• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Jul 2009; 75(13): 4539–4549.
Published online May 1, 2009. doi:  10.1128/AEM.01336-08
PMCID: PMC2704834

Phylogeny and Virulence of Naturally Occurring Type III Secretion System-Deficient Pectobacterium Strains[down-pointing small open triangle]


Pectobacterium species are enterobacterial plant-pathogenic bacteria that cause soft rot disease in diverse plant species. Previous epidemiological studies of Pectobacterium species have suffered from an inability to identify most isolates to the species or subspecies level. We used three previously described DNA-based methods, 16S-23S intergenic transcribed spacer PCR-restriction fragment length polymorphism analysis, multilocus sequence analysis (MLSA), and pulsed-field gel electrophoresis, to examine isolates from diseased stems and tubers and found that MLSA provided the most reliable classification of isolates. We found that strains belonging to at least two Pectobacterium clades were present in each field examined, although representatives of only three of five Pectobacterium clades were isolated. Hypersensitive response and DNA hybridization assays revealed that strains of both Pectobacterium carotovorum and Pectobacterium wasabiae lack a type III secretion system (T3SS). Two of the T3SS-deficient strains assayed lack genes adjacent to the T3SS gene cluster, suggesting that multiple deletions occurred in Pectobacterium strains in this locus, and all strains appear to have only six rRNA operons instead of the seven operons typically found in Pectobacterium strains. The virulence of most of the T3SS-deficient strains was similar to that of T3SS-encoding strains in stems and tubers.

The genus Pectobacterium (formerly Erwinia) contains both narrow- and broad-host-range bacterial plant pathogens that cause soft rot, stem rot, wilt, and blackleg in species belonging to over 35% of plant orders (20). Four Pectobacterium species have been described: Pectobacterium atrosepticum, Pectobacterium betavasculorum, Pectobacterium carotovorum, and Pectobacterium wasabiae (9). The recently described organism P. carotovorum subsp. brasiliensis is genetically distinct from previously described Pectobacterium taxa; approximately 82% of its genes are shared with P. atrosepticum, and 84% of its genes are shared with P. carotovorum subsp. carotovorum, while 13% of its genes are found in neither P. atrosepticum nor P. carotovorum subsp. carotovorum (7, 10, 20). To date, only P. carotovorum subsp. carotovorum and P. atrosepticum have been reported to occur in the same field (14, 21). P. carotovorum subsp. carotovorum is found worldwide, and P. atrosepticum is found in cool climates; while P. carotovorum subsp. brasiliensis has been found only in Brazil, Israel, and the United States, it is likely to have a wider distribution (20). Compared to the ecology and genetics of P. carotovorum subsp. carotovorum and P. atrosepticum, little is known about the ecology and genetics of P. betavasculorum, P. wasabiae, or P. carotovorum subsp. brasiliensis.

Pectobacterium strains isolated from potato are diverse based on serology, genome structure, and fatty acid composition (5, 35). Previous epidemiological studies of pectolytic Enterobacteriaceae were complicated by the diversity of this group and the lack of tools capable of placing all isolates into clades. For example, Gross et al. (14) were unable to classify over 50% of Pectobacterium isolates obtained from potato, and Pitman et al. (23) were unable to type 13% of their isolates. Novel PCR-based methods potentially capable of classifying all Pectobacterium isolates have been described, but they were developed prior to the recognition of P. carotovorum subsp. brasiliensis (1, 34).

The main virulence determinants of Pectobacterium are the pectolytic enzymes secreted through the type II secretion system. Although these enzymes are required for development of symptoms, many other virulence genes have been shown to contribute to Pectobacterium pathogenicity, including the type III secretion system (T3SS) genes, the cfa gene cluster, and the type IV secretion system genes (3, 15, 19). Recent genomic analysis showed that some of these gene clusters, such as the cfa and type IV secretion system cluster genes, as well as genes important for interactions with insects, are present in only some Pectobacterium species (10). Thus, Pectobacterium species appear to use different genetic tools to overcome plant host barriers and to interact with insect vectors.

Many gram-negative pathogenic bacteria secrete virulence proteins, known as effectors, through the T3SS into host cells. Once inside host cells, the effectors manipulate host defenses and promote bacterial growth (13). Unlike many other gram-negative plant pathogens, Pectobacterium does not require the T3SS for pathogenicity. Rather, this secretion system makes a small, but measurable, contribution to the early stages of P. carotovorum growth in leaves of the model plant Arabidopsis thaliana (26) and contributes to the virulence of P. atrosepticum on potato (15). Recently, we isolated Pectobacterium strains that lack the T3SS from potatoes and also found P. wasabiae and P. carotovorum subsp. brasiliensis on potatoes in Wisconsin (35). The first goal of this study was to determine if P. wasabiae and P. carotovorum subsp. brasiliensis are common in agricultural fields or if soft rot disease is typically caused by P. carotovorum subsp. carotovorum and P. atrosepticum, which have been the focus of nearly all previous studies of potato soft rot, stem rot, and blackleg disease. Second, since we recently isolated a strain lacking the T3SS (35), we also aimed to determine if strains lacking the T3SS are common in infected potatoes and if these strains tend to be less virulent on potato stems and tubers than strains encoding a T3SS.


Bacterial strains and growth media.

