cDNA-AFLP analysis of differential gene expression in the prokaryotic plant pathogen Erwinia carotovora

Alia Dellagi1, Paul R. J. Birch1, Jacqueline Heilbronn1, Gary D. Lyon1 and Ian K. Toth1

Department of Fungal and Bacterial Plant Pathology, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, UK1

Author for correspondence: Ian K. Toth. Tel: +44 1382 562731. Fax: +44 1382 562426. e-mail: itoth{at}scri.sari.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
For studies of differential gene expression in prokaryotes, methods for synthesizing representative cDNA populations are required. Here, a technique is described for the synthesis of cDNA from the potato pathogens Erwinia carotovora subsp. atroseptica (Eca) and Erwinia carotovora subsp. carotovora (Ecc) using a combination of short oligonucleotide (11-mer) primers that were known to anneal to conserved sequences in the 3' regions of enterobacterial genes. Specific PCR amplifications with primers designed to anneal to 14 known genes from either Eca or Ecc revealed the presence of the corresponding transcripts in cDNA, suggesting that the cDNA represented a broad genomic coverage. cDNA-amplified fragment length polymorphism (cDNA-AFLP) was used to identify differentially expressed genes in Eca, including one that shows significant similarity, at the protein level, to an avirulence gene from Xanthomonas campestris pv. raphani. Northern analysis was used to confirm that differentially amplified cDNA fragments were derived from differentially expressed genes. This is the first report of the use of cDNA-AFLP to study differential gene expression in prokaryotes.

Keywords: prokaryotic differential gene expression, Erwinia, plant pathogen, cDNA-AFLP

Abbreviations: Eca, Erwinia carotovora subsp. atroseptica; Ecc, Erwinia carotovora subsp. carotovora; AFLP, amplified fragment length polymorphism

The GenBank accession numbers for the EL1, EL2, EL3, EP5, EP22, EP26, EP11 and EP21 sequences determined in this work are AJ274641AJ274648, respectively.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
The enterobacterial plant pathogens Erwinia carotovora subsp. atroseptica (Eca) and Erwinia carotovora subsp. carotovora (Ecc) are responsible for soft rot diseases in a number of crops worldwide. Eca has a host range limited to potato, causing soft rot of tubers and blackleg of stems. In contrast, Ecc has a wide host range that includes potato, but fails to cause blackleg disease (Pérombelon, 1992 ). Although our knowledge of the molecular processes involved in pathogenicity is increasing (Pérombelon & Salmond, 1995 ), little is known about mechanisms that control the onset of infection or host specificity in these pathogens. Only through a comprehensive understanding of the nature of their interactions with host plants, and the identification of genes involved in such interactions, will we be able to develop novel control strategies to protect potato production.

Traditionally, transposon mutagenesis has been used as a method to identify genes involved in bacterial processes such as pathogenicity (reviewed by Mills, 1985 ). This technique has yielded important information about a number of genes involved in disease development in Er. carotovora (e.g. genes encoding pectolytic enzymes or involved in secretory and regulatory processes; Pérombelon & Salmond, 1995 ). However, ‘hot-spots’ for transposon insertion, secondary transposition with possible genome rearrangements, low throughput and the generation of lethal mutations limit the use of this method (Mills, 1985 ). In essence, many subtle interactions between the pathogen and its host may be lost when the pathogen is disabled by mutagenesis. An alternative approach to studying plant–pathogen interactions is to investigate changes in gene expression during the interaction.

