SEARCH

SEARCH BY CITATION

Keywords:

  • Botrytis cinerea;
  • rpoB;
  • star-shaped stoma;
  • Vitis vinifera;
  • xylem

Abstract

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

Molecular sequencing (rpoB) and standard pathological and microbiological methods identified Pseudomonas syringae pv. syringae (Pss) as the causal agent of bacterial inflorescence rot of grapevines (Vitis vinifera) in three vineyards in Tumbarumba, NSW, Australia in 2006 and 2007. Pss strains from shrivelled berries and necrotic inflorescences of diseased grapevines were used to inoculate leaves and inflorescences of potted cv. Semillon grapevines. Pss caused disease symptoms similar to those experienced in the field, including angular leaf lesions, longitudinal lesions in shoot tissues and rotting of inflorescences from before flowering until shortly after fruit set. High humidity promoted symptom severity. The necrotic bunch stem and leaf lesions were susceptible to the development of Botrytis cinerea infections. Cryo-scanning electron microscopy (cryoSEM) indicated that Pss entered leaves and inflorescence tissues via distorted, open, raised stomata surrounded by folds of tissue that appeared as ‘star-shaped’ callose-rich complexes when viewed by UV light microscopy. In necrotic tissues, cryoSEM revealed Pss within petiole parenchyma cells and air-filled rachis xylem vessels. This is the first report of inflorescence and hence fruit loss caused by Pss in grapevines. The disease is described as ‘bacterial inflorescence rot’ and regarded as one that expands the previously reported pathology of grapevines caused by P. syringae. This study also indicated that infection by Pss might promote destructive B. cinerea infections when the fungus is already present but latent, although further experimentation is needed to prove such an interaction.


Introduction

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

In the springs of 2006 and 2007, previously unreported disease symptoms, including decaying inflorescences and necrotic lesions in the leaves and shoots, were observed in Vitis vinifera cvs Sauvignon Blanc, Pinot Noir, Chardonnay and Riesling in vineyards in Tumbarumba, a cool viticultural region in southern New South Wales, Australia. In some vineyards, losses of inflorescences reduced fruiting by up to 60%. The region is prone to spring frosts, and water sprinklers, commonly used to prevent foliar damage, promote very humid foliar microclimates. Laboratory examination of diseased shoots, leaves, inflorescences and nascent berries revealed large numbers of oxidase-negative fluorescent pseudomonads.

The first record of Pseudomonas syringae (pathovar unspecified) on grapevines (V. vinifera cv. Cereza) occurred in Argentina where it caused necrotic lesions in leaf blades, veins, petioles, shoots, rachis and tendrils, but was considered a weak pathogen (Klingner et al., 1976). Since then there have been more reports of P. syringae in V. vinifera cultivars, viz. bark necrosis in cv. Vernaccia in Sardinia (Cugusi et al., 1986); bacteriosis in Azerbaijan (Samedov et al., 1988); and angular leaf lesions in cvs Verdelho, Merlot, Cabernet Sauvignon, Viognier, Sauvignon Blanc and Chardonnay in Australia (Hall et al., 2002).

Hall et al. (2002) reported that bacterial leaf spot of V. vinifera was first observed during wet spring weather in the northern Adelaide Hills in October 2000. Bacterial ooze was observed when necrotic tissue was cut under water and the bacteria isolated were identified as P. syringae (pv. unspecified) on the basis of fatty acid methyl ester analysis (FAME) and lodged in the Australian Collection of Plant Pathogenic Bacteria (ACPPB) as DAR 73915 and DAR 75241. After another wet spring in 2001, infection was more severe than in 2000 and P. syringae was also recovered from stem lesions, but there was no apparent effect on fruit set or yield. Pseudomonas syringae infections were also observed in 10 more vineyards in the same district and in the Southern Vales region, 50 km south of Adelaide. Given the similarities of the South Australian and Tumbarumba leaf and shoot symptoms, P. syringae strains DAR 73915 and DAR 75241 were later acquired for comparison with the strains isolated in Tumbarumba.

