (Central)1070·20153·33
Rio Claro (South-East)11100·436100·00
Barrackpore-Penal (South)29260·467100·00

Spatial population genetic structure

According to the θ estimator, the population from Gasparillo (Central) was significantly differentiated (P < 0·05) from the other three populations, and Barrackpore-Penal (South) fields were also significantly different from Maloney-Macoya (North) (Table 3).

Table 3.   Genotypical differentiation estimated by θ statistics (Weir, 1996) between Ralstonia solanacearum district populations
  1. NS: not significant.

  2. aThe H0 hypothesis: ‘there is no differentiation between the populations’ was tested by comparing θ of observed populations to that of data sets whose individuals were randomly mixed. Here the actual data were compared to 1000 randomized data sets.

  3. Significant at *P < 0·05, **P < 0·01 and ***P < 0·001.


Genetic distances between regional populations (estimated by Weir’s θ) were not significantly correlated with physical distance between populations (r = 0·369; P = 0·383). The possible occurrence of recombination in the population was tested by analysing linkage disequilibrium in the data sets. The statistics on the Index of Association (Ia and rd) rejected the null hypothesis of random mating in all regional populations (Table 4). However, the proportion of compatible loci was not significantly different from the recombined data sets in Gasparillo, Maloney and Barrackpore-Penal, indicating that the populations in these areas may have experienced recombining events.

Table 4.   Estimates of linkage disequilibrium for Trinidad district populations of Ralstonia solanacearum, on polymorphic loci (15 loci considered, unless indicated)
Populations n a g obs a PrCLb P c I a d r d d P c
  1. a n: number of isolates; gobs: number of genotypes observed.

  2. bProportion of compatible locus pairs.

  3. c P-values were estimated from 1000 randomizations. NS, not significant, *significant, **highly significant.

  4. d I a and rd are observed Index of Association (Burt et al., 1996).

  5. eAnalysis done on 11 polymorphic loci.

  6. fAnalyses on Gasparillo were done on the eight polymorphic loci.


Gene diversity and genetic structure related to phylotype and sequevar

In these analyses, two populations were considered: phylotype I (n = 21) and phylotype II (n = 34), and subpopulations sampled in the four regions. The following analyses focus on the South and South-East subpopulations only, because sample sizes were too small in the North and no phylotype I strain was isolated from the Central region.

Differentiation across phylotypes I and II

Phylotype I and II populations were significantly differentiated from each other in the whole collection (θ = 0·056**) as well as in the South (θ = 0·147***). Linkage disequilibrium was estimated by (i) proportion of phylogenetically compatible pairs of loci (PrCL), (ii) index of association (Ia and rd), and (iii) standardized index of association inline image. The null hypothesis of random mating was rejected by all estimators for the South-East phylotype II population, whereas it was accepted by all estimators for the South phylotype II population (Table 5). Within phylotype I, index of association could not reject the random mating hypothesis, whereas PrCL did. Within phylotype II (total and Central population), the index of association rejected the random mating hypothesis whereas PrCl did not.

Table 5.   Estimates of linkage disequilibrium for Ralstonia solanacearum populations of different phylotypes, on polymorphic loci (15 loci considered, unless indicated)
Phylotype (population)Region (subpopulation) n a g obs a PrCLb P c I a d r d d P c I A s e P MC e P PAR e
  1. a n: number of isolates; gobs: number of genotypes observed.

  2. bProportion of compatible locus pairs.

  3. c P-values were estimated from 1000 randomizations. NS: not significant, *significant, **highly significant, ***very highly significant.

  4. d I a and rd are observed Index of Association (Burt et al., 1996).

  5. eStandardized Ia calculated according to Haubold et al. (1998). PMC and PPAR are P-values returned by lian 3·5 (Haubold & Hudson, 2000) after Monte Carlo simulations and the parametric method respectively, on 1000 resamplings.

Differentiation related to the sequevar within phylotype II

Sequevar 7 (n = 14) and Sequevar 35 (n = 10) strains were significantly differentiated from each other (θ = 0·077*, 1000 randomizations). For both sequevars, the index of association test led to rejection of the null hypothesis of random mating, whereas the proportion of phylogenetically compatible loci test could not reject this hypothesis.