Bacterial strains used in this study are listed in Table Table1.1. In 2001, we isolated Pectobacterium isolates from diseased potato tubers and stems from four different fields in Wisconsin (37). Additional Pectobacterium isolates were obtained from diseased potato tubers collected from three different fields in Wisconsin in 2004 (Table (Table1).1). The bacteria were isolated by the methods described by Schaad et al. (29) and Yap et al. (35). Briefly, bacteria from decayed tubers were streaked either onto crystal violet pectate (CVP) medium prepared with either Bulmer or Slendid pectate (16) or onto a raffinose (RAF) medium modified from the medium described by Segall (30), and the cultures were incubated at room temperature for at least 3 days. The modified RAF medium consists of (per liter) 10 g of raffinose, 2 g of K2HPO4, 5 g of ammonium sulfate, 0.4 g of eosin Y, 0.065 g of methylene blue, and 18 g of agar. Pectobacterium forms pits on CVP medium since it digests the pectate that is used to solidify this medium; thus, pit-forming colonies were presumed to be colonies of Pectobacterium species. Isolates obtained from pit-forming colonies on CVP medium plates were streaked onto Luria-Bertani (LB) agar. On RAF medium, Pectobacterium forms either red colonies without halos or, occasionally, metallic green colonies without halos, depending on the batch of medium. Such colonies were streaked onto LB agar and were then tested for the ability to form pits on CVP medium. All isolates were confirmed to be Pectobacterium sp. isolates using a 16S-23S intergenic transcribed spacer (ITS) PCR assay described by Toth et al. (34). Pectobacterium strains were routinely grown in LB medium at either room temperature or 37°C. Although common in potato-growing regions in Europe, the related pathogen Dickeya has not been reported on potato in Wisconsin yet, and no Dickeya isolates were obtained.

Strains and plasmids used in this study

HR assay.

Nicotiana tabacum L. cv. Xanthi NN plants used for hypersensitive response (HR) assays were grown in a greenhouse. For HR elicitation assays, bacterial cells were grown overnight on LB medium and were then suspended in sterile water to a concentration of 108 CFU/ml. The bacterial cell suspensions were infiltrated into fully expanded leaves of 6- to 7-week-old N. tabacum L. cv. Xanthi NN with a needleless 1-ml syringe (2). Plants were examined for HR elicitation 24 h after infiltration. Each assay was performed in triplicate, and sterile distilled water was infiltrated into leaves as a negative control. All plant assays were repeated at least three times.

DNA hybridization and PCR analysis.

Pectobacterium T3SS genes were detected by DNA hybridization and PCR amplification with T3SS-specific probes and primers for genes encoding regulatory, structural, and secreted proteins. For DNA hybridization assays, chromosomal DNA was digested with EcoRI, electrophoresed through a 0.8% agarose gel, and transferred to a nylon membrane (Millipore Co., Billerica, MA). Probe DNA was either PCR amplified from P. carotovorum WPP14 genomic DNA (hrpL, hrpN, hrpW-dspE/F, hecB, and a region upstream of hrcU) or amplified from plasmids containing DNA cloned from P. carotovorum WPP14 [a 5.8-kb PCR-amplified fragment containing hrpD through hrpN amplified from pThrpDN(Ecc) and a 3.5-kb PCR-amplified fragment containing hrpB through hrcC amplified from pThrpBC(Ecc)]. The primers and rregions used are described in Table Table2.2. A PCR amplification program consisting of 30 cycles of 94°C for 30 s, 54°C for 30 s, and 72°C for 30 s to 5 min, depending on the size of the amplified fragment, was used. The probes were labeled with a Gene Images Alkphos direct labeling and detection system kit (GE Healthcare Bio-Sciences Corp., Piscataway, NJ). Hybridization results were recorded on X-ray film (Kodak, Los Angeles, CA). Additionally, the presence of hrcC, which encodes the T3SS porin, was detected with the primer set listed in Table Table2.2. 16S-23S ITS-PCR and restriction fragment length polymorphism (RFLP) analyses were performed as described by Toth et al. (34). Briefly, the rRNA intergenic spacer region was amplified with primers G1 and L1, the amplified DNA was digested with RsaI (New England Biolabs, Ipswich, MA), and the DNA was analyzed by gel electrophoresis on 3% GenePure HiRes agarose gels (ISC Bioexpress, Kaysville, UT) in Tris-borate-EDTA buffer.

Oligonucleotides used in this study

Phylogenetic analyses.

Fragments of six conserved housekeeping genes, acnA (aconiate hydrase 1), gapA (glyceraldehyde-3-phosphate dehydrogenase A), icdA (isocitrate dehydrogenase), mdh (malate dehydrogenase), pgi (glucose-6-phosphate isomerase), and proA (γ-glutamylphosphate reductase), were PCR amplified from the type strains of Pectobacterium sp. and Dickeya chrysanthemi and the strains isolated in Wisconsin, including WPP2, WPP5, WPP12, WPP16, WPP20, WPP127, and WPP156 (Table (Table1),1), as described previously (20). The sequences of the PCR-amplified fragments were obtained with a BigDye Terminator kit (Perkin-Elmer). Sequence chromatogram output files were initially aligned and edited using SeqMan 5.08 (DNASTAR, Inc., Madison, WI). A phylogenetic analysis of the sequences of 51 bacterial strains was conducted using PAUP* 4.0b10 and the concatenated gene sequence data set. Yersinia sp. strains were used as outgroups in all reconstructions. We used ModelTest 3.7 to select the standard AIC model (TrN with a gamma correction for rate variation among sites [32]) as the best-fitting model of evolution (24, 25). The corresponding likelihood parameters were estimated and then applied in PAUP* 4.0b10. Two different algorithms were then used, maximum likelihood-corrected neighbor joining (NJ) and weighted maximum parsimony (MP) with substitutions weighted according to the instantaneous rate matrix or characters weighted according to their rescaled consistency index values. Bootstrap values were estimated in order to evaluate the support for each clade (1,000 replicates for maximum likelihood-corrected NJ and MP heuristic search) with sampling limited to parsimony-informative characters with a cutoff value of 50.

