Abstract
Current identification methods for the soft rot erwinias are both imprecise and time-consuming. We have used the 16S-23S rRNA intergenic transcribed spacer (ITS) to aid in their identification. Analysis by ITS-PCR and ITS-restriction fragment length polymorphism was found to be a simple, precise, and rapid method compared to current molecular and phenotypic techniques. The ITS was amplified from Erwinia and other genera using universal PCR primers. After PCR, the banding patterns generated allowed the soft rot erwinias to be differentiated from all other Erwinia and non-Erwinia species and placed into one of three groups (I to III). Group I comprised all Erwinia carotovora subsp. atroseptica and subsp. betavasculorum isolates. Group II comprised all E. carotovora subsp. carotovora, subsp. odorifera, and subsp. wasabiae and E. cacticida isolates, and group III comprised all E. chrysanthemi isolates. To increase the level of discrimination further, the ITS-PCR products were digested with one of two restriction enzymes. Digestion with CfoI identified E. carotovora subsp. atroseptica and subsp. betavasculorum (group I) and E. chrysanthemi (group III) isolates, while digestion with RsaI identified E. carotovora subsp. wasabiae, subsp. carotovora, and subsp. odorifera/carotovora and E. cacticida isolates (group II). In the latter case, it was necessary to distinguish E. carotovora subsp. odorifera and subsp. carotovora using the α-methyl glucoside test. Sixty suspected soft rot erwinia isolates from Australia were identified as E. carotovora subsp. atroseptica, E. chrysanthemi, E. carotovora subsp. carotovora, and non-soft rot species. Ten “atypical” E. carotovora subsp. atroseptica isolates were identified as E. carotovora subsp. atroseptica, subsp. carotovora, and subsp. betavasculorum and non-soft rot species, and two “atypical” E. carotovora subsp. carotovora isolates were identified as E. carotovora subsp. carotovora and subsp. atroseptica.
The genus Erwinia comprises plant pathogens belonging to the family Enterobacteriaceae. One group within the genus, the soft rot erwinias, causes soft rot diseases of many plant species worldwide (28). The most important of the soft rot erwinias commercially are E. chrysanthemi, E. carotovora subsp. carotovora, and E. carotovora subsp. atroseptica, which cause diseases of potato and other commercially important crops. However, the other subspecies, E. carotovora subsp. betavasculorum, E. carotovora subsp. odorifera, and E. carotovora subsp. wasabiae, are also important pathogens (10, 11, 19). Since the host range of the soft rot erwinias has not been elucidated fully, it is important to determine which of the pathogens is responsible for a disease before disease control can be effective. However, in current identification methods, which are based mainly on biochemical and phenotypic characteristics, rapid and accurate identification is not always possible. As a consequence, isolates may be misidentified or, when results do not fit those expected, may be labeled as “atypical.”
Currently, the major soft rot erwinias are identified by their growth and cavity formation on pectate-containing selective media, such as crystal violet pectate (CVP) (3), at differential temperatures, i.e., E. carotovora subsp. atroseptica grows at 27°C, E. carotovora subsp. carotovora grows at 27 and 33.5°C, and E. chrysanthemi grows at 27, 33.5, and 37°C. However, these identifications often remain tentative, and growth at differential temperatures is not always accurate, since some isolates grow at temperatures outside their expected range (27). Biochemical tests are currently the accepted standard for identification and taxonomy of the soft rot erwinias (7, 8, 9, 23, 35), but on a routine level they are very time-consuming (taking up to 14 days) and, when carried out by nonspecialist laboratories, do not always provide a definitive identification.
A number of other methods have been used to identify the soft rot erwinias although, as with biochemistry and growth on CVP, all have limitations. The high serological heterogeneity and cross-reactivity within and between subspecies, respectively, have limited the use of serology (7). Fatty acid profiling has been used to differentiate Erwinia species (24, 36) but has been of only limited success in identifying individual subspecies within E. carotovora (5, 29). The randomly amplified polymorphic DNA (RAPD)-PCR method lacks reproducibility (particularly between laboratories) and has not been tested on all E. carotovora subspecies (21, 26, 32). DNA-DNA hybridization is accurate but time-consuming and unsuitable for routine use, especially when large numbers of strains are involved (2, 10, 25, 34). PCR-restriction fragment length polymorphism (PCR-RFLP) analysis of a pectate lyase gene has been used but has been unsuccessful in identifying all E. carotovora subspecies (4, 15). Repetitive sequences and amplified fragment length polymorphisms have been used to fingerprint phytopathogens and is accurate, but due to the generation of 30 or more fragments for each isolate tested, both methods require computer analysis for identification, which is not available to many laboratories (20).
16S rRNA gene sequencing has been used to study the phylogenetic relationships between Erwinia species (14, 18). Although the taxonomy of the E. carotovora subspecies has been examined by this method and does allow identification of species and subspecies, sequencing for routine identification is impractical at the present time. In addition, this method approaches its limits of sensitivity below the species level (34). The multicopy 16S-23S intergenic transcribed spacer (ITS), which separates the rRNA genes, however, exhibits a greater sequence and length variation that can be exploited in a simple PCR-RFLP-based test and is suitable for differentiating below the species level (12, 13, 20, 22, 30). Jensen et al. (17) developed universal primers and conditions for amplifying the ITS from all prokaryotes, and these primers have been used subsequently to identify both human (13, 22) and plant (12) pathogens. In a limited number of cases this has been combined with RFLP analysis of the ITS product (12, 30). This study describes the use of ITS-PCR, in combination with ITS-RFLP, for the rapid and accurate identification and differentiation of the soft rot erwinias.