Aggressiveness on tomato

Wilting of seedlings was observed from 4 to 14 days after inoculation. Significant differences (P < 0·05) were observed among strains and varieties in wilting tomato seedlings (Table 6), whereas strain × variety interaction was not significant (P > 0·05). Separation of the two phylotypes also showed that aggressiveness of phylotype II isolates (mean = 37·2%) was twice that of phylotype I isolates (mean = 18·4%). Considering the average wilting rates induced by regional subpopulations, the phylotype I isolates from South (mean = 15·0%) had lower aggressiveness than phylotype I isolates from North (mean = 32·3%) as well as the phylotype II isolates from North (mean = 30·9%), South (mean = 38·8%) and South-East (mean = 41·1%). Specifically, the South phylotype I strains were weakly to not virulent on the varieties Calypso, Heatmaster and Hybrid 61.

Table 6.   Pathogenicity of Ralstonia solanacearum strains on different tomato varieties
Strains (phylotype)Area of originWilting on tomato varietya (%)Overall mean
AkashCalypsoGempackGemprideHeatmasterHybrid 61
  1. aMean percentage of wilted plants (n = 12/block) 14 days after inoculation.

RF8 (II)South54·25029·262·541·729·244·4
RF11 (II)South54·25054·283·3505056·9
RF12 (II)South79·254·279·258·354·241·761·1
RF17 (II)South 4·2029·2 4·229·2 4·211·8
RF19 (II)South2537·520·8025 8·319·4
RF23 (II)North33·316·733·345·862·520·835·4
RF24 (II)North29·229·22520·82529·226·4
RF25 (I)North29·2 8·329·245·829·233·329·2
RF26 (I)North45·812·541·75033·329·235·4
RF29 (II)South-East83·337·570·883·358·35063·9
RF35 (II)South-East29·241·741·729·233·3 8·330·6
RF37 (II)South-East41·720·850252516·729·9
RF43 (I)South37·5012·537·512·520·820·1
RF44 (I)South54·2016·712·5 4·2 4·215·3
RF49 (I)South16·7 4·229·2 4·2 4·2 4·210·4
RF51 (I)South16·70 8·3 4·20 4·2 5·6
RF56 (I)South20·812·520·858·325 4·223·6
Mean variety 38·522·134·836·830·121·1 
lsd (strains) – 22·7        
lsd (variety) – 8·8        
Mean phylotype I 31·9 4·921·527·813·211·118·4
Mean phylotype II 4231·44241·739·426·537·2


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

Although R. solanacearum has been reported from as early as the last century as a major pathogen of banana crops in Trinidad, the situation in vegetable crops has not been investigated in detail. This study is the first formal report of the populations of R. solanacearum infecting tomato crops in this island. The results showed that these populations were composed of ∼36% phylotype I and ∼65% phylotype II. Phylotype II strains belonged to mainly sequevar 7 or 35. The presence of sequevar 35, which was previously a subgroup within sequevar 5 (P. Prior, unpublished data) in Trinidad was not surprising as this sequevar has already been reported on solanaceous crops in several neighbouring countries, including the French West Indies (Wicker et al., 2007), as well as Peru, Brazil (Fegan & Prior, 2005) and the USA (Ji et al., 2007). It is therefore likely that this sequevar is probably native to Trinidad. However, the identification of sequevar 7 isolates in Trinidad was a major finding, as this sequevar, which is known to cause bacterial wilt in tomato, has never been reported outside of the south-eastern USA, to the authors’ present knowledge. Ji et al. (2007) speculated that strains from this sequevar probably originated in North America. It is possible that these strains may have been introduced to Trinidad because tomato seeds used by local farmers are commonly supplied by North American seed companies. However, with the widespread distribution in Trinidad, having been found in three regions, the possibility of this sequevar being native to Trinidad and therefore in areas outside North America cannot be ruled out. Gene genealogies positioned Trinidad sequevar 7 strains (clade 3) as directly derived from the sequevar 7-clade 3 common ancestor (Wicker et al., 2011), suggesting that sequevar 7 originated in the Caribbean rather than from North America. Recent isolations of sequevar 7 in Salvador (E. Wicker, unpublished data) may reinforce this hypothesis.