Pulsed-field gel electrophoresis (PFGE) analysis.

Agarose plugs containing chromosomal DNA of each bacterial strain were prepared and digested with I-CeuI (New England Biolabs, Inc.) as described previously (27, 31, 35). Briefly, a Pulsaphor Plus system with a hexagonal electrode array (Pharmacia, Uppsala, Sweden) was used for electrophoresis by following the manufacturer's instructions. The digested genomic DNA was separated by electrophoresis on a 1% agarose gel at 12°C for 20 h at 5 V/cm with the switch time ramping from 25 to 45 s. ProMega lambda ladder (Promega, Madison, WI) was used as the size marker, and the DNA was stained with ethidium bromide and visualized with a UV light transilluminator.

Virulence assays.

Virulence assays were performed with potato tubers and stems (Solanum tuberosum cv. Atlantic) in the laboratory and field, respectively. The relative virulence of 10 Pectobacterium strains in potato tubers (cv. Atlantic) was determined by measuring the amount of macerated potato tissue (35). Ten to 12 potato tubers were inoculated by placing 10 μl of a 108-CFU/ml bacterial suspension into 1-cm-deep holes poked into the tubers with a pipette tip. The tubers were placed in plastic bags and incubated for 2 days at 28°C. This inoculum level was chosen because it was the lowest level at which the tubers reliably developed symptoms within 2 days with the most virulent strains examined. After 2 days the inoculated tubers were cut open, and the macerated tissue was removed and weighed. Five assays with P. carotovorum subsp. carotovorum WPP14, P. carotovorum subsp. brasiliensis WPP17, P. wasabiae WPP163, and P. carotovorum subsp. brasiliensis WPP165 were conducted in the summers of 2005 and 2006 at the University of Wisconsin Hancock Agricultural Research Station using potato plants (S. tuberosum cv. Atlantic). S. tuberosum cv. Atlantic was chosen because it developed consistent and reliable symptoms in preliminary greenhouse assays. Bacterial treatments were arranged in a randomized complete block design with three blocks. Each plot was 4.5 m (16 rows) by 16.7 m, and every two rows were separated by two boarder rows in which S. tuberosum cv. Red Norland was planted. Twenty potato tubers were planted for each bacterial treatment on 23 April 2005 and 21 April 2006. In 2005 and 2006, 8 to 10 and 15 to 20 potato stems per block, respectively, were inoculated with WPP14, WPP17, WPP163, and WPP165. Prior to inoculation, the Pectobacterium strains were grown overnight on LB agar. The bacteria were scraped from the plates with sterile toothpicks, and then the plants were stab inoculated with the toothpicks. One stem per plant was inoculated with one strain, and a second stem was stabbed with a sterile toothpick as a negative control. The toothpicks were left in the stem wounds, and each wound was sealed with a strip of Parafilm, both to protect the bacteria from desiccation and to mark the inoculated and control stems. One week after inoculation, the stems were collected from the field. The stems were sliced in half, and the length of the lesion in each stem was measured.

Statistical analysis.

Statistical analysis of data was conducted using Statistical Analysis Systems (SAS Institute, Cary, NC). Analysis of variance was determined using the general linear model procedures, and means were separated with the least significant difference.


Pectobacterium strains lacking the T3SS were isolated from multiple potato fields.

In 2001, we isolated a strain which lacked a T3SS, WPP17. To determine if this strain was an anomaly or if T3SS-deficient strains could be isolated from additional field samples, we examined an additional 45 Pectobacterium isolates from potatoes with soft rot from three fields in Wisconsin in 2004. To determine which Pectobacterium isolates lacked a T3SS, we examined the abilities of the isolates to elicit the HR on tobacco plants (N. tabacum L. cv. Xanthi NN). With two exceptions, all strains isolated from the same tubers had the same HR phenotype (Tables (Tables11 and and33 and Fig. Fig.1).1). Eight representative isolates from six tubers (WPP127, WPP156, WPP161, WPP163, WPP165, WPP168, WPP172, and WPP178) were examined with DNA hybridization assays to determine if the isolates unable to elicit an HR encoded a T3SS (Fig. (Fig.2).2). DNA hybridization assays showed that the Pectobacterium isolates unable to elicit an HR, WPP161, WPP163, WPP168, and WPP172, did not contain the T3SS alternative sigma factor gene hrpL, the hrpN, hrpW, and dspE genes encoding T3SS-secreted proteins, or the gene encoding the DspE chaperone, dspF (Fig. (Fig.2).2). Additionally, DNA hybridization assays showed that the strains unable to elicit an HR lacked the genes in the T3SS cluster from hrpD to hrpB (Fig. (Fig.2).2). We also attempted to PCR amplify conserved fragments of the region encoding the T3SS, including a region of hrcC and a fragment containing hrcQR, with primers based on conserved regions of these genes, but we were unable to amplify DNA from WPP161, WPP163, WPP168, or WPP172 (not shown). Thus, the strains unable to elicit an HR lacked a key regulatory gene required for a functional T3SS, the genes encoding the T3SS machinery structure, and the genes encoding three proteins secreted via the T3SS. Consistent with the results of Yap et al. (35), we found that WPP17 genomic DNA did not hybridize to any of T3SS genes.