Methods for profiling gene expression, such as differential display (DD) RT-PCR (Liang & Pardee, 1992 ) and cDNA-amplified fragment length polymorphism (cDNA-AFLP; Bachem et al., 1996 ) have yielded important differentially expressed genes from eukaryotes. Both the purification of mRNA and synthesis of cDNA for these techniques involves an oligo(dT) primer that anneals to the poly(A) tail at the 3' end of the transcript. The low levels of polyadenylation in prokaryotic mRNAs mean that efficient methods for cDNA synthesis that exclude rRNA are lacking. Nevertheless, DD RT-PCR techniques have been reported for prokaryotes (Wong & McClelland, 1994 ; Abu Kwaik & Pederson 1996 ; Akins et al., 1998 ), although insufficient amplification products were generated to allow an entire genome to be screened for differentially expressed genes. Recently, a method has been developed for profiling gene expression in members of the Enterobacteriaceae that provides a broad genome coverage (Fislage et al., 1997 ). This method uses combinations of short (10-mer and 11-mer) oligonucleotide primers that anneal, respectively, to conserved sequences in the 5' and 3' regions of Escherichia coli genes. We report the use of these primers for representative cDNA synthesis from Eca and Ecc and the subsequent use of cDNA-AFLP to profile gene expression for the first time in bacteria.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bacterial strains, media and infiltration of potato leaves.
The bacterial strains used in this study were Ecc SCRI 193 (Salmond et al., 1986 ), Eca SCRI 1039 (Hyman et al., 1997 ) and Es. coli DH1. The media used were Luria–Bertani Broth (LB) (Miller, 1972 ) and minimal medium containing 1% polygalacturonate and 1% pectin (MM-pectin) (McMillan et al., 1994 ). Overnight cultures of Er. carotovora and Es. coli were diluted 20-fold in the same medium and left to grow (at 27 °C for Er. carotovora and at 37 °C for Es. coli) to OD600 ~0·4 before RNA extraction. Bacterial suspensions were prepared by centrifuging overnight cultures of Eca or Ecc and resuspending them in 10 mM MgSO4 to the required concentration. Detached potato leaves were vacuum-infiltrated with a suspension of either Eca or Ecc at 108 c.f.u. ml-1 for 15 min, and incubated for 1 h at 18 °C before freezing in liquid nitrogen and storing at -80 °C until RNA was extracted.

RNA extraction and cDNA synthesis.
Total RNA was prepared from bacterial cells as described by Aiba et al. (1981) . Prior to cDNA synthesis, RNA was treated with RNase-free DNase I (Pharmacia Biotech) (1 U DNase I:20 µg RNA) and incubated for 20 min at 37 °C. The RNA was then heat-denatured at 65 °C for 10 min, simultaneously inactivating DNase I. A mixture of all 11-mer oligonucleotide primers (Ea1–Ea10) at 100 ng µl-1 (Fislage et al., 1997 ) was used to generate first-strand cDNA from 10 µg DNase-I-treated RNA following the procedure described in the First-strand cDNA Synthesis Kit (Pharmacia Biotech). The second strand of cDNA was synthesized using the Universal RiboClone cDNA Synthesis System (Promega).

PCR amplification.
Approximately 200 ng genomic DNA and 1 µl of a 100-fold dilution of first-strand cDNA were subjected to 35 cycles of amplification by PCR. In addition to template DNA, the amplification mixture contained, in a final volume of 50 µl, 2·5 U Taq polymerase (Gibco-BRL), 10 mM Tris/HCl (pH 8·4), 50 mM KCl, optimized amounts of MgCl2 within the range 1·5–2·5 mM, 100 µmol of each deoxynucleotide triphosphate and 50 ng of each primer for a given gene (Table 1).


View this table:
[in this window]
[in a new window]
 
Table 1. Sequences of oligonucleotide primers designed to anneal to gene sequences derived from Eca and Ecc