The phylotype I isolates were found to be either sequevar 14 or 18, which were both also reported on aubergine in Guatemala (Sanchez Perez et al., 2008) and French Guiana (Milling et al., 2009). These concordances may be explained by introductions from common Asian sources or intra-regional movement in this part of the world.

Hence, the tomato populations in Trinidad were phylogenetically relatively close to those of other countries in the region, although phylotype I/II proportions were different: 27/73 in Martinique (Wicker et al., 2007), 61/39 in Guadeloupe (Sauboua, 2007), 45/55 in Guatemala (Sanchez Perez et al., 2008). However, they displayed two interesting specificities. First was the absence of phylotype II/sequevar 1 (Race 3/Biovar 2), whereas neighbouring Venezuela was reported to harbour solanaceous-infecting R. solanacearum populations composed of 82% bv2 (phylotype II/sequevar 1), 12% bv3 (phylotype I) and 6% bv1 (phylotype II) (Garcia et al., 1999b). These differences may be due to the absence of commercial potato production in Trinidad, and to its lowland tropical climate which may have prevented the establishment of accidentally introduced r3/bv2 strains in tomato crops. Secondly, no ‘emerging’ phylotype II/sequevar 4-NPB strains were identified in tomato fields, whereas one isolate of this group was found in 2006 in an anthurium farm in the Northern range of Trinidad (Wicker et al., 2007). This group is highly pathogenic in solanaceous crops, and is capable of spreading rapidly throughout a territory, as was observed in Martinique (Wicker et al., 2009b). Surveys are currently being undertaken in the Northern range to prevent any extension of the ‘emerging’ hot spot to main cropping areas. Additional surveys will also be needed to monitor the possible evolution of R. solanacearum populations within solanaceous cropping areas.

The broad phylogenetic diversity observed within the tomato-infecting strains of R. solanacearum in Trinidad raises several questions regarding management of bacterial wilt in the island, particularly concerning the use of resistant or tolerant varieties, in tomato crops as well as in other solanaceous vegetable crops (pepper, aubergine, potato). Aggressiveness tests on tomato clearly showed that Southern phylotype I strains were less aggressive than Northern phylotype I, but also phylotype II strains from different regions. From these results, it can be proposed that varieties Calypso, Heatmaster and Hybrid61 may be suitable to cropping fields in the South with phylotype 1 strains, due to their relatively high degree of resistance. However, alternative varieties may be needed for fields containing both phylotypes. Determination of the precise virulence/aggressiveness spectrum of the different populations of R. solanacearum on reference tomato resistance sources, such as the core worldwide collection of Lebeau et al. (2011), would be needed to aid in the selection of varieties for growing in different regions of the island.

Many studies have been done using rep-PCR to determine group clustering within numerous bacterial taxa (Roumagnac et al., 2007), including R. solanacearum (Smith et al., 1995; Frey et al., 1996; Horita & Tsuchiya, 2000; Dookun et al., 2001; Robertson et al., 2001). By contrast, population genetics estimators were rarely applied to DNA fingerprinting data (AFLP, RAPD, rep-PCR) on bacteria, mainly because bacterial population geneticists and molecular epidemiologists shifted early on to inter-laboratory portable sequence-based methods (MLST in particular), Multilocus variable number of tandem repeats (VNTRs) analysis (MLVA), or SNPs. Yet, population genetic analysis in bacteria using haploid multilocus data sets is completely justified (Haubold et al., 1998). Rep-, BOX- and ERIC-sequences were demonstrated to be widely distributed within bacterial genomes (Pukall, 2006), and their distribution in Xanthomonas and Pseudomonas strains was found to be a reflection of their genomic structure (Louws et al., 1994). Recombination in bacteria was first demonstrated by linkage disequilibrium assessments on multilocus enzyme electrophoresis (MLEE) patterns in the seminal paper of Maynard Smith et al. (1993), dealing with the medical bacteria Neisseria gonorrhae (panmictic) and N. meningitidis (epidemic structure), Rhizobium meliloti (local scale recombination), and Salmonella spp. (clonal structure), and has been reported to be widespread in bacteria (see for example Achtman, 2008 or Didelot & Maiden, 2010). Rep-PCR data were previously used to infer population genetics of the hazelnut pathogen Pseudomonas avellanae (Scortichini et al., 2006), or to compare amounts and distribution of fluorescent pseudomonads genetic diversity in suppressive and conducive soils (Ramette et al., 2006).