FIG. 1.
Sixteen of 45 Pectobacterium isolates obtained in 2004 and P. wasabiae SCRI488 were unable to elicit an HR on N. tabacum cv. Xanthi. Bacterial isolates were infiltrated into the leaves of 6- to 7-week-old N. tabacum cv. Xanthi, and results were recorded ...
FIG. 2.
Pectobacterium isolates unable to elicit an HR did not contain the genes for the Pectobacterium T3SS. (A) The P. carotovorum WPP14 T3SS is indicated by open arrows, and the border genes, hecB, and ECA2075 are indicated by filled arrows. (B) Genomic DNA ...
Description of Pectobacterium strains isolated from seven fields in Wisconsin in 2001 and 2004 based on PFGE and elicitation of an HR on tobacco leaves (N. tabacum L. cv. Xanthi NN)

Since most of the T3SS-deficient strains clustered with P. wasabiae in our phylogenetic analysis (see Fig. Fig.5),5), we assayed P. wasabiae strain SCRI488 to determine if it also lacked a T3SS. We found that SCRI488 was unable to elicit an HR in tobacco and that SCRI488 genomic DNA did not hybridize to DNA encoding the T3SS (Fig. (Fig.11 and and2).2). Thus, P. wasabiae SCRI488 also lacks a T3SS entirely or encodes an atypical T3SS. Since SCRI488 was isolated in Japan, T3SS-deficient P. wasabiae may be widespread.

FIG. 5.
Character and substitution weight parsimony phylogeny generated by using concatenated sequences of six housekeeping genes of 51 isolates, including 4 Brenneria sp. isolates, 10 Dickeya sp. isolates, 4 Yersinia sp. isolates, and 33 Pectobacterium sp. isolates. ...

Multiple deletions of genes in the T3SS locus have occurred.

The flanking regions of the T3SS gene cluster are conserved in the three sequenced Pectobacterium genomes (3, 10). To determine if the regions surrounding the T3SS gene cluster were present in the T3SS-deficient isolates, two genes flanking the T3SS gene cluster (hecB encoding a TpsB transporter homolog and ECA2075 encoding a lysR homolog) were amplified from P. carotovorum subsp. carotovorum WPP14 and were used as probes in DNA hybridization assays (Fig. (Fig.3).3). Genomic DNA from most of the T3SS-deficient strains hybridized to both probes, suggesting that the T3SS genes, but not the surrounding genes, were not present in these strains. However, WPP172 and WPP17 lacked hecB and ECA2075, respectively; thus, additional genes are not present in these two T3SS-deficient strains. We attempted to PCR amplify the DNA between ECA2075 and hecB in the T3SS-deficient strains that contain both of these genes, but we were unable to amplify any fragments from this region.

FIG. 3.
Genomic DNA from WPP17 and WPP172 were unable to hybridize to two genes flanking the T3SS region, ECA2075 and hecB, respectively. Genomic DNA was digested with SacII and EcoRV and then hybridized with hecB or ECA2075 probes. The presence of T3SS and the ...

T3SS-deficient P. wasabiae strains were capable of causing disease in potato stems and tubers.

The relative virulence of a Pectobacterium strain was evaluated by inoculating it into potato tubers and stems and then weighing the amount of macerated tuber tissue or measuring the length of the stem lesion. The strains assayed could be grouped into three classes based on their tuber maceration ability (Fig. (Fig.4).4). The tuber maceration amounts obtained for P. carotovorum subsp. brasiliensis WPP17 and P. wasabiae SCRI488 were significantly lower, while P. carotovorum subsp. carotovorum WPP14 and P. carotovorum subsp. brasiliensis WPP165 macerated tuber tissue most efficiently. The abilities of rest of the strains to macerate potato tubers were intermediate. No correlation between the aggressiveness of tuber maceration and the presence of the T3SS was observed.

FIG. 4.
Presence of the Pectobacterium T3SS was not correlated with the ability to macerate potato tubers (A) or cause disease in stems (B). For the inoculated tubers, the bars indicate the amounts of tissue macerated, and the error bars indicate the standard ...

To determine if the T3SS-deficient strains were capable of causing disease in stems, we inoculated field-grown potato stems (cv. Atlantic) with T3SS-encoding and T3SS-deficient strains, including P. carotovorum subsp. carotovorum WPP14, P. carotovorum subsp. brasiliensis WPP165, P. carotovorum subsp. brasiliensis WPP17, and P. wasabiae WPP163, and measured lesion lengths 1 week after inoculation (Fig. (Fig.4).4). The lesion lengths were similar for WPP14, WPP165, and the T3SS-deficient strain WPP163 (P = 0.05), except for the second trial in 2005. In contrast, the ability of WPP17 to cause stem rot in potato was significantly less than that of the other three Pectobacterium strains (P = 0.05). In the negative controls, only the toothpick stab wounds were observed. Thus, these three isolates of P. carotovorum subsp. carotovorum, P. carotovorum subsp. brasiliensis, and P. wasabiae caused comparable levels of tuber and stem rot, and the T3SS is not required for Pectobacterium to cause soft rot disease in tubers or stems.

Multiple Pectobacterium species were present in fields and, occasionally, in individual diseased tubers.

