SEARCH

SEARCH BY CITATION

Keywords:

  • nested PCR;
  • Phaseolus vulgaris;
  • real-time PCR;
  • vacuum filtration

Abstract

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

Molecular-based methods such as PCR have greatly improved detection of bacteria in environmental samples. However, the sensitivity of PCR is not high when compared to agar plating assays, and inhibitors from plants are often a problem. Pre-enriching bacteria on agar media (BIO-PCR) can increase the sensitivity of PCR by more than 100% and reduce the effects of inhibitors. To further increase the sensitivity and also reduce the labour needed for BIO-PCR, a high throughput 96-well membrane BIO-PCR technique is described for ultra-sensitive detection of Pseudomonas syringae pv. phaseolicola (PSP) (syn. P. phaseolicola) in washings of seeds and leaves of Phaseolus vulgaris, using available classical PCR primers and newly designed real-time primers and probe. The primers and probe, designed from a tox-argK chromosomal cluster of the PSP-specific phaseolotoxin gene, were confirmed to be specific to PSP. Samples (1·2 mL) were filtered under vacuum in 96-well membrane plates. After incubating on soft agar medium for 48–52 h, each well is washed with 200 µL of sterile water and used immediately for nested (two-step) PCR or real-time PCR or stored at −20°C. Results of assaying spiked seed washings showed that classical PCR was unable to detect PSP at mean concentrations of 40 colony forming units (cfu) mL−1. BIO-PCR detected PSP in five out of six samples at 40 mean cfu mL−1 but none at mean concentrations of 4·2 and 0·4 mean cfu mL−1. In contrast, membrane BIO-PCR detected the bacterium in all six samples tested containing as few as 0·4 mean cfu mL−1. The sensitivity of detection from leaf washings was lower but the results were similar, classical and BIO-PCR were negative from all three levels of inoculum while membrane BIO-PCR detected three out of three samples at 80 mean cfu mL−1 and one out of three at 40 mean cfu mL−1.


Introduction

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

Detecting bacterial pathogens in seeds and other plant materials such as symptomless leaves is often difficult because the population of the target organism is often less than other bacteria. Classical isolation methods can be very sensitive but not when samples contain few numbers of the target organism and large numbers of other bacteria (Schaad, 1989). Serological techniques are rapid and inexpensive but are generally insensitive and lack specificity (Hampton et al., 1990). With recent advances in PCR technology, considerable progress is being made in the detection of small numbers of microorganisms even in the presence of numerous other organisms. The sensitivity of classical PCR can be increased with the use of a second round of PCR using nested primers (Schaad et al., 1995). However, PCR has several limiting factors including PCR inhibitors often present in natural plant samples (Rossen et al., 1992; Prosen et al., 1993; Schaad et al., 1999; Weller et al., 2000), and relatively low sensitivity due to extremely small sample size requirements. Sensitivity can be further increased with techniques such as immunomagnetic separation (IMS-PCR) (Skjerve & Olsvik, 1991; Fratamico et al., 1992; van de Wolf et al., 2000) or enrichment PCR (BIO-PCR) (Schaad et al., 1995; Ito et al., 1998; Weller et al., 2000). Both techniques can reduce the effect of PCR inhibitors, but BIO-PCR has the added advantage of including the recovery of a viable culture of the target organism. This technique combines biological amplification on common or semiselective media (agar or liquid) with direct PCR; DNA extraction is not required, nor is a centrifugation step needed for concentrating the target. Typically, samples of 0·1 mL are plated onto each of 6–8 agar medium plates and the resulting cell growth removed from half of the plates by washing with water after 24–72 h incubation, depending upon the growth rate of the target organism. The pooled samples containing cells of the target organism are used immediately for PCR or stored at −20°C for use later. BIO-PCR has a major advantage over classical agar plating techniques. Whereas large numbers of other organisms often overgrow the target organism on agar media before the organism can be recognized, this does not happen in BIO-PCR because colonies of the target organism are harvested when they are just visible and colonies of saprophytes are relatively small. The sensitivity of BIO-PCR (normally 5–10 mean colony forming units (cfu) mL−1 of sample) can be 10 to 100-fold more sensitive than classical PCR (Schaad et al., 1995; Ito et al., 1998; Schaad & Frederick, 2002). Such sensitivity, however, is still not good enough for detecting regulated and/or quarantined plant pathogenic bacteria which are often present in very small numbers in asymptomatic tissues (Webster et al., 1983; Ozakman & Schaad, 2003).