 
cDNA-AFLP.
Double-stranded cDNA (50 ng) was incubated with 0·5 U each of EcoRI and MseI in the presence of OnePhorAll (OPA) buffer (Pharmacia Biotech) and 10 µg BSA ml-1 for 3 h at 37 °C, according to Vos et al. (1995) . The digestion products (10 µl) were incubated with 5 pmol each of an MseI-adaptor and an EcoRI-adaptor (which were annealed as described by Vos et al., 1995 ), 1 mM ATP and 1 U T4 DNA ligase (Gibco) in OPA buffer at 37 °C for 3 h. The adaptors were: 5'-CTCGTAGACTGCGTACC-3' and 3'-CTGACGCATGGTTAA-5' (EcoRI), and 5'-GACGATGAGTCCTGAG-3' and 3'-TACTCAGGACTCAT-5' (MseI). The EcoRI (Eco) primer was radiolabelled at 37 °C for 1 h. The labelling reaction contained 3·5 ng primer, 0·125 U T4 polynucleotide kinase, 0·1 µl 5xForward Reaction Buffer provided with the enzyme (Gibco), and 0·5 µCi (18·5 kBq) [{gamma}-33P]ATP, and the total volume was adjusted to 0·5 µl with sterile distilled water. Following digestion of the cDNA and ligation of EcoRI and MseI adaptors, 2 µl of the ligation mixture was used as a template for primary PCR amplification with the non-selective MseI (M00) (5'-GATGAGTCCTGAGTAA-3') and EcoRI (E00) (5'-GACTGCGTACCAATTC-3') primers of Vos et al. (1995) . The PCR contained 2·5 µl AmpliTaq LD buffer and 1 U AmpliTaq LD (Perkin Elmer), 200 µM of all four dNTPs and 50 ng of each adaptor primer, and the total volume was adjusted to 25 µl with sterile distilled water. The PCR was performed under the following conditions: 32 cycles of 30 s denaturing at 94 °C, 30 s annealing at 60 °C and 1 min extension at 72 °C. The amplification products were diluted 300-fold and 2 µl used as a template in a second PCR with selective primers with 2-base extensions. The EcoRI primers were E11 (extension AA) and E19 (GA), and the MseI primers were M14 (AT), M15 (CA), M16 (CC) and M17 (CG). The reaction contained 1 µl Perkin Elmer AmpliTaq LD buffer, 200 µM of all four dNTPs, 12·5 ng unlabelled EcoRI primer, 3·5 ng labelled primer (contained in the 0·5 µl of labelling reaction), 15 ng MseI primer and 1 U Taq polymerase (Gibco), and the total volume was adjusted to 11 µl with sterile distilled water. The selective PCR was performed under the following conditions: (i) 1 cycle of 30 s denaturing at 94 °C, 30 s annealing at 65 °C, 1 min extension at 72 °C; (ii) 11 cycles over which the annealing temperature was reduced from 72 °C by 0·7 °C each cycle; (iii) 23 cycles of 30 s denaturing at 94 °C, 30 s annealing at 56 °C and 1 min extension at 72  °C. All amplifications were performed in a PE-9600 thermocycler (Perkin Elmer). AFLP products were electrophoresed through a 6% polyacrylamide denaturing gel; the gel was then dried on Whatman paper and exposed to autoradiographic film (Kodak) to visualize results. A flow diagram of the cDNA-AFLP procedure for studying Erwinia gene expression is given in Fig. 1.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 1. The cDNA-AFLP procedure for studying Eca gene expression, adapted from Bachem et al. (1996) . Using total RNA extracted from Er. carotovora, first-strand cDNA was synthesized using a mixture of the ten 11-mer primers described by Fislage et al. (1997) (a). Second-strand cDNA was synthesized using standard protocols (b). The double-stranded cDNA was restriction-digested using a combination of a 6-bp-cutting enzyme, EcoRI (left) and a 4-bp-cutting enzyme, MseI (right) (c). Adaptors (Ec=EcoRI; M=MseI) were ligated to the ends of the molecules (d). Non-selective primers which anneal to the adaptor sequences (E00=EcoRI; M00=MseI) were used to PCR-pre-amplify the cDNA molecules (e). Specific primers which anneal to the adaptors and which possess 2 nucleotide extensions were used to PCR-amplify subsets of the cDNA fragments. The EcoRI primer was end-labelled with [33P]dATP (f). PCR amplification products were electrophoresed on polyacrylamide gel and visualized by autoradiography.

 
Isolation of amplified cDNA products.
Following development, autoradiographic films were repositioned on polyacrylamide gels and the segments corresponding to differentially amplified cDNAs were excised. The gel fragments were then heated in 100 µl sterile distilled water at 65 °C for 10 min and a PCR performed with 10 µl of the eluted DNA as template and the same primers as those used to generate the cDNA-AFLP profile. Depending on the efficiency of elution, one or two rounds of PCR were necessary to obtain sufficient DNA for sequencing (using an ABI Cycle Sequencer; Perkin Elmer) and for generating radiolabelled probes for Northern analysis (see below).