MATERIALS AND METHODS
Bacterial strains and media.
Bacterial strains used in this study are listed in Tables 1 and 2. Reference strains are species type strains obtained from the National Collection of Plant Pathogenic Bacteria (NCPPB; York, United Kingdom). In cases in which type strains were not available from the NCPPB, strains identified by the NCPPB as belonging to certain species and subspecies were used. In addition, a number of uncharacterized and “atypical” strains were investigated. Bacterial strains were stored in freezing medium at −80°C (1). All cultures used in the study were maintained on nutrient agar at 18°C (Oxoid, Basingstoke, United Kingdom). When required, Erwinia species and saprophytes were grown at 27°C, and other enterobacteria were grown at 37°C in Luria-Bertani broth (LB) for 18 h with shaking.
TABLE 1.
Soft rot erwinia strains together with ITS-PCR and ITS-RFLP patterns
| Bacterial straina | Host | Locationb | ITS-PCR ITS-RFLP patternc |
|---|---|---|---|
| E. carotovora subsp. | |||
| carotovora | |||
| SCRI 108 | Potato | Finland | ECCa-1 |
| SCRI 295 | Potato | Scotland | ECCa-1 |
| SCRI 112 (NCPPB 1746)v | Potato | Japan | ECCa-2 |
| SCRI 193 | Potato | U.S. | ECCa-2 |
| SCRI 120 (NCPPB 1231)v | Sunflower | Uganda | ECCa-3 |
| SCRI 133 | Potato | Israel | ECCa-3 |
| SCRI 134 | Cyclamen | Israel | ECCa-3 |
| SCRI 144 | Potato | Tasmania | ECCa-4 |
| SCRI 212, 234, and 244 | Potato | Scotland | ECCa-4 |
| SCRI 258 | Potato | Israel | ECCa-5 |
| SCRI 126, 270, 273, 275, 296, 350, 352, and 353 | Potato | United Kingdom | ECCa |
| SCRI 137 and 138 | Potato | Arizona (U.S.) | ECCa |
| SCRI 146, 148, 152, 154, 159, 160, 164, and 156 | Potato | Tasmania | ECCa |
| SCRI 171, 178, and 280 | Potato | Peru | ECCa |
| SCRI 205 | Sunflower | Mexico | ECCa |
| SCRI 256, 344, 317, and 343 | Soil or water | Scotland | ECCa |
| SCRI 318 | Rutabaga | Scotland | ECCa |
| SCRI 324 | Potato | Canada | ECCa |
| SCRI 327 and 329 | Potato | U.S. | ECCa |
| SCRI 174 | Potato | Peru | ECCb-3 |
| SCRI 167 | Water | Israel | ECCb-4 |
| atroseptica | |||
| SCRI 33 and 37 | Soil or water | Scotland | ECA |
| SCRI 47, 54, 59, 83, 93, 1035, 1039, and 1040 | Potato | Scotland | ECA |
| SCRI 135 | Potato | Arizona (U.S.) | ECA |
| Z141, Z144, Z148, and Z153 | Potato | Sweden | ECA |
| 581-1.1 and 1342-47 | Potato | Spain | ECA |
| 310 (88.33), 315 (86.14.11), and 494 | Potato | France | ECA |
| IPO 723, IPO 848, IPO 852, IPO 856, IPO 861, IPO 161, N88.30, and N88.46 | Potato | The Netherlands | ECA |
| betavasculorum | |||
| SCRI 479 (NCPPB 2795)vT, SCRI 908 (NCPPB 2792)v, SCRI 909 (NCPPB 2793)v, SCRI 910 (NCPPB 2794)v, and SCRI 911 (NCPPB 3075)v | Sugar beet | U.S. | ECB |
| odorifera | |||
| SCRI 482 (NCPPB 3840), SCRI 487, SCRI 914 (NCPPB 3842), and SCRI 916 (NCPPB 3844) | Chicory | France | ECOa |
| SCRI 912 (NCPPB 3839), SCRI 913 (NCPPB 3841), and SCRI 915 (NCPPB 3843) | Chicory | France | ECOb |
| wasabiae | |||
| SCRI 481 (NCPPB 3701), SCRI 488, SCRI 917 (NCPPB 3702), SCRI 918 (NCPPB 3703), and SCRI 919 (NCPPB 3704) | Horseradish | Japan | ECW |
| E. cacticida SCRI 484 (NCPPB 3849) and SCRI 923 (NCPPB 3848) | Giant cactus | U.S. | E. cacticida |
| E. chrysanthemi pv. | |||
| Unspecified | |||
| SCRI 418 | Potato | Peru | ECHa-2 |
| SCRI 4033 and SCRI 4064 (NCPPB 3090) (biovar 3)v | Rice | Japan | ECHa-2 |
| SCRI 4076 (NCPPB 1125)v∗ | Pineapple | Malaysia | ECHb-2 |
| SCRI 4000 (IPO 645) and SCRI 4004 (IPO 649) | Potato | The Netherlands | ECHc-2 |
| SCRI 417, 4037, and 4039 | Potato | Peru | ECHd-1 |
| SCRI 4044 (IPO 502) (biovar 7) | Potato | The Netherlands | ECHc-3 |
| dianthicola | |||
| SCRI 4073 (NCPPB 453) | Carnation | United Kingdom | ECHa-2 |
| SCRI 401 (NCPPB 426) | Carnation | Denmark | ECHd-3 |
| SCRI 409 (NCPPB 518)v∗ | Carnation | Denmark | ECHf-3 |
| zeae | |||
| SCRI 4072 (NCPPB 2547) (biovar 7)v∗ | Maize | India | ECHa-2 |
| SCRI 4071 (NCPPB 2541) (biovar 7)v∗ | Maize | U.S. | ECHd-1 |
| SCRI 4078 (NCPPB 1065)v∗ | Maize | Egypt | ECHd-1 |
| dieffenbachiae SCRI 4074 (NCPPB 1514)v∗ | Dieffenbachia | Germany | ECHa-2 |
T, type strain; v, authenticity of the culture was verified by the NCPPB; ∗, pathogenicity of the culture was confirmed by the NCPPB.