In this study Pseudomonas syringae pv. phaseolicola (PSP) (syn. P. phaseolicola) was chosen as a model pathogen because the organism is seed-borne and there is a zero tolerance for PSP infection in bean seed lots (Webster et al., 1983). Also, classical PCR primers specific to sequences within the PSP-specific phaseolotoxin gene were available (Prosen et al., 1993; Schaad et al., 1995) for designing real-time PCR primers and probes. To further increase the sensitivity of PCR, membrane filtration has been combined with BIO-PCR in a high throughput format for ultra sensitive detection of bacteria in samples of water and leaf, and seed washings. Both classical nested PCR and a real-time PCR-based fluorescent 5′ nuclease assay system (Schoen et al., 1996; Schaad et al., 1997, 1999) were compared for detection of amplified products.

Materials and methods

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

Bacterial strains and pathogenicity tests

Bacteria used in this study and their sources are listed in Table 1. All PSP strains were tested for pathogenicity using bean (Phaseolus vulgaris cv. Hystyle) plants as described below, except the plants were at the 3–4 leaf stage and symptoms were recorded after 10 days.

Table 1.  Strains of Pseudomonas syringae pv. phaseolicola (PSP), other pathovars of P. syringae, species of Pseudomonas, and unknown saprophytic fluorescent pseudomonads isolated from bean seed used in this study and their reaction to PSP tox gene-specific real-time PCR primers and probea
StrainbSpecies/pathovarSourcecAmplificationd
  • a

    Real-time PCR run for 40 cycles.

  • b

    Number in parenthesis is the original source from which the culture was obtained.

  • c

    ICPB, International Collection of Phytopathogenic Bacteria, USDA/ARS, Ft. Detrick, MD, USA; ATCC, American Type Culture Collection, Manassas, VA, USA.

  • d

    +, Fluorescence rose above baseline between 18 and 25 cycles; –, fluorescence failed to rise above baseline after 40 cycles.

FC-22, 30P. syringae pv. phaseolicolaN.W. Schaad, ID, USA+
FC-1 (C-199)N.W. Schaad, ID, USA+
FC-42 (132)M. Ozakman, Turkey+
FC-52 (PM-180)ICPB+
FC-74 (1281A)″ race-1J. Taylor, UK+
FC-75 (882)″ race-2J. Taylor, UK+
FC-76 (1301B)″ race-3J. Taylor, UK+
FC-77 (1348B)″ race-4J. Taylor, UK+
FC-78 (34120)″ race-5J. Taylor, UK+
FC-79 (34569)″ race-6J. Taylor, UK+
FC-80 (1449B)″ race-7J. Taylor, UK+
FC-81 (2656A)″ race-8J. Taylor, UK+
FC-82 (2709B)″ race-9J. Taylor, UK+
993G. Franc, WY, USA+
FC-41″ (non-fluorescent)J. Taylor, Tanzania
FC99 (B3A)pv. syringaeS. Hutchenson, MD, USA
FC-103 (301D)pv. syringaeS. Hutchenson, MD, USA
FC-101 (SD 302)pv. syringaeS. Hutchenson, MD, USA
FC-105 (PS 17)pv. syringaeS. Hutchenson, MD, USA
FC-53 (PS 270)pv. syringaeICPB
FC-54 (PS 22)pv. tomatoICPB
FC-59 (JL1)pv. tomatoDhanbantari, Canada
FC-68 (LG-2)pv. glycineaICPB
FC-55 (PT130)pv. tabaciICPB
FC-58 (PT-5)pv. tabaciICPB
FC-127 (12623)P. putidaATCC
FC-62 (PF-107)P. fluorescensICPB
FC-120 (13525)P. fluorescens biotype AATCC
FC-23Saprophyte, bean seedsK. Mohan, ID, USA
FC-24Saprophyte, bean seedsK. Mohan, ID, USA
FC-25Saprophyte, bean seedsK. Mohan, ID, USA
FC-26Saprophyte, bean seedsK. Mohan, ID, USA
FC-27Saprophyte, bean seedsK. Mohan, ID, USA
FC-28Saprophyte, bean seedsK. Mohan, ID, USA
FC-29Saprophyte, bean seedsK. Mohan, ID, USA

Preparation of natural inoculum

To obtain natural inoculum, bean plants with several immature pods (3–7 mm long) and flowers were atomized with a suspension containing approximately 1 × 106 mean cfu mL−1 of PSP strains FC-74 or FC-1 and incubated in a lighted dew chamber (Pervical model E-54U-DL) at 18°C nights and 24°C days (12 h) until characteristic water-soaked lesions began to appear after 7–10 d on several leaves and pods. The plants were then transferred to a lighted, walk-in growth chamber set at 20°C nights and 24°C days (12 h) and with intermittent (30 s every 2 h) misting for three weeks. After an additional 30–32 d without misting, all watering ceased and the resulting mature pods with lesions were allowed to dry on the plants. Pods with dark green greasy-appearing lesions were collected and stored at room temperature in small coin envelopes. The seeds were removed from the pods, and lesion tissue cut out with a scalpel, pooled, and ground to a fine powder with a pestle in a sterile mortar. The seeds and ground tissue were stored separately in a Petri dish at room temperature until needed.