Isolates obtained in 2001 and 2004 were used to determine if infected plants from individual fields were typically infected with one or multiple species. Two previously described DNA-based methods, multilocus sequence analysis (MLSA) (20) and 16S-23S ITS-PCR-restriction fragment length polymorphism (RFLP) analysis (34), were used to determine the species of the isolates (Fig. (Fig.55 and Fig. Fig.6).6). Our MLSA relied on building phylogenetic trees from concatenated gene sequences of six housekeeping genes, acnA, gapA, icdA, mdh, pgi, and proA, and it provided robust support for clades corresponding to the four previously described Pectobacterium species (clades 2 through 5) and for P. carotovorum subsp. brasiliensis (clade 1) (20). Sequences of seven new Pectobacterium isolates and five type strains (four Pectobacterium strains and one Dickeya strain) were added to the MLSA developed by Ma et al. (20). The new isolates were chosen because they produced PFGE or 16S-23S ITS-PCR-RFLP patterns not produced by strains included in the previously described phylogenetic tree and therefore could not easily be identified to the species level. For example, WPP127 and WPP156, both of which were placed in the P. carotovorum subsp. carotovorum clade (Fig. (Fig.5),5), produced a novel 16S-23S ITS-PCR-RFLP pattern (Fig. (Fig.6),6), and their PFGE patterns were different from each other (Table (Table33 and Fig. Fig.7).7). The tree in Fig. Fig.55 lacks sequences of 14 Pectobacterium isolates used by Ma et al. (20); thus, approximately two-thirds of the Pectobacterium data are shared by the two trees.

FIG. 6.
16S-23S ITS-PCR-RFLP patterns of Pectobacterium strains. The 16S-23S ITS region was amplified with primers L1 and G1, and the amplified DNA was digested with RsaI and then analyzed by gel electrophoresis. The strains used are indicated above the lanes, ...
FIG. 7.
Four representative I-CeuI patterns for the 42 Pectobacterium isolates obtained from diseased potato tubers in Wisconsin in 2004. WPP14 was isolated in 2001 from a diseased potato plant with aerial stem rot. Fragments more than 1,000 kb long are not shown. ...

All of the isolates from Wisconsin potato fields were identified as either P. carotovorum or P. wasabiae and are in clade 1, 2, or 5. The 22 Pectobacterium isolates included in both this analysis and the analysis of Ma et al. (20) were placed in the same clades in both trees. Pectobacterium species assignments for strains are shown in Table Table11 and are based on a combination of MLSA and PFGE results. Some strains with identical PFGE patterns from different sources were included in our MLSA (such as WPP161 and WPP168) and previous MLSA (such as WPP14 and Ecc380) (20) and were found to have nearly identical DNA sequences; thus, strains with identical PFGE patterns from the same tuber or stem were assumed to be clonal. We found strains belonging to multiple clades in all seven fields, and twice we isolated strains belonging to multiple clades from single tubers (Table (Table11).

We examined the isolates using the 16S-23S ITS-PCR-RFLP method as described by Toth et al. (34) with the hope that a method that is faster and simpler than MLSA could be used to reliably identify Pectobacterium species. We found two ITS-PCR-RFLP patterns for P. carotovorum subsp. carotovorum strains, one of which was described previously (34) (Fig. (Fig.6).6). Similarly, we found two patterns for P. carotovorum subsp. brasiliensis strains, one of which was described previously (34). The pattern for WPP17, which is at the base of the P. carotovorum subsp. brasiliensis clade and which lacks a T3SS, was yet a third pattern, which was identical to a pattern described by Toth et al. (34). We also found two ITS-PCR-RFLP patterns for P. wasabiae, one of which was described previously (34) and one of which was identical to the pattern obtained for some P. carotovorum subsp. carotovorum strains. WPP19, which is at the base of the P. wasabiae clade, had a pattern similar to that obtained for the majority of the P. wasabiae strains and also similar to that of some of the P. carotovorum subsp. carotovorum strains. Since P. carotovorum subsp. carotovorum strains, such as WPP14, and P. wasabiae strains, such as WPP161, have identical ITS-PCR-RFLP patterns, the 16S-23S ITS-PCR-RFLP method could not be used to reliably place Pectobacterium isolates into taxa.

To determine if Pectobacterium strains isolated from individual tubers were clonal and to determine if PFGE patterns could be correlated with Pectobacterium clades, genomic DNA was digested with endonuclease I-CeuI and separated by PFGE (Fig. (Fig.7).7). I-CeuI specifically targets a 26-bp sequence on 23S rRNA genes of bacterial rrn operons; thus, the number of digested fragments represents the number of rrn operons on the chromosome. PFGE profiles were obtained for 42 of the 45 isolates obtained in 2004 examined (Table (Table33 and Fig. Fig.7).7). As expected, digestion of all Pectobacterium isolates yielded five or six fragments ranging from 50 to 1,000 kb long (Fig. (Fig.7)7) and a large fragment over 3,000 kb long (not shown) (35). A total of four distinct I-CeuI PFGE pulsotypes were observed, and all isolates from each individual tuber had identical patterns. Unfortunately, we were unable to obtain PFGE data for WPP168, which was a T3SS-deficient isolate obtained from a tuber also containing T3SS-encoding isolates. Strains with identical PFGE pulsotypes were also obtained from fields A and C; thus, isolates with the same pulsotype may be obtained from multiple locations. We gave the PFGE pulsotypes arbitrary numbers, and only pulsotypes 1 and 8 were seen in 2001 and 2004. P. carotovorum Ecc71, a model strain used for many years to study P. carotovorum genetics, was also a pulsotype 1 strain (35). The P. wasabiae PFGE pattern differed from the other patterns in that only five fragments smaller than 1,000 kb were apparent, suggesting that strains in the P. wasabiae clade contain six rRNA operons rather than the seven operons found in the other Pectobacterium species. WPP17, which is most closely related to P. carotovorum subsp. brasiliensis and which also lacks a T3SS, also appears to contain only six rRNA operons (35).