U.S., United States.
Superscripts: a to f, ITS-PCR patterns; 1 to 5, ITS-RFLP patterns. Abbreviations: ECA, E. carotovora subsp. atroseptica; ECB, E. carotovora subsp. betavasculorum; ECC, E. carotovora subsp. carotovora; ECO, E. carotovora subsp. odorifera; ECW, E. carotovora subsp. wasabiae; ECH, E. chrysanthemi.
TABLE 2.
Other soft rot and non-soft rot erwinia strainsa
| Organism | Subspecies, pathovars, and/or strain(s) |
|---|---|
| Erwinia rhapontici | SCRI 421 (NCPPB 139)v, SCRI 423 (NCPPB 1578)vT, and SCRI 472 |
| Erwinia uredovora | SCRI 432 (NCPPB 800)vT and SCRI 433 (NCPPB 802)v |
| Erwinia cypripedii | SCRI 440 (NCPPB 752) and SCRI 478 (NCPPB 2636)∗ |
| Erwinia quercina | SCRI 442 and 477 |
| Erwinia rubrifaciens | SCRI 445 (NCPPB 2020)vT and SCRI 446 (NCPPB 2021)v |
| Erwinia nigrifluens | SCRI 476 (NCPPB 564)vT |
| Erwinia amylovora | SCRI 444, SCRI 906 (NCPPB 595)vs, and SCRI 907 (NCPPB 686)v |
| Erwinia salicis | SCRI 474 (NCPPB 447)vT and SCRI 905 (NCPPB 2535)v |
| Erwinia persicina | SCRI 480 (NCPPB 3375) |
| Erwinia mallotivora | SCRI 490 (NCPPB 2851)vT∗ and SCRI 903 (NCPPB 2852)v∗ |
| Erwinia psidii | SCRI 492 (NCPPB 3555)v and SCRI 902 (NCPPB 3556)v |
| Erwinia tracheiphila | SCRI 493 (NCPPB 2452)vT∗ and SCRI 900 (NCPPB 2133) |
| Pantoea agglomerans | SCRI 435, SCRI 459, and SCRI 462 |
| Pantoea stewartii | SCRI 475 (NCPPB 449)v∗ |
| Pantoea ananatis | SCRI 485 (NCPPB 1846)vT∗ and SCRI 922 (NCPPB 544) |
| Enterobacter cancerogenus | SCRI 489 (NCPPB 2176)T and SCRI 920 (NCPPB 2177)v |
| Enterobacter dissolvens | SCRI 921 (NCPPB 1850)vT |
| Enterobacter nimipressuralis | SCRI 491 (NCPPB 2045)vT and SCRI 901 (NCPPB 440) |
| Xanthomonas albilineans | PDDCC 196 |
| Xanthomonas campestris | pv. campestris UPB 142, pv. phaseoli 095, pv. phaseoli var. fuscans 180, pv. manihotis UPB 025, pv. pruni ATCC 10016, pv. ricini UPB 075, and pv. vesicatoria UPB 127 |
| Pseudomonas syringae | pv. syringae SCRI 777 (NCPPB 1071), pv. morsprunorum SCRI 776 (NCPPB 1458), and pv. phaseolicola 882 race 2 |
| Agrobacterium radiobacter | subsp. tumefaciens SCRI 527 (NCPPB 2404) |
| Ralstonia solanacearum | SCRI 783 (NCPPB 2314) |
| Clavibacter xyli | subsp. xyli 1 J1, 5 BIB, and 7 IISN |
| Comamonas sp. | S102 and S160 |
| Janthinobacter sp. | S126 and S159 |
| Aeromonas sp. | S91 and S291 |
| Saprophytic bacteria | 1, 2, 3, 4, 6, 7, 8, 9, 10, 11, and 12 |
| Escherichia coli | 25 and 26 |
| Salmonella enterica serovar Enteritidis | |
| Proteus mirabilis | |
| Citrobacter freundii | 34 |
Superscripts are as defined in Table 1, footnote c, with the addition of the following: s, authenticity and virulence confirmed by the NCPPB.
Biochemical and phenotypic tests.
Biochemical tests—acid production from α-methyl glucoside, palatinose, sorbitol, melibiose, and lactose, the production of phosphatase and indole, reducing substances from sucrose, the utilization of citrate, and growth in 5% NaCl and on nutrient agar at 37°C—were done as described previously (8). Cavity formation on CVP medium at 27, 33.5, and 37°C was also assessed as described earlier (27).
Isolation of genomic DNA.
Bacterial genomic DNA (gDNA) was extracted and purified using a DNA Mini Kit as described by the manufacturer (Qiagen, Crawley, United Kingdom). Extracted DNA was electrophoresed through a 1.2% agarose gel in Tris-borate-EDTA (TBE) buffer and stained with ethidium bromide (0.5 μg ml−1) as described previously (31). DNA was stored at −20°C until required.