Media

King's medium B (KB) agar (King et al., 1954) and semi-selective MSP agar (Mohan & Schaad, 1987) were prepared as described. Because preliminary results showed that PSP grew more rapidly on filters placed over soft agar than those placed over standard solid agar, all media for membrane BIO-PCR were made with 1% agar.

Classical and real-time PCR primers and probe

All PCRs, regardless of the method, were performed directly without extracting DNA (direct PCR) unless stated otherwise. Classical PCR using external primers P 3·1 and P 5·1 and nested primers P 3·2 and P 5·2 resulted in the expected 0·5 kb product for classical PCR and 0·45 kb product for nested PCR (Schaad et al., 1995), respectively. PCR was performed in 25 µL using 10 mM Tris-HCl (pH 8·3), 50 mM KCl, 1·5 mM of MgCl2, 0·001% gelatin, 80 µm each of dATP, dCTP, dGTP, dTTP and 0·2 units of Taq DNA polymerase. For nested PCR, 1 µL of a 10-fold diluted product of the external PCR product was used as template. In each experiment a Southern blot using a 0·45 kb fragment of the 2·6 kb EcoRI segment was performed to confirm the identity of the nested PCR amplified product, as described by Schaad et al. (1995).

For real-time PCR, the following primers and probe were selected based upon specificity to P. syringae pv. phaseolicola: Forward primer PsF-tox 188: 5′ GGGGTGGGACGTGTTAT 3′ Reverse primer PsR-tox 557: 3′ GGCTGGCAGATTGATGTTC 5′ Probe PsF-tox 286P: 5′ ACCATCCGAATGCCAGTAATGCC 3′

These real-time primers and probe were designed from sequences within the same 2·6 kb EcoR1 fragment within the tox-argK chromosomal cluster of the des-like open reading frame (Hatziloukas et al., 1995) of the PSP-specific phaseolotoxin gene (accession U27310) from which the above PSP classical primers were designed (Schaad et al., 1995). For real-time PCR (TaqMan®), the probe was labeled with two fluorescent dyes: a reporter dye (FAM) and a quencher dye (TAMRA) (Schaad et al., 1999). When the probe is left intact in the reaction mix the quencher dye inhibits the reporter dye and there is no emission of fluorescence. If the probe hybridizes with the target DNA, the 5′ nuclease activity of the Taq polymerase cleaves the probe and frees the reporter dye. The emission of fluorescence produced by the free reporter dye is proportional to the amount of amplified DNA, and monitored in real time by the ABI 7700 Sequence Detection System (Applied Biosystems). PCR was performed in 25 µL reactions containing reagents at the following concentrations: TaqMan buffer1X, MgCl2 3 mM, dNTPs 250 µm, primers 0·4 µm, probe 0·1 µm, AmpErase 0·5 U, AmpliTaq Gold 1 U, and template, 12·8 µL. The PCR profile was: 2 min at 50°C, 10 min at 95°C and 40 cycles of 15 sec at 95°C followed by 1 min at 60°C. Results were recorded as a simple positive or as cycle threshold (Ct) values. The Ct value is defined as the PCR cycle number at which time the signal (fluorescence) of the probe raises above background; negative, no fluorescence observed.

Specificity of TaqMan real-time PCR primers and probe

To test for specificity of the newly designed primers and probe for the real-time PCR assay, cells from cultures growing overnight on KB agar plates were transferred to PCR tubes, containing 10 µL water, with a sterile tooth pick and tested by real-time PCR for 40 cycles, as described above. Sixteen strains of PSP were tested, including one of each of nine races and a non-fluorescent strain (Table 1). Strains of several closely related pathovars of P. syringae including syringae, tomato, glycinea and tabaci and distantly related pseudomonads P. putida, P. fluorescens, and seven unknown saprophytic bacteria recovered from bean seeds on KB agar were included (Table 1).