The purpose of this study was to identify and characterize, using molecular and standard pathological and microbiological methods, bacteria isolated from the diseased grapevine tissue, and to establish their pathogenicity towards V. vinifera under conditions of high and low humidity. The distribution of the bacteria on and within the leaf and inflorescence tissues and the morphology of stomata, putative points of bacterial entry to the plants, were examined by light and electron microscopy.

Materials and methods

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

Vineyard symptoms and bacterial isolation

In December 2006 and 2007, around 30 days after flowering, six healthy and six necrotic inflorescences were collected from each of three Tumbarumba vineyards with the following disease symptoms: necrotic leaf spots with yellow haloes and angular lesions delineated by veins; dark longitudinal lesions on petioles and rachises; and brown ‘soggy’ abscised necrotic inflorescences (Fig. 1).

image

Figure 1.  Symptoms of bacterial inflorescence rot in commercial vineyards from which Pseudomonas syringae pv. syringae was isolated in 2006. (a) Sauvignon Blanc detached necrotic inflorescence (dni) and bacterial ooze on the leaf petiole. (b) Pinot Noir leaf with splits through necrotic lesions in blade, dark longitudinal lesions (L) on petiole and shoot and detached necrotic inflorescence (dni). (c) Leaf with dark spots (s) with yellow chlorotic halos; and angular necrotic lesions delineated by veins resulting in leaf splits. (d) TEM of bacteria in naturally infected vines in vineyard; main photo: bacteria in both transverse and longitudinal section, within decayed phloem companion cell, Sauvignon Blanc rachis; inset 1: bacteria inside cortical cell of inflorescence pedicel, Sauvignon Blanc; inset 2: extracellular bacteria between two cortical cell walls, Pinot Noir lignified shoot. Scale bars = 1 μm; b = bacterium.

Download figure to PowerPoint

The vineyards, located within 10 km of each other, had been selected as typical of the Tumbarumba region, being of similar elevation and producing grapes for premium table wines and/or sparkling wines. Tumbarumba (148°00′44″E, 35°46′41″S; elevation 645 m) has a mild and semihumid climate (20·1°C, 1318 growing degree days and 492 mm rainfall from October to April). Disease severity was very high: pre-bunch counts had been undertaken and it was estimated that 60% of inflorescences had been lost by December in 2006. Similar symptoms had been reported from these and other Tumbarumba vineyards for around 5 years prior to this study, but a physiological cause had been suspected previously.

The plant samples were surface-sterilized with calcium hypochlorite solution (1% Cl) for 3 min, rinsed three times with sterile deionized water (SDW) and placed on dichloran rose-bengal chloramphenicol agar (DRBC; Oxoid); nutrient agar (Oxoid) with benomyl (30 μg mL−1) (NA-B); or pseudomonas-selective CCF agar (PS; Oxoid), and incubated in darkness at 25°C. Bacterial ooze from diseased shoots and rachises was also streaked without further treatment onto DRBC, NA-B or PS. Bacteria were streaked onto King’s B agar (KB) and tested for Gram stain, motility and fluorescence on PS under UV light (365 nm).

Two representative cv. Sauvignon Blanc bacterial strains, DAR 77819 (from an abscised 4-mm-diameter shrivelled berry) and DAR 77820 (from a necrotic rachis), collected in December 2006, were selected for identification by the biolog microlog 1 Version 4.20 bacterial identification system and FAME analysis (Sherlock Microbial Identification System, Version 4.0, MIDI Inc.).