E. carotovora subsp. atroseptica-specific PCR.
E. carotovora subsp. atroseptica gDNA was amplified using the E. carotovora subsp. atroseptica-specific primers ECA1f and ECA2r (6) obtained from MWG Biotech (Milton Keynes, United Kingdom) using a modified protocol described earlier (16).
PCR amplification and restriction digestion of the ITS.
After the extraction of gDNA, the ITS was amplified using the primers G1 (5′-GAAGTCGTAACAAGG-3′) and L1 (5′-CAAGGCATCCACCGT-3′) as described by Jensen et al. (17). For each strain, 10 μl of the amplified product was digested with each of the restriction enzymes AluI, CfoI, HaeIII, HhaI, HpaII, MseI, MspI, MboI, RsaI, Sau3aI, TaqI, and ThaI, as described by the manufacturer (Life Technologies, Paisley, United Kingdom). Digested samples (10 μl) were electrophoresed through a 2% NuSieve GTG agarose gel (Flowgen, Ashby de la Zouch, United Kingdom) in TBE buffer for 1.5 h and stained as described above. Gel images were digitized and band sizes analyzed by Gel Compar software (Applied Maths, Kortrijk, Belgium). Fragment sizes were determined by comparison to a 1-kb Plus DNA molecular weight marker (Life Technologies, Paisley, United Kingdom).
RESULTS AND DISCUSSION
The main aim of this study was to develop a method, or to utilize an existing method, for the simple, rapid, and accurate identification and differentiation of the soft rot erwinias. This was accomplished by using a combination of ITS-PCR and ITS-RFLP.
After PCR amplification of the ITS (ITS-PCR) from all Erwinia species, other plant and/or animal pathogens, and plant and/or soil saprophytes, characteristic banding patterns were generated on NuSieve agarose gels (Fig. 1) derived from multicopy rRNA operons (20). The majority of bands were clearly visible each time that a PCR was carried out (primary bands). However, fainter bands did appear on occasion (secondary bands), but these proved less reliable for identification. Band sizes were thus recorded for primary products only (Table 3). Bands from a number of isolates were sequenced to reveal that each consisted of DNA from the ITS region with deletions and/or insertions determining the size of each band (data not shown). ITS-PCR generated unique patterns for all bacterial species tested (Fig. 1), and in most cases, these patterns were similar for isolates within a species. Patterns generated for human pathogens were similar but not identical to those described by Jensen et al. (17), a result that might have been due to the use of different strains (data not shown).
FIG. 1.
ITS-PCR amplification patterns of species belonging to Erwinia, Pantoea, Enterobacter, and other Enterobacteriaceae genera. Lane 1, E. carotovora subsp. atroseptica SCRI 1039; 2, E. carotovora subsp. betavasculorum SCRI 479; 3, ECCa, E. carotovora supsp. carotovora SCRI 244; 4, ECCb, E. carotovora subsp. carotovora SCRI 167; 5, ECOa, E. carotovora subsp. odorifera SCRI 914; 6, ECOb, E. carotovora subsp. odorifera SCRI 915; 7, E. carotovora subsp. wasabiae SCRI 481; 8, E. cacticida SCRI 484; 9, Enterobacter cancerogenus SCRI 489; 10 and 11, E. cypripedii SCRI 478 and SCRI 440; 12 and 13, E. rhapontici SCRI 421 and SCRI 472; 14, ECHa, E. chrysanthemi SCRI 418; 15, ECHb, E. chrysanthemi SCRI 4071; 16, ECHc, E. chrysanthemi SCRI 4004; 17, ECHd, E. chrysanthemi SCRI 4037; 18, ECHe, E. chrysanthemi SCRI 4044; 19, ECHf, E. chrysanthemi SCRI 409; 20 and 21, Pantoea ananatis SCRI 485 and SCRI 922; P. stewartii SCRI 475; 23 and 24, P. agglomerans SCRI 435 and SCRI 459; 25, Enterobacter nimipressuralis SCRI 491; 26, Enterobacter dissolvens SCRI 921; 27, Erwinia uredovora SCRI 433; 28, E. quercinia SCRI 442; 29, E. rubrifaciens SCRI 445; 30, E. nigrifluens SCRI 476; 31 and 32, E. amylovora SCRI 444 and SCRI 906; 33, E. tracheiphila SCRI 493; 34, E. mallotivora SCRI 490; 35, E. salicis SCRI 474; 36, E. psidii SCRI 492; 37, E. persicinus SCRI 480. M, 1-kb Plus ladder (Life Technologies).
TABLE 3.