BIO-PCR

BIO-PCR was conducted as described (Schaad et al., 1995). Samples of 100 µL were spread onto each of eight KB and MSP agar plates and incubated at 28°C for 32–34 h and 48–50 h, respectively, to obtain pin point-size colonies of the target PSP organism. Results of dilution plating assays (using KB agar) of entire single colonies removed from MSP agar after 48 h at 23°C showed an average of 30 000 total cells present. Depending upon the experiment, each of five plates of each medium were washed three times with 1 mL of water and the washings pooled into a single sample of 15 mL or each plate was washed twice with 1 mL of water and the washings of each plate kept separate. The samples were kept on ice and 2·8 µL used immediately for direct PCR (without extraction of DNA) or stored at −20°C. The remaining three plates were incubated for an additional 24–48 h for determining mean cfu mL−1 and/or recovery of PSP.

Membrane BIO-PCR

A special 96-well micro titre plate (Fig. 1) fitted with surfactant-free 0·45 µ pore size cellulose nitrate filter (Sartorius Corp.) was designed to allow the membrane to be flush with the bottom of the plate and allow complete contact with the agar medium. The plates were made by special order by Corning-Costar and provided individually in a sterile package. Surfactant-free membranes were needed to prevent killing small numbers of bacteria by the surfactant (unpublished data). DNA from 1–2 dead cells cannot be detected by standard PCR but 1–2 viable cells can be detected by BIO-PCR protocols. Aliquots of 400 µL of each sample were dispensed into triplicate wells of each of two plates. The plates were then placed on a 96 well vacuum suction device (Whatman Polyfiltronics) to remove the liquid and the same operation repeated after adding two additional 400 µL aliquots so that 1·2 mL of the sample was filtered through each well. Separately, 100 µL of each dilution were spread onto each of five plates of KB and MSP agar plates and incubated under the same conditions to determine mean cfu mL−1 of each sample. After filtering the samples, one of the 96 well plates was placed onto KB or MSP soft agar in the original sterile plastic container (Fig. 1), slightly tapped to allow contact with the agar, and the lid added to keep the plate sterile. Alternately the plates were placed onto soft agar in a 20 × 40 cm Pyrex baking dish and covered with aluminum foil. After 32–34 and 48–52 h for KB and MSP, respectively, each well of the micro titre plate was washed with 200 µL of sterile water using a multi-channel micro pipette and the sample transferred to a microfuge tube for PCR or archived at −20°C. The second 96 well plate was used for membrane PCR without incubating on agar media. Each well was washed and 12·8 µL used for classical real-time PCR, as above.

image

Figure 1. Membrane plate consisting of 96-wells removed from soft agar incubation chamber. Note marks of plate on the agar.

Download figure to PowerPoint

Determination of sensitivity

Pure cultures in water

To evaluate the membrane technique for detection of PSP in water, an overnight culture growing in liquid NBY was adjusted to 0·1 OD at 600 nm and 10-fold serial dilutions made to 10−10 to obtain samples with one or fewer cfu per 100 µL. To determine actual number of cfu of PSP present, samples of 10−5, 10−6, and 10−8 dilutions were kept on ice and 100 µL aliquots of each plated onto each of three plates of KB agar. After incubating at 23°C for 55–60 h, the numbers of cfu were determined. Dilutions were stored at −20°C for PCR later. All samples were assayed by classical PCR, BIO-PCR, and membrane BIO-PCR using 12·8 µL of template. Both classical and real-time PCR were used for detection of PSP. For classical PCR, nested primers and a Southern blot were performed, as described above.

Seeds spiked with natural inoculum

For recovery of PSP from bean seed (cv. Hystyle), samples of PSP-spiked bean seed washes were used. Seed wash samples were spiked with natural PSP-inoculum obtained from dried lesions of infected pods, as described above. For sensitivity tests, a sample of 2·8 mg of the finely ground pod material was suspended in 2 mL of sterile water and kept at 4°C for 30 min. The seed pod suspension was then diluted in sterile water to 10−3, and aliquots of 100 µL of each 10-fold dilution spread onto each of three KB agar plates and incubated at 23°C to determine mean cfu mg−1 tissue. After determining the approximate mean cfu mg−1 tissue, another sample was weighed, suspended in 2 mL water, incubated for 30 min, and 100 µL of the undiluted and each of the 10-fold dilutions used to spike four samples of 250 seeds in 250 mL of cold sterile water. Non-spiked samples of 250 seeds in 250 mL of water served as a negative control. All samples were incubated at 4°C for 17 h as described by Mohan & Schaad (1987). From each sample, 100 µL aliquots of seed soak fluid were plated onto each of 20 plates of MSP. In a preliminary experiment with seed washings, samples for BIO-PCR were incubated on both KB and MSP agars. However, since BIO-PCR results were more consistent using the semi-selective MSP agar, MSP was used for BIO-PCR and KB was used for control colony counts. For BIO-PCR, after 48–52 h of incubation at 23°C, five plates were washed three times each with 1 mL of sterile water and then pooled for PCR. The other 15 plates were incubated for 72 h at 23°C to determine mean cfu mL−1. For membrane BIO-PCR, three times 400 µL of each sample of seed soak fluid (1·2 mL total) were filtered into triplicate wells of each of two 96-well plates, as described above. After incubating for 48–52 h on MSP soft agar, each well of the micro titre plate was washed with 200 µL of sterile water as above and used immediately for classical and real-time PCR or stored at −20°C.