This study is the first to assess genetic structure of R. solanacearum populations at the regional scale, addressing questions related to spatial partition of genetic diversity, and genetic structure related to phylogenetic status. Due to low sample sizes at the field level, the partition of genetic diversity among and within fields was not considered. The results revealed that genetic diversity in Trinidad of R. solanacearum was relatively high, compared with other plant pathogenic bacteria such as Pseudomonas avellanae (Scortichini et al., 2006) or Xylella fastidiosa (Colleta-Filho & Machado, 2002). Populations were not regionally structured, as indicated by the absence of isolation by distance, suggesting that regular gene flow occurred between the regions sampled, maybe by irrigation networks, or by movement of people, equipment and plant materials among farms which can facilitate spread of strains within the island.

This study also points out the very specific features of the R. solanacearum population of Gasparillo (Central). This population, comprising phylotype IIA/sequevar 7 isolates only, displayed a relatively low genetic diversity, and highly significant differentiation from all the other locations. It may be that these strains were introduced from abroad, or that they constitute an ecological group adapted to specific soils. The soil of the tomato growing region of Gasparillo has a unique calcareous siltstone lithology as opposed to alluvium, shale or sand in other major tomato growing areas in Trinidad. This soil is fertile due to the presence of high levels of the potassium rich mineral glauconite and the region has had a long history of vegetable cultivation, with tomato being one of the main crops. Population structures were mostly clonal, but estimators revealed that the South population might have experienced recombination events.

This study also gives new insights on genetic structure of R. solanacearum populations related to phylotype. Phylotype I and phylotype II isolates could not be separated by cluster analysis (data not shown), as was already observed on Korean isolates with AFLP markers (Jeong et al., 2007). Moreover, phylotype I and II populations in Trinidad displayed similar gene and genotype diversity indices. However, both groups were significantly differentiated at the genotypic level, and differed by their level of clonality. Phylotype I total population was nearly-recombining, whereas phylotype II was more clonal. At the same time, the South populations of phylotype I and II were both recombining and nearly-recombining, whereas the phylotype II South-East population was strongly clonal. From these analyses, it appears that the South population displays a higher gene diversity and nearly-recombining genetic structure. It is difficult to explain the differences between the South and South-East populations because both areas have similar small-scale tomato production systems generally involving rotation with other vegetable crops. Other factors such as soil physical, chemical and biological characteristics must be investigated for possible influences on population genetic structure of the pathogen. It was also shown that sequevar 7 isolates formed a significantly differentiated group within the Trinidad phylotype II population. Ecological and phenotypical specificities of this group (in terms of host range and virulence/aggressiveness features) remain to be elucidated.

The conjunction of high gene diversity and clonal structure suggests that R. solanacearum genetic structure may be in concordance with the epidemic model proposed by Maynard Smith et al. (1993), where occasional outcrossing provides genetic variability, but multiple asexual cycles lead to linkage disequilibria and a tendency to clonality. Tomato-infecting isolates, according to this model, may represent only a fraction of the R. solanacearum population prevalent in a field. The report of recombination events in natural populations is new for R. solanacearum, although there is evidence that this bacterium is capable of entering into a competent state in planta (Bertolla et al., 1999) and is subject to horizontal gene transfers (HGT), both in vitro (Coupat et al., 2008; Guidot et al., 2009) and in competition experiments (Coupat-Goutaland et al., 2011). Multilocus sequence analysis revealed recently that phylotypes differ greatly in recombination levels, with phylotype I being highly recombinogenic (Wicker et al., 2011). Further investigations are needed to confirm the level of recombination in the populations of R. solanacearum at the local field and regional scales by using hierarchical designs and other more powerful approaches, such as MLST (Pérez-Losada et al., 2006) and VNTRs (Vergnaud & Pourcel, 2006) or even metagenomic sequencing approaches.