Product sizes for different soft rot erwinias following PCR amplification of the 16S-23S ITS (ITS-PCR) and digestion of PCR products with either CfoI or RsaI (ITS-RFLP)
| Method and groupa | Primary fragments (bp) |
|---|---|
| ITS-PCR | |
| Group I | |
| ECA | 540, 575, 620, 740 |
| ECB | 540, 575, 620, 720 |
| Group II | |
| ECCa | 540, 575, 740 |
| ECCb | 540, 575, 600, 740 |
| ECOa | 540, 575, 740 |
| ECOb | 540, 575, 600, 740 |
| ECW | 535, 575, 720 |
| E. cacticida | 550, 580, 740 |
| Group III | |
| ECHa | 440, 590, 690 |
| ECHb | 445, 590, 720, 1170 |
| ECHc | 450, 590, 695 |
| ECHd | 450, 580, 690 |
| ECHe | 450, 575, 605, 670, 700 |
| ECHf | 450, 470, 490, 580, 595, 685, 710 |
| ITS-RFLP | |
| Group Ix | |
| ECA | 115, 153, 225, 345 |
| ECB | 115, 153, 305, 420 |
| Group IIy | |
| ECC1 | 180, 355 |
| ECC2 | 180, 210, 355 |
| ECC3 | 195, 355, 520 |
| ECC4 | 115, 180, 225, 355 |
| ECC5 | 115, 180, 225, 445 |
| ECO | 180, 210, 355 |
| ECW | 165, 200, 355 |
| E. cacticida | 180, 355, 565 |
| Group IIIx | |
| ECH1 | 115, 145, 255, 455 |
| ECH2 | 145, 185, 255, 455 |
| ECH3 | 155, 180, 255, 455 |
Superscripts are as defined in Table 1, footnote c, with the addition of the following: x, digested with CfoI; y, digested with RsaI.
Isolates of E. carotovora subspecies, E. chrysanthemi, and the closely related E. cacticida yielded unique banding patterns that clearly distinguished them from other Erwinia and non-Erwinia species tested (Fig. 1, lanes 1 to 8 and lanes 14 to 19). Primary products ranged from three to seven in number and from 440 to 1,170 bp in size. Within the soft rot erwinias, three PCR groups were distinguished based on differences in their banding patterns (groups I to III). Group I comprised E. carotovora subsp. atroseptica and subsp. betavasculorum and generated two patterns, each characteristic of a subspecies (Fig. 1, lanes 1 to 2). However, the difference between band sizes, i.e., a slight shift in the larger band, was deemed insufficient for a reliable identification. Group II comprised E. carotovora subsp. carotovora, subsp. odorifera, and subsp. wasabiae and E. cacticida and generated four different patterns (Fig. 1, lanes 3 to 8). Two of the patterns were present in both E. carotovora subsp. carotovora and E. carotovora subsp. odorifera, while the other two patterns were characteristic of E. carotovora subsp. wasabiae and E. cacticida, respectively. Again, small differences in band sizes were insufficient to distinguish reliably the members within group II. Groups I and II were clearly related based on the banding patterns generated. Group III comprised all E. chrysanthemi isolates and generated six different but related patterns (Fig. 1, lanes 14 to 19; Table 2), which differed from those of groups I and II (Fig. 1, lanes 1 to 8; Table 2). The variation within group III made identification difficult, especially since strains from other non-Erwinia species, e.g., Pantoea agglomerans (strain SCRI 459) and Enterobacter nimipressuralis (strain SCRI 491), gave similar patterns (Fig. 1, lanes 23 to 25).
To improve the level of discrimination between the soft rot erwinias further, ITS-PCR products were digested with either one or two of 12 four-base cutting restriction enzymes (ITS-RFLP). Of these enzymes, two (CfoI and RsaI), used individually, produced restriction patterns that allowed identification and differentiation of most species and subspecies within the soft rot groups (I to III) (Fig. 2). Double digests did not improve the level of discrimination and were not used further. However, whether CfoI or RsaI was used individually depended on the ITS-PCR group (I to III), i.e., CfoI was used only after identification of group I and III isolates, whereas RsaI was used only for group II isolates. This prevented misidentification; e.g., the patterns generated for E. carotovora subsp. carotovora, subsp. odorifera, and subsp. wasabiae (group II) using CfoI were identical to that of E. carotovora subsp. atroseptica (group I) (data not shown). After identification of groups I and III using ITS-PCR, digestion with CfoI clearly distinguished E. carotovora subsp. atroseptica from E. carotovora subsp. betavasculorum isolates (Fig. 2, lanes 1 to 2) and distinguished the E. chrysanthemi (group III) isolates from non-Erwinia species (Fig. 2, lanes 3 to 5). E. carotovora subsp. atroseptica and subsp. betavasculorum produced subspecies-specific patterns that were identical for all isolates tested. E. chrysanthemi isolates produced only three characteristic patterns (E. chrysanthemi1, E. chrysanthemi2, and E. chrysanthemi3) from the six ITS-PCR patterns generated. RsaI digestion of group II products produced characteristic patterns that could identify isolates of E. carotovora subsp. wasabiae, subsp. carotovora, subsp. odorifera, and subsp. carotovora and E. cacticida. However, it was not always possible to differentiate E. carotovora subsp. carotovora and subsp. odorifera, since some E. carotovora subsp. carotovora strains appear to be too closely related to E. carotovora subsp. odorifera at the ITS DNA level (Fig. 2, lanes 6 to 13). This was evident by the generation of identical banding patterns in many, but not all, cases using both ITS-PCR and ITS-RFLP (Fig. 1 and 2) and has been shown previously by DNA-DNA hybridization (10) and RFLP analysis (4, 15). It was thus necessary to differentiate E. carotovora subsp. odorifera and subsp. carotovora isolates using the α-methyl glucoside test: positive for E. carotovora subsp. odorifera and negative for E. carotovora subsp. carotovora (10), which could take a further 3 to 5 days. The E. carotovora subsp. carotovora isolates tested produced five different restriction patterns (E. carotovora subsp. carotovora1, carotovora2, carotovora3, carotovora4, and carotovora5), while E. carotovora subsp. wasabiae and E. cacticida isolates produced their own characteristic patterns. All E. carotovora subsp. odorifera isolates produced a single pattern that was identical to E. carotovora subsp. carotovora2.