Leaves

To evaluate the detection of PSP from plant leaf surfaces, the recovery of the bacterium from bean leaves was tested using real-time PCR, BIO-PCR and membrane BIO-PCR. To obtain leaf samples with PSP, six bean plants at the 2–3 leaf-stage, for each treatment, were sprayed with a dilution of 10−4, 10−5 and 10−6 of a pure culture of PSP prepared as above for recovery from water. Six 100 µL aliquots of each dilution were plated onto KB agar plates and incubated for 48 h at 23°C to determine the mean cfu mL present in the inoculum used to spray the leaves. After 2–3 h, when the leaves were dry, one leaf from each of six plants of each treatment (different dilutions) was collected in a quart size ‘Zip Lock’™ plastic bag containing 50 mL of washing buffer (0·1 M KH2PO4, 0·1% Bacto peptone, 0·1% Tween 20, pH 7) and sonicated (low power) for 10 min. One millilitre aliquots of sonicated leaf washings were kept at −20°C for real-time PCR. For BIO-PCR, 100 µL aliquots of sonicates were plated onto each of eight MSP agar plates and incubated at 23°C. For each of the six leaf samples, five of the plates of MSP were washed after 50–52 h, respectively and assayed by real-time PCR. The three other plates (controls) were incubated for 3 days to determine the mean cfu mL−1 present in the leaf washing. For membrane BIO-PCR, 1 mL of sonicate from each of the six samples was filtered on each of three wells of two 96-well membrane plates and one plate incubated on soft MSP agar for 2 days and the other for 3 days, as above. Each well was then washed with 100 µL of sterile water and the washings stored at −20°C. For PCR, the six membrane washings were diluted to10−2 and 12·8 µL assayed by real-time PCR, as above.

Results

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

Pathogenicity

All strains of PSP, including non-fluorescent FC-41, produced water-soaked lesions surrounded by a yellow halo after 7–10 days.

Specificity of real-time PCR primers and probe

Of the 16 strains of PSP tested, all were positive using the newly designed tox gene-specific real-time PCR primers and probe except strain FC 41, while all 13 other pseudomonads were negative, as were six unidentified saprophytic bacteria morphologically similar to PSP isolated from beans (Table 1).

Determination of sensitivity

Pure cultures in water

Membrane BIO-PCR resulted in a significant increase in sensitivity over BIO-PCR in detecting samples of pure cultures containing extremely small numbers of PSP. In the first series of experiments, with samples containing a mean of 17 cfu mL−1 (10−8 dilution) PSP was detected by BIO-PCR and membrane BIO-PCR but not by classical or membrane PCR using ethidium bromide stained gels and Southern analysis. With samples containing 12 or fewer cfu mL−1 (10−8 and 10−9 dilutions), membrane BIO-PCR detected 13 of 16 samples whereas BIO-PCR detected PSP in only 14 of 24 samples. In a later experiment with samples containing a mean of 12 cfu mL−1, PSP was detected by membrane BIO-PCR using both real-time and classical detection techniques.

Seeds spiked with natural inoculum

Membrane BIO-PCR (96-well plate) combined with nested PCR and Southern blot analysis (Tables 2 & 3) or real-time PCR protocols (Table 3) were able to detect samples containing a mean of 4·2 cfu mL−1 or less. All other methods were negative at such low concentrations.

Table 2.  Comparison of sensitivity of detection of Pseudomonas syringae pv. phaseolicola (PSP) in spiked bean seed washings using direct PCR, BIO-PCR, and 96-well membrane BIO-PCR by classical PCRa
PSP, mean cfu mL−1DirectMembraneBIO-PCRf
PCRcBIO-PCRe
  • a Washings of samples of 250 bean seeds soaked in 250 mL water for 2 h were spiked with dilutions of PSP from powdered infected bean pods. Actual colony forming units per mL (cfu mL−1) determined by spreading 100 µL onto each of five KB agar plates. All assays done using classical nested PCR with tox primers.

  • b

    Negative control consisted of soaking 250 seeds without adding PSP.