Northern hybridization.
Total RNA (10 µg) was denatured, separated by gel electrophoresis and blotted onto Amersham Hybond-N+ (Pharmacia Biotech) membranes following the procedure of Fourney et al. (1988) . Hybridization was performed with DNA probes in 5xSSPE, 0·2% SDS, 500 µg denatured herring sperm DNA (Boehringer Mannheim) ml-1 and 5xDenhardt’s solution (Sambrook et al., 1989 ). After hybridization, filters were washed twice with 5xSSC, 0·5% SDS at 65 °C for 20 min and twice with 1xSSC, 0·5% SDS at 65 °C for 20 min. Probes were generated using, as template, PCR products purified through Wizard columns (Promega) by a random-primed reaction with High Prime (Boehringer Mannheim). Unincorporated nucleotides were eliminated from the probe by purification through Sephadex Nick-Columns (Pharmacia Biotech).


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Synthesis of representative cDNA from Er. carotovora
An alternative approach to transposon mutagenesis for studying the role of genes in physiological, morphological and biochemical processes in bacteria is to investigate changes in gene expression. For any study of gene expression it is desirable to synthesize cDNA representative of all the genes that are expressed in a particular condition. RNA prepared from Eca, Ecc and Es. coli after growth to mid-exponential phase in LB medium (Miller, 1972 ) or MM-pectin medium (McMillan et al., 1994 ), was converted to first-strand cDNA using a combination of all ten 11-mer primers that anneal to conserved 3' regions in enterobacterial genes (Fislage et al., 1997 ). The media, LB and MM-pectin, were chosen because they represent different conditions for growth of Erwinia spp. and thus potentially stimulate differential gene expression. LB is a standard laboratory complete medium and pectin is a component of plant cell walls that is known to be degraded by a battery of enzymes during plant infection by Erwinia spp. (Pérombelon & Salmond, 1995 ).

To investigate whether the cDNA generated from Eca and Ecc was representative of genes expressed under the conditions used, we designed pairs of oligonucleotide primers that anneal to 14 previously reported Er. carotovora genes (Table 1) to test for the presence of these sequences in each cDNA population by PCR amplification. PCR was performed using the cDNAs as templates with each of the primer pairs in Table 1. As controls, the DNase-I-treated RNA populations used for cDNA synthesis, and Eca, Ecc and Es. coli genomic DNAs were also used as PCR templates. The results of the PCRs are given in Table 2. Fig. 2 shows a typical PCR amplification result for primers that anneal to the mopB gene. In all cases, PCR was performed on duplicate samples of independently synthesized cDNAs and gave reproducible results. All primer pairs derived from Eca gene sequences amplified DNA fragments of expected sizes (Table 1) from genomic DNA and cDNA templates of Eca (Table 2). Similarly, Ecc-derived primer pairs amplified DNA fragments of expected sizes from genomic DNA and cDNA templates of Ecc. No amplification products were generated from Es. coli cDNA or genomic DNA templates. In addition, no amplification products were obtained from DNase-treated RNA samples, demonstrating the absence of contaminating genomic DNA (results not shown). In some cases, primer pairs (Y1/Y2, Mop F/R, Ogl F/R, CelN F/R; Table 1) amplified DNA fragments of expected sizes from both Eca and Ecc cDNA templates, indicating that they annealed to conserved regions of analogous genes within each subspecies (Table 2). Interestingly, the profile of PCR amplification for the ogl gene suggested that it is differentially expressed in Ecc but not in Eca under the conditions used in this study. In addition, the rffDG operon, which was not detectable in Ecc, appeared to be differentially expressed in Eca. Successful amplification from cDNA templates with all primer pairs indicated that the combination of all ten 11-mers had resulted in the synthesis of representative first-strand cDNA. In addition, amplification of a product of over 1000 bp in size (rffDG) implied that primer extension is extremely efficient during the process of reverse transcription.


View this table:
[in this window]
[in a new window]
 
Table 2. PCR amplification from Eca and Ecc cDNA templates using primer pairs given in Table 1

 


View larger version (59K):
[in this window]
[in a new window]
 
Fig. 2. PCR amplification with primers designed to the mopB gene (see Table 1). Templates were cDNA (lanes 1), total RNA (lanes 2) or genomic DNA (lanes 3) from Eca (lanes A), Ecc (lanes B) or Es. coli (lanes C). Lane M contains a size marker (1 kb ladder; Gibco). The arrow to the left of the gel indicates the amplification product of size 656 bp.