FIG. 2.
ITS-RFLP amplification patterns of soft rot Erwinia species digested with CfoI (lanes 1 to 5) and RsaI (lanes 6 to 13). Lane 1, E. carotovora subsp. atroseptica SCRI 1039; 2, E. carotovora subsp. betavasculorum SCRI 479; 3, ECH1, E. chrysanthemi SCRI 417; 4, ECh2, E. chrysanthemi SCRI 4000; 5, ECH3, E. chrysanthemi SCRI 4044; 6, ECC1, E. carotovora subsp. carotovora SCRI 108; 7, ECC2, E. carotovora subsp. carotovora SCRI 193; 8, ECC3, E. carotovora subsp. carotovora SCRI 120; 9, ECC4, E. carotovora subsp. carotovora SCRI 212; 10, ECC5, E. carotovora subsp. carotovora SCRI 258; 11, E. carotovora subsp. odorifera SCRI 912; 12, E. carotovora subsp. wasabiae SCRI 481; 13, E. cacticida SCRI 484. M, 1-kb Plus ladder (Life Technologies).
Individual isolates within E. chrysanthemi and E. carotovora subsp. carotovora generated several different banding patterns with ITS-PCR and ITS-RFLP, respectively, showing a higher level of diversity within these than within other species or subspecies. This increased diversity has been shown previously at both the phenotypic and the genotypic levels (4, 7, 15, 21, 23, 26, 35). Although it is likely that new isolates could generate further patterns, a combination of ITS-PCR and ITS-RFLP would still allow the identification of these pathogens and their differentiation from other bacterial strains, including other soft rot erwinias.
To determine the effectiveness of ITS-PCR and ITS-RFLP in practice, 60 unidentified isolates from Australia (believed to be soft rot erwinias due to their growth and cavity formation on CVP) were tested. After ITS-PCR, all three soft rot groups were present within these isolates: group I (13 isolates), group II (44 isolates), group III (one isolate), and cavity-forming non-soft rot erwinias (2 isolates). After PCR, ITS-RFLP identified all group I isolates as E. carotovora subsp. atroseptica (13 isolates), group II isolates as E. carotovora subsp. carotovora (26 isolates), and E. carotovora subsp. carotovora/odorifera (18 isolates), and group III isolates as E. chrysanthemi (1 isolate). These isolates were also extensively tested using phenotypic, biochemical (including α-methyl glucoside) and E. carotovora subsp. atroseptica-specific PCR-based tests, which confirmed the above results and identified all 18 of the E. carotovora subsp. carotovora/odorifera isolates as E. carotovora subsp. carotovora (data not shown). This latter result was consistent with the use of the α-methyl glucoside test alone. ITS-PCR plus ITS-RFLP thus proved to be considerably faster, more convenient, and more accurate than other identification methods (CVP-differential temperature, biochemistry, and E. carotovora subsp. atroseptica-specific PCR) used in our laboratory and used in the present study to validate the new method.
This success was further supported when 12 “atypical” isolates (10 “atypical” E. carotovora subsp. atroseptica and 2 “atypical” E. carotovora subsp. carotovora isolates) were tested. Of 10 “atypical” E. carotovora subsp. atroseptica isolates tested by ITS-PCR, 2 were placed in group I, 6 were placed in group II, and 2 gave non-soft rot erwinia patterns. ITS-RFLP identified the group I isolates as E. carotovora subsp. atroseptica (one isolate) and E. carotovora subsp. betavasculorum (one isolate) and the group II isolates as E. carotovora subsp. carotovora (six isolates). After ITS-PCR, the two “atypical” E. carotovora subsp. carotovora isolates were placed in groups I and II and were identified as E. carotovora subsp. atroseptica and E. carotovora subsp. carotovora by using ITS-RFLP. Again, the identification of the isolates was confirmed using phenotypic, biochemical, and E. carotovora subsp. atroseptica-specific PCR-based tests. Only 2 of the 12 strains tested were correctly identified at the time of isolation, although correct identifications, at that time, may have provided important information about the epidemiology of these organisms and their role in disease. In addition, using ITS-PCR/RFLP we were able to identify non-soft rot species, which were selected initially for their cavity formation on CVP and which originally led to misidentification of disease symptoms.
The Erwinia genus includes a diverse group of pathogens that cause disease on a wide variety of plants (28). However, visual disease symptoms are not always sufficient to make an unequivocal identification of the pathogen involved. This is especially true when any one of a number of soft rot erwinias can lead to the development of similar disease symptoms on a common host, e.g., E. carotovora subsp. atroseptica, subsp. carotovora, and subsp. wasabiae and E. chrysanthemi all cause similar diseases on potato (11, 28). Thus, to ensure that a correct diagnosis is made and that steps are taken toward reducing disease spread, identification systems are required. While this can be achieved using detection systems that identify an organism directly from plant extract, e.g., PCR-based diagnostics (33), some soft rot erwinias, such as E. carotovora subsp. carotovora, are widely distributed in the environment and on plant surfaces, which can lead to false-positive results. The direct detection of a soft rot erwinia in plant extract is thus not necessarily an indication of its role in disease, although it is a useful initial screen. Ultimately, the isolation of the causative agent is needed, followed by a robust identification method. ITS-PCR plus ITS-RFLP is such a method and has proved to be the most simple, rapid, and with the exception of biochemical testing, most accurate method currently available for the identification of the soft rot erwinias. This is especially true when large numbers of isolates are tested. ITS-PCR alone is a useful method for the rapid identification of soft rot erwinia isolates (requiring 24 h, including DNA extraction) where an assumption has been made as to the species and/or subspecies present on a given host plant, e.g., E. carotovora subsp. atroseptica or subsp. carotovora or E. chrysanthemi on potato. However, it is recommended that ITS-PCR be used in conjunction with ITS-RFLP in all cases so as to obtain an accurate identification (requiring 48 h, including DNA extraction). When E. carotovora subsp. carotovora and subsp. odorifera cannot be differentiated, it would still be necessary to undertake an α-methyl glucoside test, adding an additional 3 to 5 days to the identification. However, this single biochemical test is relatively simple to perform and requires minimal labor time compared to performing multiple biochemical tests, which can take up to 14 days. The method also appears to be suitable for the identification of other Erwinia species, although more testing is required to confirm this. Based on this information, sequence analysis of the ITS may allow the phylogenetic relationships between different soft rot and other Erwinia species to be determined.