In this work we examined strains collected from seven different fields in 2001 and/or 2004. We tested three methods for typing the Pectobacterium isolates and found that MLSA provided the most unambiguous species identification. ITS-PCR-RFLP analysis resulted in identical patterns for different species, while PFGE resulted in numerous patterns per species. We used a combination of these methods to demonstrate that P. carotovorum subsp. brasiliensis, P. carotovorum subsp. carotovorum, and P. wasabiae can be found in single fields and that these taxa are sometimes present together in single infected plants. Because multiple taxa were found in each field and in some individual samples, we decided to halt analysis of field samples until a method that can differentiate strains in Pectobacterium clades directly from field samples can be developed.

We also found that Pectobacterium strains lacking the T3SS can be isolated from diseased tubers and that the T3SS-deficient strains are still virulent in potato tubers and stems. A recent draft sequence of WPP163 shows that the T3SS is indeed not present in this strain (N. T. Perna, unpublished data; NCBI Genome Project ID 31293). Since we used high inoculum levels, we cannot rule out the possibility that the T3SS-encoding Pectobacterium strains may be able to better colonize plants or cause disease when bacteria are inoculated at lower concentrations, under environmental conditions, or into hosts or tissues that were not examined. It should not be assumed the T3SS-deficient P. wasabiae isolates are secondary invaders since P. wasabiae was the only species isolated from some samples, and P. wasabiae is capable of causing disease in both stems and tubers.

Recent genomic analyses of three Pectobacterium genomes revealed that gene clusters identified as clusters important for virulence in P. atrosepticum are not present in one or both sequenced P. carotovorum genomes (10). Similarly, our results show that Pectobacterium sp. strains lacking the T3SS, which has been reported to play a role in Pectobacterium virulence (15, 26), can be isolated from diseased tubers. The T3SS-deficient strains are likely to have lost a T3SS rather than to never have had one since closely related T3SS genes are present in the same relative chromosomal position in both Dickeya and Pectobacterium, suggesting that the T3SS was acquired by the common ancestor of these two soft rot genera. The mosaic of the virulence genes present in these Pectobacterium species suggests that different strains use different tools to overcome plant barriers. P. wasabiae WPP172, but not the other P. wasabiase strains, also lacks at least one gene, hecB, that flanks the T3SS gene cluster in Pectobacterium and Dickeya and that plays a role in Dickeya virulence (28). Finally, both the T3SS-deficient strains in the P. wasabiae clade and P. carotovorum WPP17 appear to contain only six rRNA operons, rather than the seven rRNA operons found in all other Pectobacterium strains analyzed. Spontaneous deletion of rrn operons has been reported for Yersinia species as well (6). It appears that deletion of an rrn operon, as well as deletion of the T3SS, occurred at least twice in the genus Pectobacterium.

Strains lacking a functional T3SS have been reported for the plant pathogens Pseudomonas syringae (22) and Erwinia pyrifoliae (18). P. wasabiae differs from these plant pathogens since it lacks a T3SS, but it is still virulent. Also, in contrast to T3SS-deficient P. syringae strains, which are closely related to each other and separate from pathogenic P. syringae (22), T3SS-deficient Pectobacterium strains fall into two species, P. carotovorum and P. wasabiae. The T3SS-deficient Pectobacterium strains may be most analogous to the animal pathogen Pseudomonas aeruginosa. Strains of this human pathogen isolated from patients with chronic lung infections are typically T3SS deficient, even though 90% of environmental P. aeruginosa isolates encode a T3SS (17).

The reported host range of Pectobacterium expands regularly, but it is challenging to identify the Pectobacterium species isolated from newly identified hosts, as well as from commonly studied hosts, such as potato. Of the three DNA-based methods that we used to characterize isolates, MLSA provided the most unambiguous results and allowed us to place isolates into defined clades with strong branch support (20), while the other two methods could not be used to unambiguously place isolates into taxa. All three methods tested suffer from the limitation that they can be used only with pure cultures. However, combination of the MLSA data with genome sequence data should aid in the development of sensitive assays capable of detecting and classifying Pectobacterium strains directly from field samples, thereby allowing tracking of these taxa in the environment that is more efficient than the tracking that can be done with currently available assays.

The MLSA method appears to be robust for classification of Pectobacterium. One-third of the Pectobacterium isolates in the MLSA tree differ from the isolates used previously (20), and this did not change the clades into which the remaining Pectobacterium isolates were placed. We defined clade 2 as a clade that includes both P. carotovorum subsp. carotovorum and P. carotovorum subsp. P. carotovorum subsp. odoriferum (SCRI482). P. carotovorum subsp. odoriferum was established based on DNA hybridization with three P. carotovorum strains (8), one of which was later determined not to be a P. carotovorum strain (9). Further review of P. carotovorum phylogeny is needed to determine if maintaining this subspecies designation is appropriate. As shown previously (20), Brenneria isolates did not form a monophyletic clade. Sequence data for additional Brenneria isolates are required to resolve how they are related to each other and to the soft rot enterobacteria.


We thank Jane and Jeff Breuer for their invaluable assistance with our plant experiments. We also thank Ruth Genger, Courtney Jahn, Maria del Pilar Marquez Villavicencio, Mafmudije Selimi, Mee-Ngan Yap, Ralph Reedy, Jennifer Apodaca, and Bryan Biehl for their assistance and suggestions.

This study was supported by National Science Foundation award 0412599 (“BE/GenEn: Genome-Enabled Analyses of Natural Populations of Pathogens on Natural Hosts”) and funds from the Wisconsin Potato and Vegetable Growers.