This study provides the first comprehensive information on populations of R. solanacearum affecting tomatoes in a southern Caribbean country, showing that great variability exists between neighbouring islands (Wicker et al., 2009b). It is now very important to conduct region-wide studies throughout the Caribbean and Central America to guide the development of regionally adapted control strategies, including bacterial wilt resistance breeding and suppressive rotational crops.


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

The authors would like to express thanks to Dr Ralph Phelps and Ms Adanna Piggott for their assistance provided during sample collection and pathogenicity testing of isolates. This work was partially funded by The University of the West Indies and European Regional Development Funds (FEDER) of the European Union, Région Martinique (Programme ‘Gestion intégrée des Bioagresseurs’), and Région Réunion (Programme ‘Lutte génétique contre les maladies émergentes chez les solanées maraîchères’ (GENETOM).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Achtman M, 2008. Evolution, population structure, and phylogeography of genetically monomorphic bacterial pathogens. Annual Review of Microbiology 62, 5370.
  • Agapow P-M, Burt A, 2001. Indices of multilocus linkage disequilibrium. Molecular Ecology Notes 1, 1012.
  • Bennett RS, Milgroom MG, Bergstrom GC, 2005. Population structure of seedborne Phaeosphaeria nodorum on New York wheat. Phytopathology 95, 3005.
  • Bertolla F, Frostegard A, Brito B, Nesme X, Simonet P, 1999. During infection of its host, the plant pathogen Ralstonia solanacearum naturally develops a state of competence and exchanges genetic material. Molecular Plant–Microbe Interactions 12, 46772.
  • Bonnet E, Van De Peer Y, 2002. zt: a software tool for simple and partial Mantel tests. Journal of Statistical Software (online) 7, 112. [].
  • Boucher CA, Barberis PA, Trigalet AP, Demery DA, 1985. Transposon mutagenesis of Pseudomonas solanacearum: isolation of Tn5-induced avirulent mutants. Journal of General Microbiology 131, 244957.
  • Briant AK, 1932. Tomato diseases in Trinidad. Tropical Agriculture 9, 6371.
  • Burt A, Carter DA, Koenig GL, White TJ, Taylor JW, 1996. Molecular markers reveal cryptic sex in the human pathogen Coccioides immitis. Proceedings of the National Academy of Sciences, USA 93, 7703.
  • Chong KX, Yang RC, Yeh FC, 1994. Nucleotide divergence between populations of trembling aspen (Populus tremuloides) estimated with RAPDs. Current Genetics 26, 3746.
  • Colleta-Filho HD, Machado MA, 2002. Evaluation of the genetic structure of Xylella fastidiosa populations from different Citrus sinensis varieties. Applied and Environmental Microbiology 68, 37316.
  • Coupat B, Chaumeille-Dole F, Fall S et al. , 2008. Natural transformation in the Ralstonia solanacearum species complex: number and size of DNA that can be transferred. FEMS Microbiology Ecology 66, 1424.
  • Coupat-Goutaland B, Bernillon D, Guidot A, Prior P, Nesme X, Bertolla F, 2011. Ralstonia solanacearum virulence increased following large interstrain gene transfers by natural transformation. Molecular Plant–Microbe Interactions 24, 497505.
  • Didelot X, Maiden MCJ, 2010. Impact of recombination on bacterial evolution. Trends in Microbiology 18, 31522.
  • Dookun A, Saumtally S, Seal S, 2001. Genetic diversity in Ralstonia solanacearum strains from Mauritius using restriction fragment length polymorphisms. Phytopathologische Zeitschrift 149, 515.
  • Fegan M, 2005. Bacterial wilt diseases of banana: evolution and ecology. In: Allen C, Prior P, Hayward AC, eds. Bacterial Wilt Disease and the Ralstonia solanacearum Species Complex. St Paul, MN: APS Press, 37986.