ACKNOWLEDGMENTS
This work was funded by the Scottish Executive Environment and Rural Affairs Department (SEERAD) and the British Potato Council.
We are grateful to Trevor Wicks and Barbara Morgan, The University of Adelaide, Adelaide, South Australia, Australia, for supplying isolates.
REFERENCES
- 1.Birch P R J, Hyman L J, Taylor R, Opio A F, Bragard C, Toth I K. RAPD PCR-based differentiation of Xanthomonas campestris pv. phaseoli and Xanthomonas campestris pv. phaseoli var. fuscans. Eur J Plant Pathol. 1997;103:809–814. [Google Scholar]
- 2.Brenner D J, Fanning G R, Steigerwalt A G. Deoxyribonucleic acid relatedness among species of Erwinia and between Erwinia species and other enterobacteria. J Bacteriol. 1972;110:12–17. doi: 10.1128/jb.110.1.12-17.1972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cuppels D, Kelman A. Evaluation of selective media for isolation of soft rot bacteria from soil and plant tissue. Phytopathology. 1974;64:468–475. [Google Scholar]
- 4.Darrasse A, Priou S, Kotoujansky A, Bertheau Y. PCR and restriction fragment length polymorphism of a pel gene as a tool to identify Erwinia carotovora in relation to potato diseases. Appl Environ Microbiol. 1994;60:1437–1443. doi: 10.1128/aem.60.5.1437-1443.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.De Boer S H, Sasser M. Differentiation of Erwinia carotovora subsp. carotovora and Erwinia carotovora subsp. atroseptica on the basis of cellular fatty-acid composition. Can J Microbiol. 1986;32:796–800. [Google Scholar]
- 6.De Boer S H, Ward L J. PCR detection of Erwinia carotovora subsp. atroseptica associated with potato tissue. Phytopathology. 1995;85:854–858. [Google Scholar]
- 7.De Boer S H, Verdonck L, Vruggink H, Harju P, Bång H O, De Ley J. Serological and biochemical variation among potato strains of Erwinia carotovora subsp. atroseptica and their taxonomic relationship to other E. carotovora strains. J Appl Bacteriol. 1987;63:487–495. [Google Scholar]
- 8.Dye D W. A taxonomic study of the genus Erwinia. II. The “carotovora” group. N Z J Sci. 1969;12:81–97. [Google Scholar]
- 9.Dye D W. A numerical taxonomy of the genus Erwinia. N Z J Agric Res. 1981;24:223–229. [Google Scholar]
- 10.Gallois A, Samson R, Ageron E, Grimont P A D. Erwinia carotovora subsp. odorifera subsp. nov., associated with odorous soft rot of chicory (Cichorium intybus L.) Int J Syst Bacteriol. 1992;42:582–588. [Google Scholar]
- 11.Goto M, Matsumoto K. Erwinia carotovora subsp. wasabiae subsp. nov. isolated from diseased rhizomes and fibrous roots of Japanese horseradish (Eutrema wasabi maxim) Int J Syst Bacteriol. 1987;37:130–135. [Google Scholar]
- 12.Guasp C, Moore E R B, Lalucat L, Bennasar A. Utility of internally transcribed 16S–23S rDNA spacer regions for the definition of Pseudomonas stutzeri genomovars and other Pseudomonas species. Int J Syst Evol Microbiol. 2000;50:1629–1639. doi: 10.1099/00207713-50-4-1629. [DOI] [PubMed] [Google Scholar]
- 13.Gurtler V, Barrie H D. Typing of Staphylococcus aureus strains by PCR-amplification of variable-length 16S–23S rDNA spacer regions: characterization of spacer sequences. Microbiology. 1995;141:1255–1265. doi: 10.1099/13500872-141-5-1255. [DOI] [PubMed] [Google Scholar]
- 14.Hauben L, Moore E R, Vauterin L, Steenackers M, Mergaert J, Verdonck L, Swings J. Phylogenetic position of phytopathogens within the Enterobacteriaceae. Syst Appl Microbiol. 1998;21:384–397. doi: 10.1016/S0723-2020(98)80048-9. [DOI] [PubMed] [Google Scholar]
- 15.Helias V, Le Roux A-C, Bertheau Y, Andrivon D, Gauthier J-P, Jouan B. Characterisation of Erwinia carotovora subspecies and detection of Erwinia carotovora subsp. atroseptica in potato plants, soil and water extracts with PCR-based methods. Eur J Plant Pathol. 1998;104:685–699. [Google Scholar]
- 16.Hyman L J, Dewasmes V, Toth I K, Perombelon M C M. Improved PCR detection sensitivity of Erwinia carotovora subsp. atroseptica in potato tuber peel extract by prior enrichment on a selective medium. Lett Appl Microbiol. 1997;25:143–147. [Google Scholar]