[down-pointing small open triangle]Published ahead of print on 1 May 2009.


1. Avrova, A. O., L. J. Hyman, R. L. Toth, and I. K. Toth. 2002. Application of amplified fragment length polymorphism fingerprinting for taxonomy and identification of the soft rot bacteria Erwinia carotovora and Erwinia chrysanthemi. Appl. Environ. Microbiol. 68:1499-1508. [PMC free article] [PubMed]
2. Bauer, D. W., A. J. Bogdanove, S. V. Beer, and A. Collmer. 1994. Erwinia chrysanthemi hrp genes, and their involvement in soft rot pathogenesis and elicitation of the hypersensitive response. Mol. Plant-Microbe Interact. 7:573-581. [PubMed]
3. Bell, K. S., M. Sebaihia, L. Pritchard, M. T. G. Holden, L. J. Hyman, M. C. Holeva, N. R. Thomson, S. D. Bentley, L. J. C. Churcher, K. Mungall, R. Atkin, N. Bason, K. Brooks, T. Chillingworth, K. Clark, J. Doggett, A. Fraser, Z. Hance, H. Hauser, K. Jagels, S. Moule, H. Norbertczak, D. Ormond, C. Price, M. A. Quail, M. Sanders, D. Walker, S. Whitehead, G. P. C. Salmond, P. R. J. Birch, J. Parkhill, and I. K. Toth. 2004. Genome sequence of the enterobacterial phytopathogen Erwinia carotovora subsp. atroseptica and characterization of virulence factors. Proc. Natl. Acad. Sci. USA 101:11105-11110. [PMC free article] [PubMed]
4. Burkholder, W. H., L. A. McFadden, and A. W. Dimock. 1953. A bacterial blight of chrysanthemums. Phytopathology 43:522-526.
5. De Boer, S. H., and M. Sasser. 1986. Differentiation of Erwinia carotovora subsp. carotovora and E. carotovora subsp. atroseptica on the basis of cellular fatty acid composition. Can. J. Microbiol. 32:796-800.
6. Deng, W., V. Burland, G. I. Plunkett, A. Boutin, G. F. Mayhew, P. Liss, N. T. Perna, D. J. Rose, B. Mau, S. Zhou, D. C. Schwartz, J. D. Fetherston, L. E. Lindler, R. R. Brubaker, G. V. Plano, S. C. Straley, K. A. McDonough, M. L. Nilles, J. S. Matson, F. R. Blattner, and R. D. Perry. 2002. Genome sequence of Yersinia pestis KIM. J. Bacteriol. 184:4601-4611. [PMC free article] [PubMed]
7. Duarte, V., S. H. De Boer, L. J. Ward, and M. C. de Oliveira. 2004. Characterization of atypical Erwinia carotovora strains causing blackleg of potato in Brazil. J. Appl. Microbiol. 96:535-545. [PubMed]
8. Gallois, A., R. Samson, E. Ageron, and P. A. D. Grimont. 1992. Erwinia carotovora subsp. odorifera subsp. nov., associated with odorous soft rot of chicory (Cichorium intybus L.). Int. J. Syst. Bacteriol. 42:582-588.
9. Gardan, L., C. Gouy, R. Christen, and R. Samson. 2003. Elevation of three subspecies of Pectobacterium carotovorum to species level: Pectobacterium atrosepticum sp. nov., Pectobacterium betavasculorum sp. nov. and Pectobacterium wasabiae sp. nov. Int. J. Syst. Evol. Microbiol. 53:381-391. [PubMed]
10. Glasner, J. D., M. Marquez-Villavicencio, H.-S. Kim, C. E. Jahn, B. Ma, B. S. Biehl, A. I. Rissman, B. Mole, X. Yi, C.-H. Yang, J. L. Dangl, S. R. Grant, N. T. Perna, and A. O. Charkowski. 2008. Niche-specificity and the variable fraction of the Pectobacterium pan-genome. Mol. Plant-Microbe Interact. 21:1549-1560. [PubMed]
11. Goto, M., and K. Matsumoto. 1987. Erwinia carotovora subsp. wasabiae subsp. nov. isolated from diseased rhizomes and fibrous roots of Japanese horseradish (Eutrema wasabi Maxim.). Int. J. Syst. Bacteriol. 37:130-135.
12. Graham, D. C., and W. J. Dowson. 1960. The coliform bacteria associated with potato black-leg and other soft rots. I. Their pathogenicity in relation to temperature. Ann. Appl. Biol. 48:51-57.
13. Grant, S. R., E. J. Fisher, J. H. Chang, B. M. Mole, and J. L. Dangl. 2006. Subterfuge and manipulation: type III effector proteins of phytopathogenic bacteria. Annu. Rev. Microbiol. 60:425-449. [PubMed]
14. Gross, D. C., M. L. Powelson, K. M. Regner, and G. K. Radamaker. 1991. A bacteriophage-typing system for surveying the diversity and distribution of strains of Erwinia carotovora in potato fields. Phytopathology 81:220-226.
15. Holeva, M. C., K. S. Bell, L. J. Hyman, A. O. Avrova, S. C. Whisson, P. R. J. Birch, and I. K. Toth. 2004. Use of a pooled transposon mutation grid to demonstrate roles in disease development for Erwinia carotovora subsp. atroseptica putative type III secreted effector (DspE/A) and helper (HrpN) proteins. Mol. Plant-Microbe Interact. 17:943-950. [PubMed]
16. Hyman, L. J., L. Sullivan, I. K. Toth, and M. C. M. Perombelon. 2001. Modified crystal violet pectate medium (CVP) based on a new polypectate source (Slendid) for the detection and isolation of soft rot erwinias. Potato Res. 44:265-270.