  • c

    Sample is 12·8 µL of seed washings used directly as PCR template.

  • d

    Number samples positive/number samples tested.

  • e

    Sample of 12·8 µL of plate washing obtained as follows: for each dilution 100 µL of seed washing were spread on each of five plates of MSP agar; the plates were incubated for 2 days and washed three times with 1 mL of sterile water; 12·8 µL of the pooled 15 mL were used as template in each PCR reaction.

  • f

    Sample of 12·8 µL of well washing obtained as follows: three times 400 µL of each of four seed washing samples (1·2 mL total) were filtered in triplicate wells of each of two membrane plates (six wells per dilution). After incubating for 2 days on soft MSP agar, each well was washed with 200 µL of sterile water and 12·8 µL of each of the well washings was used as template in each PCR reaction.

Negative Controlb0/2d0/60/3
400/35/66/6
4·20/30/56/6
0·40/30/36/6
Table 3.  Comparison between the sensitivity of direct PCR, BIO-PCR, and membrane BO-PCR for detection of Pseudomonas syringae pv. phaseolicola (PSP) in spiked washings of bean seeds by classical nested and real-time PCRa
TreatmentDirect PCRc,d,eBIO-PCRfMembrane BIO-PCRg
Real-timeClassical nestedReal-timeClassical nestedReal-timeClassical nested
  • a Washings of samples of 250 bean seeds soaked in 250 mL water for 2 h were spiked with dilutions of PSP prepared from powdered infected bean pods. Actual colony forming units per mL (cfu mL−1) determined by spreading 100 µL onto each of five KB agar plates.

  • b

    Negative controls consisted of soaked seed without adding PSP.

  • c

    +, Fluorescence rose above background within 40 cycles; –, fluorescence failed to rise above background within 40 cycles.

  • d

    Sample of 12·8 µL of seed washings used directly as PCR template without extracting DNA.

  • e

    +, 0·45 kb DNA band observed; –, no DNA band observed.

  • f

    Sample of 12·8 µL of agar plate washing obtained as follows: for each dilution 100 µL of seed washing were spread on each of five plates of MSP, the plates were incubated for 2 days and washed three times with 1 mL of sterile water, 12·8 µL of the 15 mL pooled sample were used as template in each PCR reaction.

  • g

    Sample of 12·8 µL of microplate well washing obtained as follows: three times 400 µL of each seed washings (1·2 mL total) were filtered in a well of the membrane plate (six wells per dilution), the plate was then incubated for 2 days on soft MSP agar and each well washed with 200 µL of sterile water and 12·8 µL of the well washing was used directly as PCR template without extracting DNA.

Negative controlbe
910 cfu/ml−1+++++
26 cfu/ml−1+++
2·0 cfu/ml−1++
<1·0 cfu/ml−1+
Detection from leaves

The mean concentration of inoculum present in the three leaf washing treatments, 10−4, 10−5, and 10−6, was 5500, 500, and 55 cfu mL−1, respectively. The inoculum recovered in the leaf washings for each respective treatment was 80, 40, and zero mean cfu mL−1, respectively. The classical PCR and BIO-PCR assays were negative for PSP for all three treatments. In contrast, for membrane real-time BIO-PCR three out of three samples for treatment 1 (80 mean cfu mL−1) were positive for PSP after 2 days of incubation with Ct values of 27·4, 29·8, and 33·1. After 3 days of incubation, Ct vales were 28·5, 29·7, and 29·7. For treatment 2 (40 mean cfu mL−1), one of three was positive after 2 and 3 days with a Ct values of 34·4 and 31·3, respectively. For the third treatment (zero mean cfu mL−1) all samples were negative after both 2 and 3 days incubation.

Discussion

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

Pseudomonas syringae pv. phaseolicola is a highly regulated seed-borne pathogen (Grogan & Kendrick, 1967; Webster et al., 1983). Several techniques for detecting PSP in seeds are available, including isolation on semi-selective agar media (Mohan & Schaad, 1987), serology (van Vuurde & van den Bovenkamp, 1989), and PCR (Prosen et al., 1993; Audy et al., 1995; Schaad et al., 1995). Membrane BIO-PCR using classical nested PCR or real-time PCR worked as well for ultra sensitive detection of PSP in seed washings as in pure culture in water. Use of bean seed washings containing known numbers of PSP showed the greatest sensitivity when the 96 well membrane BIO-PCR technique was combined with classical nested PCR and a Southern blot analysis. A detection limit of 26 mean cfu mL−1 by BIO-PCR is in agreement with the sensitivity originally reported for BIO-PCR to detect PSP in commercial seed lots (Schaad et al., 1995). Also, the results agreed with previous preliminary results showing the detection of 1–3 cfu of PSP per millilitre of pure culture using membrane BIO-PCR when combined with nested PCR and Southern blot analysis (Schaad et al., 1996).