All the 2006 Tumbarumba grape strains (i.e. 18 from six diseased inflorescences collected from each of three vineyards), plus two cultures from 2007 [MW844 (diseased inflorescence) and MW850 (dormant bud)]; and Australian P. syringae strains from the NSW Industry & Investment (NSW I&I) culture collection: P. syringae pv. syringae (Pss) DAR 73915, DAR 75527 and DAR 75241 (V. vinifera leaves, Adelaide Hills, Australia) and DAR 72042 (apple, Batlow, NSW, Australia); and Pseudomonas syringae pv. morsprunorum (Psm) DAR 33417, DAR 33418 and DAR 33419 [Prunus avium (wild cherry) leaves, Armidale, NSW, Australia) were tested for LOPAT (levan, oxidase, potato rot, arginine dihydrolase and tobacco leaf hypersensitivity reaction (HR). Bacteria positive for levan and tobacco leaf HR but negative for the other tests (i.e. + −−− +) were identified as P. syringae (Lelliott & Stead, 1987). The ability to induce HR on tobacco leaves was tested by infiltrating the second and third fully expanded leaves of tobacco (Nicotiana tabacum) with bacterial suspensions (5 × 108 cells mL−1). The inoculation site was punctured with a 26-gauge needle and inoculum was introduced with a needleless syringe held perpendicular to the leaf. SDW and an avirulent strain of P. fluorescens were injected in the same way.

The strains were also tested for gelatin liquefaction and aesculin hydrolysis to distinguish pathovars within P. syringae. Strains positive or negative for both tests were identified as Pss or Psm, respectively (Latorre & Jones, 1979; Table 1).

Table 1.   Characteristics of Tumbarumba grape (Vitis vinifera) bacterial inflorescence necrosis strains compared with those of Australian Pseudomonas syringae pv. syringae (Pss) and Pseudomonas syringae pv. morsprunorum (Psm) strains
StrainHostOriginPathogenic to grape leafaGelatin liquefactionbAesculin hydrolysisb16S rRNArpoBGenBank accession number
NucleotidesAmino acids
  1. aNecrotic lesions on attached young Chardonnay leaves after inoculation with 100 μL of 105 cells mL−1. No hypersensitivity reactions were observed. Two days after inoculation (a.i.), brown spots appeared along the veins, progressing from the point of inoculation to the (necrotic) leaf extremities. Seven days a.i. the leaf blades were covered in dry angular necrotic leaf lesions which often split, leaving holes. At 13 days a.i. all the inoculated leaves had sporulating Botrytis cinerea infection and had senesced.

  2. bAll strains were LOPAT (group Ia, + −−− +) and thus identified as P. syringae (Lelliott et al., 1966). Strains positive or negative for both gelatin liquefaction and aesculin hydrolysis tests were identified as Pss or Psm, respectively (Latorre & Jones, 1979).

  3. nt: nucleotide.

DAR 77819V. vinifera necrotic rachisTumbarumba, NSW, Australia, 2006+++P. syringaePssPssHM370531
DAR 77820V. vinifera shrivelled berryTumbarumba, NSW, Australia, 2006+++P. syringae (identical to DAR 77819)Pss (identical to DAR 77819)(identical to DAR 77819)HM370532
MW953Potted V. vinifera Semillon, shoot lesionPathogenesis study (Koch’s postulates)+++P. syringae (identical to DAR 77819)Pss (identical to DAR 77819)(identical to DAR 77819)HM370533
MW844V. vinifera necrotic rachisTumbarumba, NSW, Australia, 2007+++P. syringae (identical to DAR 77819)Pss (1/728 nt different from DAR 77819)(identical to DAR 77819)HM370534
MW850V. vinifera dormant budTumbarumba, NSW, Australia, 2007+++P. syringae (1/666 nt different from DAR 77819)Pss (3/728 nt different from DAR 77819)(1 aa different from DAR 77819)HM370535
DAR 73915V. vinifera leavesAdelaide Hills, South Australia, 2002+++P. syringae (1/666 nt different from DAR 77819)Pss (1/728 nt different from DAR 77819, identical to MW844)(identical to DAR 77819)HM370536
DAR 72042Apple leavesBatlow, NSW, Australia, 1997+++P. syringae (identical to DAR 77819)Pss (4/728 nt different from DAR 77819)(1 aa different from DAR 77819)HM370537
DAR 33419Prunus sp. (wild cherry) leavesArmidale, NSW, Australia, 1975+P. syringae (1 nt different from DAR 77819)Psm (28/728 nt different from DAR 77819)(identical to DAR 77819)HM370538