  • Fegan M, Prior P, 2005. How complex is the ‘Ralstonia solanacearum species complex’. In: Allen C, Prior P, Hayward AC, eds. Bacterial Wilt Disease and the Ralstonia solanacearum Species Complex. St Paul, MN: APS Press, 44962.
  • Fegan M, Prior P, 2006. Diverse members of the Ralstonia solanacearum species complex cause bacterial wilts of bananas. Australasian Plant Pathology 35, 93101.
  • Fegan M, Holloway G, Hayward AC, Timmis J, 1998. Development of a diagnostic test based on the polymerase chain reaction (PCR) to identify strains of R. solanacearum exhibiting the biovar 2 genotype. In: Prior P, Allen C, Elphinstone J, eds. Bacterial Wilt Disease – Molecular and Ecological Aspects. Paris: Springer-Verlag/INRA Editions, 3443.
  • Frey P, Smith JJ, Albar L et al. , 1996. Bacteriocin typing of Burkholderia solanacearum race 1 in the French West Indies and correlation with genomic variation of the pathogen. Applied and Environmental Microbiology 62, 4739.
  • Garcia R, Garcia A, Delgado L, 1999a. Bacterial disease of tomato caused by the biovar 2 A of Ralstonia solanacearum in several localities in Merida State – Venezuela. Revista Forestal Venezolana 43, 1839. [Spanish].
  • Garcia R, Garcia A, Delgado L, 1999b. Distribution, incidence and variability of Ralstonia solanacearum, causal agent of bacterial wilt of potato, in Merida state, Venezuela. Bioagro 11, 1223. [Spanish]
  • Gillings MR, Fahy P, 1994. Genomic fingerprinting: towards a unified view of the Pseudomonas solanacearum species complex. In: Hayward AC, Hartman GL, eds. Bacterial Wilt: the Disease and its Causative Agent, Pseudomonas solanacearum. Wallingford: CAB International, 95112.
  • Guidot A, Coupat B, Fall S, Prior P, Bertolla F, 2009. Horizontal gene transfer between Ralstonia solanacearum strains detected by comparative genomic hybridization on microarrays. ISME Journal 3, 54962.
  • Haubold B, Hudson RR, 2000. lian 3˙0: detecting linkage disequilibrium in multilocus data. Bioinformatics 16, 8478.
  • Haubold B, Travisano M, Rainey PB, Hudson RR, 1998. Detecting linkage disequilibrium in bacterial populations. Genetics 150, 13418.
  • Hayward AC, 1994. The hosts of Pseudomonas solanacearum. In: Hayward AC, Hartman GL, eds. Bacterial Wilt: the Disease and its Causative Agent, Pseudomonas solanacearum. Wallingford: CAB International, 924.
  • Horita M, Tsuchiya K, 2000. Genetic diversity of Japanese strains of Ralstonia solanacearum. Phytopathology 91, 399407.
  • Jaunet TX, Wang JF, 1999. Variation in genotype and aggressiveness of Ralstonia solanacearum race 1 isolated from tomato in Taiwan. Phytopathology 89, 3207.
  • Jeong Y, Kim J, Kang Y, Lee S, Hwang I, 2007. Genetic diversity and distribution of Korean isolates of Ralstonia solanacearum. Plant Disease 91, 127787.
  • Ji PS, Allen C, Sanchez PA et al. , 2007. New diversity of Ralstonia solanacearum strains associated with vegetable and ornamental crops in Florida. Plant Disease 91, 195203.
  • Jukes TH, Cantor CR, 1969. Evolution of protein molecules. In: Munro HN, ed. Mammalian Protein Metabolism. New York, NY: Academic Press, 21132.
  • Kelman A, 1954. The relationship of pathogenicity in Pseudomonas solanacearum to colony appearance on tetrazolium medium. Phytopathology 44, 6935.
  • Lebeau A, Daunay MC, Frary A et al. , 2011. Bacterial wilt resistance in tomato, pepper, and eggplant: genetic resources respond to diverse strains in the Ralstonia solanacearum species complex. Phytopathology 101, 15465.
  • Louws FJ, Fulbright DW, Stephens CT, de Bruijn FJ, 1994. Specific genomic fingerprints of phytopathogenic Xanthomonas and Pseudomonas pathovars and strains generated with repetitive sequences and PCR. Applied and Environmental Microbiology 60, 228695.