 
To investigate the detection of Erwinia gene expression in planta, cDNA was generated from RNA extracted from leaves of the potato cv. Stirling before, and 1 h after, infiltration with a bacterial suspension of Eca SCRI 1039. The PCR results are given in Table 2. No amplification was obtained from samples derived from uninfected leaf tissue with any of the primers tested or from DNase-I-treated RNA samples (indicating the absence of contaminating genomic DNA). In contrast, cDNA samples derived from infiltrated leaves yielded clear PCR amplification products using primers designed to anneal to gene sequences derived from Eca (with the exception of rffDG), indicating the presence of the corresponding transcripts 1 h after interaction with the plant (Table 2).

cDNA-AFLP from Eca after growth in LB or MM-pectin media
To test whether the synthesized cDNA, as described above, could be used for the isolation of differentially expressed genes, we prepared templates from Eca grown on LB and MM-pectin media for cDNA-AFLP (see Fig. 1 for procedure). Until the present work, the cDNA-AFLP technique, derived from the DNA fingerprinting method of Vos et al. (1995) , had only been used for profiling differential gene expression in eukaryotes (Bachem et al., 1996 ).

First-strand cDNAs from Eca were converted to double-stranded cDNA and used to prepare cDNA-AFLP templates. An AFLP template from Eca genomic DNA was prepared for comparison. Following PCR amplifications with eight independent MseI/EcoRI primer combinations, radiolabelled PCR products were visualized by PAGE. Amplification products specific to a template, as well as many common to all templates, were observed (Fig. 3). A mean of 34±9 amplification products were detected from the genomic DNA template. In contrast, means of 26±10 and 17±3 amplification products were obtained from cDNA templates derived from LB and MM-pectin, respectively (Fig. 3). The larger number of amplification products generated from genomic DNA may have been due to either of two factors: (i) additional DNA fragments present in this AFLP template were derived from non-transcribed portions of the genome, or (ii) gene sequences were detected that were not expressed on LB or MM-pectin media. Similarly, the larger number of amplification products generated from LB-derived template than from MM-pectin-derived template (Fig. 3) may be representative of more complex gene expression on a complete medium than on a minimal medium. On three separate occasions, fresh cDNA-AFLP templates generated from independent RNA extractions, after growth of Eca on LB and MM-pectin media, yielded reproducible amplification profiles when PCR amplified with the same primer combinations (results not shown). In all cases, PCR amplification was performed using MseI and EcoRI primers with two nucleotide extensions. A total of 256 combinations of such primers can be made, suggesting that as many as 6656 amplification products may be generated from LB-derived template and 4352 from MM-pectin-derived template. Given that there are approximately 4300 expressed genes in the closely related enterobacterium Es. coli, cDNA-AFLP appears to provide a broad coverage of the Eca transcriptome. Nevertheless, it is unlikely that all amplification products are derived from independent genes or that all genes contain an EcoRI site. It is therefore advisable to employ more than one restriction enzyme combination, and to include 6-bp-cutting enzymes such as PstI, which recognizes sites with a greater G+C content and may thus, in some cases, increase the chances of digesting G+C-rich cDNA sequences.



View larger version (84K):
[in this window]
[in a new window]
 
Fig. 3. AFLP fingerprints of genomic DNA from Eca (lanes 1), cDNA from Eca after growth in MM-pectin medium (lanes 2) and cDNA from Eca after growth in LB medium (lanes 3). (a), (b) and (c) show amplifications with the E11 primer in combination with M16, M14 and M17, respectively (see Methods). The arrow points to a differentially amplified cDNA fragment, the sequence of which contains an ORF that shows significant similarity, at the protein level, to an avirulence gene from X. campestris pv. raphani (Parker et al., 1993 ).

 
Eleven differentially amplified PCR products specific to either LB or MM-pectin-derived cDNA-AFLP templates, and five commonly amplified products, were excised from polyacrylamide gels, reamplified and directly sequenced following purification. In all cases, unambiguous sequences were obtained, suggesting that only a single species of cDNA molecule was present in each excised band. Of the eleven differentially amplified cDNA fragments, five showed significant similarity to known bacterial genes (EL1, EL2, EL3, EP5 and EP22; Table 3), one matched a previously sequenced bacterial gene of unknown function (EP26) and five made no matches in international databases (results not shown) using BLAST searches (NCBI web server; Altschul et al., 1990 ). Of the cDNA fragments amplified under all conditions, four showed significant similarity to known bacterial sequences (EP11, EP21, EP12 and EP2), including two of 188 bp (EP12) and 130 bp (EP2) that showed homology to portions of Erwinia rhapontici and Er. carotovora 16S rRNA, respectively.