17. Jain, M., D. Ramirez, R. Seshadri, J. F. Cullina, C. A. Powers, G. S. Schulert, M. Bar-Meir, C. L. Sullivan, S. A. McColley, and A. R. Hauser. 2004. Type III secretion phenotypes of Pseudomonas aeruginosa strains change during infection of individuals with cystic fibrosis. J. Clin. Microbiol. 42:5229-5237. [PMC free article] [PubMed]
18. Jock, S., W.-S. Kim, M.-A. Barny, and K. Geider. 2003. Molecular characterization of natural Erwinia pyrifoliae strains deficient in hypersensitive response Appl. Environ. Microbiol. 69:679-682. [PMC free article] [PubMed]
19. Lehtimaki, S., A. Rantakari, J. Routtu, A. Tuikkala, J. Li, O. Virtaharju, E. T. Palva, M. Romantschuk, and H. T. Saarilahti. 2003. Characterization of the hrp pathogenicity cluster of Erwinia carotovora subsp. carotovora: high basal level expression in a mutant is associated with reduced virulence. Mol. Genet. Genomics 270:263-272. [PubMed]
20. Ma, B., M. E. Hibbing, H.-S. Kim, R. M. Reedy, I. Yedidia, J. Breuer, J. Breuer, J. D. Glasner, N. T. Perna, A. Kelman, and A. O. Charkowski. 2007. The host range and molecular phylogenies of the soft rot enterobacterial genera Pectobacterium and Dickeya. Phytopathology 97:1150-1163. [PubMed]
21. Maher, E. A., S. H. De Boer, and A. Kelman. 1986. Serogroups of Erwinia carotovora involved in systemic infection of potato plants and infestation of progeny tubers. Am. Potato J. 63:1-11.
22. Mohr, T. J., H. Liu, S. Yan, C. E. Morris, J. A. Castillo, J. Jelenska, and B. A. Vinatzer. 2008. Naturally occurring nonpathogenic isolates of the plant pathogen Pseudomonas syringae lack a type III secretion system and effector gene orthologues. J. Bacteriol. 190:2858-2870. [PMC free article] [PubMed]
23. Pitman, A. R., P. J. Wright, M. D. Galbraith, and S. A. Harrow. 2008. Biochemical and genetic diversity of pectolytic enterobacteria causing soft rot disease of potatoes in New Zealand. Aust. J. Plant Pathol. 37:559-568.
24. Posada, D., and T. R. Buckley. 2004. Model selection and model averaging in phylogenetics: advantages of akaike information criterion and bayesian approaches over likelihood ratio tests. Syst. Biol. 53:793-808. [PubMed]
25. Posada, D., and K. A. Crandall. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817-818. [PubMed]
26. Rantakari, A., O. Virtaharju, S. Vahamiko, S. Taira, E. T. Palva, H. T. Saarilahti, and M. Romantschuk. 2001. Type III secretion contributes to the pathogenesis of the soft-rot pathogen Erwinia carotovora: partial characterization of the hrp gene cluster. Mol. Plant-Microbe Interact. 14:962-968. [PubMed]
27. Ribot, E. M., C. Fitzgerald, K. Kubota, B. Swaminathan, and T. J. Barret. 2001. Rapid pulsed-field gel electrophoresis protocol for subtyping of Camplobacter jejuni. J. Clin. Microbiol. 39:1889-1894. [PMC free article] [PubMed]
28. Rojas, C. M., J.-H. Ham, W.-L. Deng, J. J. Doyle, and A. Collmer. 2002. HecA is a member of a class of adhesins produced by diverse pathogenic bacteria and contributes to the attachment, aggregation, epidermal cell killing, and virulence phenotypes of Erwinia chrysanthemi EC16 on Nicotiana clevelandii seedlings. Proc. Natl. Acad. Sci. USA 99:13142-13147. [PMC free article] [PubMed]
29. Schaad, N. W., J. B. Jones, and W. Chun. 2001. Laboratory guide for identification of plant pathogenic bacteria, 3rd ed. APS Press, St. Paul, MN.
30. Segall, R. H. 1971. Selective medium for enumerating Erwinia species commonly found in vegetable packinghouse waters. Phytopathology 61:425-426.
31. Swaminathan, B., T. J. Barrett, S. B. Hunter, and R. V. Tauxe. 2001. PulseNet: the molecular subtyping network for foodborne bacterial disease surveillance, United States. Emerg. Infect. Dis. 7:382-389. [PMC free article] [PubMed]
32. Tamura, K., and M. Nei. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10:512-526. [PubMed]
33. Thomson, S. V., D. C. Hildebrand, and M. N. Schroth. 1981. Identification and nutritional differentiation of the Erwinia sugarbeet pathogen from members of Erwinia carotovora and Erwinia chrysanthemi. Phytopathology 71:1037-1042.
34. Toth, I. K., A. O. Avrova, and L. J. Hyman. 2001. Rapid identification and differentiation of the soft rot erwinias by 16S-23S intergenic transcribed spacer-PCR and restriction fragment length polymorphism analyses. Appl. Environ. Microbiol. 67:4070-4076. [PMC free article] [PubMed]
35. Yap, M.-N., J. D. Barak, and A. O. Charkowski. 2004. Genomic diversity of Erwinia carotovora subsp. carotovora and its correlation with virulence. Appl. Environ. Microbiol. 70:3013-3023. [PMC free article] [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...