If such ultra sensitivity is not needed, the more rapid and less costly real-time PCR technique using multi-well plates and the high throughput ABI 7700 Sequence Detection System is the method of choice. As expected, the target pathogen was not detected at concentrations of 2 mean cfu mL−1 by standard BIO-PCR (no membrane) even with nested primers and a Southern blot. However, PSP was detected at 2 mean cfu mL−1 with membrane BIO-PCR regardless of the detection method used to detect the amplified product. To detect such small numbers of cells it is necessary to use surfactant-free membranes because small numbers of cells are killed by the surfactant (N.W. Schaad, unpublished data). Such sensitivity is a 10-fold increase over the detection obtained by BIO-PCR and nested primers and nearly a 500-fold increase over BIO-PCR and real-time PCR. When large numbers of samples are to be assayed, the 96-well plate allows two replications of 40–45 samples, including controls.

Real-time PCR has many advantages over classical nested PCR, including speed and freedom from contamination (Schaad & Frederick, 2002). The higher cost of a fluorescent-based detection system is more than offset by the significant savings in labour costs resulting from the additional 6–8 h needed to do the classical assay. Real-time PCR has other advantages including being a closed system and providing quantifiable results (Schaad & Frederick, 2002). A disadvantage of the real-time PCR system was the need to dilute the natural bean seed washes 10-fold or more; cleavage of the amplified product was apparently inhibited by plant materials in the seed wash. This thereby prevented fluorescence from occurring and certainly contributed to the lower sensitivity observed with real-time PCR over classical nested PCR. Both the PSP-specific tox-argK gene cluster (Hatziloukas et al., 1995) classical primers (Schaad et al., 1995) and the newly designed real-time primers and probe are highly specific but have a disadvantage in sensitivity since the target sequence exists as a single copy. The real-time primers and probe maintained the specificity of the classical primers (Schaad et al., 1995). However, the failure of the non-fluorescent strain to react suggests the primers and probe may be too specific. Also, a clonal group of nontoxigenic (lack argK gene cluster) strains of PSP has recently been reported in Spain (Rivas et al., 2005; Rico et al., 2003). Although nontoxigenic and nonfluorescent strains are not commonly observed (Schaad et al., 1995) and not generally available in international culture collections, such strains may occur in small numbers in field samples and simply not be recognized. This suggests that a multiplex PCR assay including primers and probe to the common Tox positive strains and less common Tox- and nonfluorescent strains may be needed.