  • Ludwig W, Strunk O, Westram R et al. , 2004. ARB: a software environment for sequence data. Nucleic Acids Research 32, 136371.
  • Mahbou Somo Toukam G, Cellier G, Wicker E et al. , 2009. Broad diversity of Ralstonia solanacearum strains in Cameroon. Plant Disease 93, 112330.
  • Mantel N, 1967. The detection of disease clustering and a generalized regression approach. Cancer Research 27, 20920.
  • Martin DP, Lemey P, Lott M, Moulton V, Posada D, Lefeuvre P, 2010. RDP3: a flexible and fast computer program for analyzing recombination. Bioinformatics 26, 24623.
  • Maynard Smith J, Smith NH, O’rourke M, Spratt BG, 1993. How clonal are bacteria? Proceedings of the National Academy of Sciences, USA 90, 43848.
  • Milling A, Meng FH, Denny TP, Allen C, 2009. Interactions with hosts at cool temperatures, not cold tolerance, explain the unique epidemiology of Ralstonia solanacearum race 3 biovar 2. Phytopathology 99, 112734.
  • Nei M, 1973. Analysis of gene diversity in subdivided populations. Proceedings of the National Academy of Sciences, USA 70, 33213.
  • Opina N, Tavner F, Hollway G et al. , 1997. A novel method for development of species and strain-specific DNA probes and PCR primers for identifying Burkholderia solanacearum (formerly Pseudomonas solanacearum). Asia Pacific Journal of Molecular Biology and Biotechnology 5, 1930.
  • Pérez-Losada M, Browne EB, Madsen A, Wirth T, Viscidi RP, Crandall KA, 2006. Population genetics of microbial pathogens estimated from multilocus sequence typing (MLST) data. Infection, Genetics and Evolution 6, 97112.
  • Poussier S, Prior P, Luisetti J, Hayward AC, Fegan M, 2000. Partial sequencing of the hrpB and endoglucanase genes confirms and expands the known diversity within the Ralstonia solanacearum species complex. Systematic and Applied Microbiology 23, 47986.
  • Prior P, Fegan M, 2005a. Diversity and molecular detection of Ralstonia solanacearum race 2 strains by multiplex PCR. In: Allen C, Prior P, Hayward AC, eds. Bacterial Wilt Disease and the Ralstonia solanacearum Species Complex. St Paul, MN: APS Press, 40514.
  • Prior P, Fegan M, 2005b. Recent development in the phylogeny and classification of Ralstonia solanacearum. In: Momol T, Jones JB, eds. Proceedings of the First International Symposium on Tomato Diseases. Acta Horticulturae , 695, 12736.
  • Prior P, Steva H, 1990. Characteristics of strains of Pseudomonas solanacearum from the French West Indies. Plant Disease 74, 137.
  • Pukall R, 2006. DNA fingerprinting techniques applied to the identification, taxonomy and community analysis of prokaryotes. In: Stackebrandt E, ed. Molecular Identification, Systematics, and Population Structure of Prokaryotes. Berlin: Springer-Verlag, 5182.
  • Ramette A, Moënne-Loccoz Y, Défago G, 2006. Genetic diversity and biocontrol potential of fluorescent pseudomonads producing phloroglucinols and hydrogen cyanide from Swiss soils naturally suppressive or conducive to Thielaviopsis basicola-mediated black root rot of tobacco. FEMS Microbiology Ecology 55, 36981.
  • Robertson AE, Fortnum BA, Wood TC, Kluepfel DA, 2001. Diversity of Ralstonia solanacearum in the Southeastern United States. Contributions to Tobacco Research 19, 32331.
  • Rodríguez F, Oliver JL, Marín A, Medina JR, 1990. The general stochastic model of nucleotide substitution. Journal of Theoretical Biology 142, 485501.
  • Rorer JB, 1911. A bacterial disease of bananas and plantains. Phytopathology 1, 2752.