View this table:
[in this window]
[in a new window]
 
Table 3. Similarities at the protein level between cDNA-AFLP fragment sequences and sequences in databases

 
Northern analysis reveals that differentially amplified cDNA fragments are derived from differentially expressed genes
Three differentially amplified cDNA fragments larger than 200 bp (EL1, EL2, EL3; Table 3) and one cDNA-AFLP fragment of 118 bp (EP26) were used as radiolabelled probes to screen Northern filters of RNA extracted from Eca and Ecc grown on either LB or MM-pectin media. As a control, the 188 bp fragment of 16S rRNA sequence (a commonly amplified cDNA-AFLP band; see Table 3) was also used as a radiolabelled probe (Fig. 4). In the case of two of the larger fragments and the 118 bp fragment, clear differential hybridization was observed, which was in agreement between Eca and Ecc and which confirmed the differential amplification from Eca. The other cDNA fragment over 200 bp in size gave no differential hybridization. The Northern result for one of the cDNA fragments (EL2; Table 3), which was expressed in both Eca and Ecc specifically in LB medium but not MM-pectin medium, is shown in Fig. 4. An ORF in this sequence showed significant similarity at the protein level (6·4 e-25) with an avirulence gene from Xanthomonas campestris pv. raphani (Parker et al., 1993 ). In X. campestris pv. raphani, the product from this gene confers an avirulence phenotype in most interactions with Arabidopsis thaliana accessions. On the basis of Southern hybridization, related sequences were previously thought to be absent from Ecc (Parker et al., 1993 ). This sequence is being further investigated to see whether it also elicits an avirulence response and may thus play a role in either pathogenicity or host specificity in Er. carotovora.



View larger version (48K):
[in this window]
[in a new window]
 
Fig. 4. Northern blot hybridization of RNA isolated from Eca (lanes A) and Ecc (lanes B) after growth in LB medium (lanes 1) or MM-pectin medium (lanes 2). The two 32P-radiolabelled probes used were: (upper panel) the cDNA-AFLP fragment that was differentially amplified from template prepared after growth on LB medium but not from template prepared after growth on MM-pectin medium (indicated in Fig. 3), and which shows similarity to the X. campestris pv. raphani avirulence gene; and (lower panel) a cDNA-AFLP fragment that was amplified from templates prepared after growth on both LB and MM-pectin media and which shows strong homology to a portion of Es. coli 16S rRNA.

 
In this work, we present a method for synthesizing cDNA from the bacterial plant pathogen Er. carotovora, which gives a broad genome coverage. Such a method will allow cDNA libraries to be constructed and differential gene expression to be analysed, providing an alternative route to transposon mutagenesis for gene isolation and characterization. For example, included in the battery of genes required for plant infection by Er. carotovora are some encoding pectic enzymes that are induced only when grown in plant extract or in planta (McMillan et al., 1994 ). Differential cDNA screening methods, such as the suppression subtractive hybridization technique (Diatchenko et al., 1996 ; Birch et al., 1999 ) may allow such previously unidentified genes to be isolated.

Adaptation of bacteria to their environment can be highly efficient, involving many metabolic and physiological changes. This work shows that it is possible to reproducibly profile gene expression in a bacterial plant pathogen under different environmental conditions, and to isolate differentially regulated sequences using a modification of the cDNA-AFLP protocol of Bachem et al. (1996) .