- 17.Jensen M A, Webster J A, Straus N. Rapid identification of bacteria on the basis of polymerase chain reaction-amplified ribosomal DNA spacer polymorphism. Appl Environ Microbiol. 1993;59:945–952. doi: 10.1128/aem.59.4.945-952.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kwon S-W, Go S-J, Kang H W, Ryu J-C, Jo J-K. Phylogenic analysis of Erwinia species based on 16S rRNA gene sequences. Int J Syst Bacteriol. 1997;47:1061–1067. doi: 10.1099/00207713-47-4-1061. [DOI] [PubMed] [Google Scholar]
- 19.Lelliot R A, Dickey R S. Genus VII. Erwinia. In: Krieg N T, Holt J G, editors. Bergey's manual of determinative bacteriology. 8th ed. Baltimore, Md: The Williams & Wilkins Co.; 1984. pp. 469–471. [Google Scholar]
- 20.Louws F J, Rademaker J L W, de Bruijn F J. The three Ds of PCR-based genomic analysis of phytobacteria: diversity, detection and disease diagnosis. Annu Rev Phytopathol. 1999;37:81–125. doi: 10.1146/annurev.phyto.37.1.81. [DOI] [PubMed] [Google Scholar]
- 21.Maki-Valkama T, Karjalainen R. Differentiation of Erwinia carotovora subsp. atroseptica and carotovora by RAPD-PCR. Ann Appl Biol. 1994;125:301–309. [Google Scholar]
- 22.Mendoza M, Meugnier H, Bes M, Etienne J, Freney J. Identification of Staphylococcus species by 16S–23S rDNA spacer PCR analysis. Int J Syst Bacteriol. 1998;48:1049–1055. doi: 10.1099/00207713-48-3-1049. [DOI] [PubMed] [Google Scholar]
- 23.Mergaert J, Verdonck L, Kersters K, Swings J, Boeufgras J-M, De Ley J. Numerical taxonomy of Erwinia species using API systems. J Gen Microbiol. 1984;130:1893–1910. [Google Scholar]
- 24.Moline H E, Wells J M. Differentiation of the soft rot Erwinia species on the basis of cellular fatty-acid composition. Phytopathology. 1987;77:1767–1767. [Google Scholar]
- 25.Murata N, Starr M P. Intrageneric clustering and divergence of Erwinia strains from plants and man in the light of deoxyribonucleic acid segmental homology. Can J Microbiol. 1974;20:1545–1565. doi: 10.1139/m74-242. [DOI] [PubMed] [Google Scholar]
- 26.Parent J-G, Lacroix M, Page D, Vezina L. Identification of Erwinia carotovora from soft rot diseased plants by random amplified polymorphic DNA (RAPD) analysis. Plant Dis. 1996;80:494–499. [Google Scholar]
- 27.Perombelon M C M, Lumb V M, Hyman L J. A rapid method to identify and quantify soft rot erwinias on seed potato tubers. EPPO Bull. 1987;17:25–35. [Google Scholar]
- 28.Perombelon M C M, Salmond G P C. Bacterial soft rots. In: Singh U S, Singh R P, Kohmoto K, editors. Pathogenesis and host specificity in plant diseases. Vol. 1. Oxford, England: Pergamon Press; 1995. pp. 1–17. [Google Scholar]
- 29.Persson P, Sletten A. Fatty acid analysis for the identification of Erwinia carotovora subsp. atroseptica and E. carotovora subsp. carotovora. EPPO Bull. 1995;25:151–156. [Google Scholar]
- 30.Riffard S, Presti F L, Normand P, Forey F, Reyrolle M, Etienne J, Vandenesch F. Species identification of Legionella via intergenic 16S–23S ribosomal spacer PCR analysis. Int J Syst Bacteriol. 1998;48:723–730. doi: 10.1099/00207713-48-3-723. [DOI] [PubMed] [Google Scholar]
- 31.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1989. [Google Scholar]
- 32.Toth I K, Bertheau Y, Hyman L J, Laplaze L, López M M, McNicol J, Niepold F, Persson P, Sletten A J, van der Wolf M, Pérombelon M C M. Evaluation of phenotypic and molecular typing techniques for determining diversity in Erwinia carotovora subspecies atroseptica. J Appl Microbiol. 1999;87:770–781. doi: 10.1046/j.1365-2672.1999.00929.x. [DOI] [PubMed] [Google Scholar]
- 33.Toth I K, Hyman L J, Wood J R. A one-step PCR-based method for the detection of economically important soft rot Erwinia species on micropropagated potato plants. J Appl Microbiol. 1999;87:158–166. [Google Scholar]
- 34.Vandamme P, Pot B, Gillis M, De Vos P, Kersters K, Swings J. Polyphasic taxonomy, a consensus approach to bacterial systematics. Microbiol Rev. 1996;60:407–438. doi: 10.1128/mr.60.2.407-438.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Verdonck J, Mergaert J, Rijckaert J, Swings J, Kersters K, DeLey J. The genus Erwinia: numerical analysis of phenotypic features. Int J Syst Bacteriol. 1987;37:4–18. [Google Scholar]
- 36.Wells J M, Moline H E. Differentiation of the soft-rotting Erwinia of the carotovora group by fatty acid composition. J Phytopathol. 1991;131:22–32. [Google Scholar]