  • Roumagnac P, Gagnevin L, Pruvost O, Achtman M, 2007. Insights into structure and evolution of bacterial species that are revealed by molecular methods. In: Tibayrenc M, ed. Encyclopedia of Infectious Diseases: Modern Methodologies. Hoboken, NJ: John Wiley and Sons, Inc., 47392.
  • Rousset F, 1997. Genetic differentation and estimation of gene flow from F-statistics under isolation by distance. Genetics 145, 121928.
  • Saitou N, Nei M, 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology Evolution 4, 40625.
  • Sambrook J, Fritsch EF, Maniatis T, 1989. Molecular Cloning: a Laboratory Manual. New York, NY: Cold Spring Harbor Laboratory Press.
  • Sanchez Perez A, Mejia L, Fegan M, Allen C, 2008. Diversity and distribution of Ralstonia solanacearum strains in Guatemala and rare occurence of tomato fruit infection. Plant Pathology 57, 112.
  • Sauboua P, 2007. Identification des souches de Ralstonia solanacearumprésentes en Guadeloupe et évaluation d’hybrides INRA de tomate pour leur résistance aux maladies à begomovirus TYLCV et PYMV, et au flétrissement bactérien. Bordeaux: ENITA Bordeaux, MSc dissertation.
  • Scortichini M, Natalini E, Marchesi U, 2006. Evidence for separate origins of the two Pseudomonas avellanae lineages. Plant Pathology 55, 4517.
  • Smith JJ, Offord LC, Holderness M, Saddler GS, 1995. Genetic diversity of Burkholderia solanacearum (synonym Pseudomonas solanacearum) race 3 in Kenya. Applied and Environmental Microbiology 61, 42638.
  • Stefanova M, 1998. Current situation of bacterial wilt (Ralstonia solanacearum Smith) in Cuba. In: Prior P, Allen C, Elphinstone J, eds. Bacterial Wilt Disease – Molecular and Ecological Aspects. Berlin: Springer, 3648.
  • Vergnaud G, Pourcel C, 2006. Multiple locus VNTR (variable number of tandem repeat) analysis. In: Stackebrandt E, ed. Molecular Identification, Systematics, and Population Structure of Prokaryotes. Berlin: Springer-Verlag, 83104.
  • Versalovic J, Schneider M, De Bruijn FJ, Lupski JR, 1994. Genomic fingerprinting of bacteria using repetitive sequence based PCR (REP-PCR). Methods in Molecular Cell Biology 5, 2540.
  • Weir BS, 1996. Genetic Data Analysis II. Sunderland, MA: Sinauer.
  • Weir BS, Cockerham CC, 1984. Estimating F-statistics for the analysis of population structure. Evolution 38, 135870.
  • Wicker E, Grassart L, Coranson-Beaudu R et al. , 2007. Ralstonia solanacearum strains from Martinique (French West Indies) exhibiting a new pathogenic potential. Applied and Environmental Microbiology 73, 6790801.
  • Wicker E, Coranson-Beaudu R, Cadasse S, William MA, 2009a. Emerging strains of Ralstonia solanacearum in the French West Indies raise new challenges to tomato breeders. Acta Horticulturae 808, 27986.
  • Wicker E, Grassart L, Coranson-Beaudu R, Mian D, Prior P, 2009b. Epidemiological evidence for the emergence of a new pathogenic variant of Ralstonia solanacearum in Martinique (French West Indies). Plant Pathology 58, 85361.
  • Wicker E, Lefeuvre P, De Cambiaire JC, Lemaire C, Poussier S, Prior P, 2011. Contrasting recombination patterns and demographic histories of the plant pathogen Ralstonia solanacearum inferred from MLSA. The ISME Journal. doi: 10.1038/ismej.2011.160.
  • Winstead NN, Kelman A, 1952. Inoculation techniques for evaluating resistance to Pseudomonas solanacearum. Phytopathology 42, 62834.

Supporting Information

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

Figure S1. Phylogenetic neighbour-joining (NJ) tree based on egl sequences. Trinidad strains isolated from tomato are written in bold.

Figure S2. Maximum parsimony tree constructed by using the software dnapars.

PPA_2572_sm_FigS1.pdf202KSupporting info item
PPA_2572_sm_FigS2.pdf191KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.