Acknowledgements

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

We thank John Taylor, Steve Hutchenson, and Bobberjung Dhanbantari for providing cultures and Vern Damsteegt and Reid Frederick for reviewing the manuscript. We thank Rogers Bros. Seed Co, Twin Falls, Idaho for supplying healthy bean seeds.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Audy P, Baraat CE, Saindon G, Huang HC, Laroche A, 1995. A rapid and sensitive PCR-based assay for concurrent detection of bacteria causing common and halo blights in bean seeds. Phytopathology 86, 3616.
  • Fratamico PM, Schultz FJ, Buchanan RL, 1992. Rapid isolation of Escherichia coli 0157:H7 from enrichment cultures of foods using an immunomagnetic separation method. Food Microbiology 9, 10513.
  • Grogan RG, Kendrick KA, 1967. The role of seed contamination in the transmission of Pseudomonas phaseolicola in Phaseolus vulgaris. Phytopathology 57, 2831.
  • Hampton R, Ball E, De Boer S, eds, 1990. Serological Methods for Identification and Detection of Viral and Bacterial Plant Pathogens . A Laboratory Manual. St. Paul, MN, USA: APS Press.
  • Hatziloukas E, Panopoulos NJ, Delis S, Prosen D, Schaad NW, 1995. An open reading frame in the 28 kb tox-argK gene cluster encodes a polypeptide with homology to fatty acid desaturases. Gene 166, 837.
  • Ito S, Ushijima Y, Fujii Y, Tanaka S, Kameya-Iwaki M, Yashiwaru S, Kishi F, 1998. Detection of viable cells of Ralstonia solanacearum in soil using a semi-selective medium and a PCR technique. Journal of Phytopathology 146, 36984.
  • King EO, Ward MK, Raney DE, 1954. Two simple media for the demonstration of pyocyanin and fluorescence. Journal of Laboratory Clinical Medicine 44, 3017.
  • Mohan K, Schaad NW, 1987. An improved agar plating assay for detecting Pseudomonas syringae pv. syringae and Pseudomonas syringae pv. phaseolicola in contaminated bean seed. Phytopathology 77, 13905.
  • Ozakman M, Schaad NW, 2003. A real-time BIO-PCR assay for detection of Ralstonia solanacearum race 3, biovar 2, in asymptomatic potato tubers. Canadian Journal of Plant Pathology 25, 2329.
  • Prosen D, Hatziloukas E, Schaad NW, Panopoulas NJ, 1993. Specific detection of Pseudomonas syringae pv. phaseolicola DNA in bean seed by polymerase chain reaction-based amplification of a phaseolotoxin gene region. Phytopathology 83, 96570.
  • Rivas LA, Mansfield JW, Tsiamis G, Jackson RW, Murill J, 2005. Changes in race-specific virulence in Pseudomonas syringae pv. phaseolicola are associated with a chimeric transposable element and rare deletion events in a plasmid-borne pathogenicity island. Applied and Environmental Microbiology 71, 377885.
  • Rico AR, Lopez R, Asensio C, Aizpun MT, Manzanera MCAS, Murillo J, 2003. Nontoxigenic strains of Pseudomonas syringae pv. phaseolicola are a main cause of halo blight of beans in Spain and escape current detection methods. Phytopathology 93, 15539.
  • Rossen L, Norskov P, Holmstron K, Rasmussen OF, 1992. Inhibition of PCR by components of food samples, microbial diagnostic assays and DNA-extraction solutions. International Journal of Food Microbiology 17, 3745.
  • Schaad NW, 1989. Detection and identification of bacteria. In: SaetterAW, SchaadNW, RothD, eds. Detection of Bacteria in Seeds. St. Paul MN, USA: APS Press, 916.
  • Schaad NW, Frederick RD, 2002. Real-time PCR and its application for plant disease diagnosis. Canadian Journal of Plant Patholology 24, 2508.
  • Schaad NW, Cheong SS, Tamaki S, Hatziloukas E, Panopoulos NJ, 1995. A combined biological and enzymatic amplification (BIO-PCR) technique to detect Pseudomonas syringae pv. phaseolicola in bean seed extracts. Phytopathology 85, 2438.
  • Schaad NW, Haziloukas E, Henricksen G, 1996. Development of membrane BIO-PCR for ultra-sensitive detection of pathogens in environmental samples. In: ASM General Meeting, New Orleans, LA, USA. Washington DC, USA: American Society for Microbiology, 391.
  • Schaad NW, Berthier-Schaad Y, Hatziloukas E, Knorr D, 1997. Development and comparison of ABI prism TM7700 sequence detection system to BIO-PCR for sensitive detection of Pseudomonas syringae pv. phaseolicola. In: ASM Conference on Molecular Diagnostics and Therapeutics, Kananaskis, AL, Canada. Washington DC, USA: American Society for Microbiology, 14.
  • Schaad NW, Berthier-Schaad Y, Sechler A, Knorr D, 1999. Detection of Clavibacter michiganensis subsp. sepedonicus in potato tubers by BIO-PCR and automated real-time fluorescence detection system. Plant Disease 83, 1095100.
  • Schoen CD, Knorr D, Leone G, 1996. Detection of potato leaf roll virus in dormant potato tubers by immunocapture and fluorogenic 5’ nuclease RT-PCR assay. Phytopathology 86, 9939.
  • Skjerve E, Olsvik O, 1991. Immunomagnetic separation of Salmonella from foods. International Journal of Food Microbiolology 14, 1118.
  • Van Der Wolf JM, Vriend SGC, Kastelein P, Nijuis EH, Van Bekkum PJ, Van Wuurde JWL, 2000. Immunofluorescense colony-staining (IFC) for detection and quantification of Ralstonia (Pseudomonas) solanacearum biovar 2 (race 3) in soil and verification of results by PCR and dilution plating. European Journal of Plant Patholology 106, 12333.
  • Van Vuurde JWL, Van Den Bovenkamp GW, 1989. Detection of Pseudomonas syringae pv. phaseolicola in beans. In: SaetterAW, SchaadNW, RothD, eds. Detection of Bacteria in Seeds. St. Paul MN, USA: APS Press, 3040.
  • Webster DM, Atkin JD, Cross JE, 1983. Bacterial blights of snap beans and their control. Plant Disease 67, 9359.
  • Weller SA, Elphinstone JG, Smith NC, Boonham N, Stead DE, 2000. Detection of Ralstonia solanacearum strains with a quantitative, multiplex, real-time, fluorogenic PCR (TaqMan) assay. Applied and Environmental Microbiology 66, 28538.