   ACKNOWLEDGEMENTS
 
The Scottish Crop Research Institute is grant-aided from the Scottish Environment and Rural Affairs Department (SERAD). A.D. was supported by the European Union (project ERBIC-15-CT-960908).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Abu Kwaik, Y. & Pederson, L. L. (1996). The use of differential display-PCR to isolate and characterise a Legionella pneumophila locus induced during intracellular infection of macrophages. Mol Microbiol 21, 543-556.[Medline]

Aiba, H., Adghya, S. & Crombrugghe, B. (1981). Evidence for two functional promoters in intact Escherichia coli cells. J Biol Chem 256, 11905-11910.[Abstract/Free Full Text]

Akins, D. R., Bourell, K. W., Caimano, M. J., Norgard, M. V. & Radolf, J. D. (1998). A new animal model for studying Lyme disease spirochetes in a mammalian host-adapted state. J Clin Invest 101, 2240-2250.[Abstract/Free Full Text]

Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. (1990). Basic local alignment search tool. J Mol Biol 215, 403-410.[Medline]

Bachem, C. W. B., Van der Hoeven, R. S., de Bruijin, S. M., Vreugdenhil, D., Zabeau, M. & Visser, R. G. F. (1996). Visualisation of differential gene expression using a novel method of RNA fingerprinting based on AFLP: analysis of gene expression during tuber development. Plant J 9, 745-753.[Medline]

Birch, P. R. J., Avrova, A. O., Lyon, G. D., Duncan, J. M. & Toth, R. L. (1999). Isolation of potato genes that are induced during an early stage of the hypersensitive response to Phytophthora infestans. Mol Plant–Microbe Interact 12, 356-361.

De Boer, S. H. & Ward, L. J. (1995). PCR detection of Erwinia carotovora subsp. atroseptica associated with potato tissue. Phytopathology 85, 854-858.

Diatchenko, L., Lau, Y.-F. C., Campbell, A. P. & 8 other authors (1996). Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci USA 93, 6025–6030.[Abstract/Free Full Text]

Fislage, R., Berceanu, M., Humboldt, Y., Wendt, M. & Oberender, H. (1997). Primer design for prokaryotic differential display RT-PCR. Nucleic Acid Res 25, 1830-1835.[Abstract/Free Full Text]

Fourney, R. M., Miakoshi, J., Day, R. S. & Paterson, R. S. (1988). Northern blotting: efficient RNA staining and transfer. Focus 10, 5-6.

Hyman, L. J., Dewasmes, V., Toth, I. K. & Pérombelon, M. C. M. (1997). Improved detection sensitivity of Erwinia carotovora subsp. atroseptica in potato tuber peel extract by prior enrichment on a selective medium. Lett Appl Microbiol 25, 143-147.

Liang, P. & Pardee, A. B. (1992). Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 257, 967-971.[Medline]

McMillan, G. P., Barett, A. M. & Pérombelon, M. C. M. (1994). An isoelectric focusing study of the effect of methyl-esterified pectic substances on the production of extracellular pectin isoenzymes by soft rot Erwinia spp. J Appl Bacteriol 77, 175-184.

Miller, J. F. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Mills, D. (1985). Transposon mutagenesis and its potential for studying virulence in plant pathogens. Annu Rev Phytopathol 23, 297-320.

Parker, J. E., Barber, C. E., Fan, M. J. & Daniels, M. J. (1993). Interaction of Xanthomonas campestris with Arabidopsis thaliana: characterization of a gene from X. c. pv. raphani that confers avirulence to most A. thaliana accessions. Mol Plant–Microbe Interact 6, 216-224.[Medline]

Pérombelon, M. C. M. (1992). Potato blackleg: epidemiology, host–pathogen interaction and control. Neth J Plant Pathol 92, 135-146.

Pérombelon, M. C. M. & Salmond, G. P. C. (1995). Bacterial soft rots. In Pathogenesis and Host Specificity in Plant Diseases, vol. 1, Prokaryotes, pp. 1-20. Edited by U. S. Singh, R. P. Singh & K. Kohmoto. Oxford: Pergamon.

Salmond, G. P. C., Hinton, J. C. D., Gill, D. R. & Pérombélon, M. C. M. (1986). Transposon mutagenesis of Erwinia using phage lambda vectors. Mol Gen Genet 203, 524-528.

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Vos, M., Hogers, R., Bleeker, M. & 8 other authors (1995). AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res 23, 4407–4414.[Abstract]

Wong, K. K. & McClelland, M. (1994). Stress-inducible gene of Salmonella typhimurium identified by arbitrarily primed PCR of RNA. Proc Natl Acad Sci USA 91, 639-643.[Abstract]

Received 26 July 1999; revised 20 September 1999; accepted 5 October 1999.