The bacterial strains from Tumbarumba were tested for ice nucleation activity by the method of Lindow et al. (1978). Briefly, ten 10-μL drops of cell suspension (108 cells mL−1 in SDW), prechilled at 5°C for 30 min, were applied to the surface of 3- × 3-cm squares of laboratory film within aluminium foil boats floating on a supercooled (−6°C) water/ethanol ice bath. Droplets of SDW and non-ice-nucleating P. fluorescens were included as controls.

Analysis for production of coronatine (a phytotoxin produced by some P. syringae pathovars; Melotto et al., 2006) by the two representative Tumbarumba strains (DAR 77819 and DAR 77820) was conducted by HPLC after extraction with acidified ethyl acetate and dissolution in 0·05% v/v trifluoroacetic acid and 10% v/v acetonitrile in water (Palmer & Bender, 1993). HPLC was conducted on a Waters 2690 separation module run by millenium software and connected to a Waters 2996 photodiode array detector operating over the range of 200–600 nm. The column was a reversed phase Varian microsorb-MV 100 C8 of particle size 5 μm and 250 × 4·6 mm. Samples were analysed in 200-μL volumes in a 250-μL injection loop with a flow rate of 2·0 mL min−1 and column temperature of 40°C. Solvent A was 0·05% trifluoroacetic acid in water and solvent B was acetonitrile. A gradient program was employed whereby solvent A and cumulative time were 90% at 0 min, 50% at 10 min, 50% at 15 min, 90% at 17 min and 90% at 22 min.

Molecular characterization of Pseudomonas strains

Direct sequencing of partial 16S rRNA (ribosomal RNA gene) and rpoB (β-subunit of RNA polymerase) genes (Tayeb et al., 2005) was performed on the two representative Tumbarumba strains (DAR 77819 and DAR 77820), NSW I&I strains (Pss DAR 72042, Pss DAR 73915, and Psm DAR 33419), P. syringae isolated in 2007 from a rachis and a dormant bud in Tumbarumba vineyards (MW844 and MW850, respectively), plus samples representative of those reisolated from surface-sterilized pot-experiment shoots to confirm Koch’s postulates (MW953) (Table 1).

Genomic DNA was extracted from a 48-h KB culture derived from a single colony and then PCR-amplified using a REDExtract-N-Amp PCR kit (Sigma-Aldrich). The two genes (16S rRNA and rpoB) were PCR-amplified using an Applied Biosystems 9700 thermal cycler.

The 16S rRNA primers were 27f (5′-AGAGTTTGATCMTGGCTCAG) and 765r (5′-CTGTTTGCTCCCCACG) (Lane, 1991). The cycle conditions were: 94°C for 3 min; 35 cycles of 94°C for 1 min, 47°C for 1 min and 72°C for 2 min; a single cycle of 47°C for 1 min; and a final extension step of 72°C for 10 min.

The rpoB PCR primers were LAPS (5′-TGGCCGAGAACCAGTTCCGCGT) and LAPS27 (5′-CGGCTTCGTCCAGCTTGTTCAG). The cycle conditions were: 94°C for 3 min; 40 cycles of 94°C for 45 s, 55°C for 1 min and 72°C for 90 s; and a final extension step 72°C for 10 min. PCR products were purified with a QIAquick PCR Purification kit (Qiagen) following the manufacturer’s directions and were sequenced, in both directions, by Australian Genome Research Facility Ltd.

The sequence data of approximately 666 bp (16S rRNA) and 728 bp (rpoB) were compared visually and aligned using the multiple alignment software clustalw. Chromatograms of DNA sequences were analysed for sequence conformity using chromas lite version 2.01. Phylogenetic trees were generated by upgma, using Jukes-Cantor corrected distances (Kumar et al., 2004); evaluation of statistical confidence for sequence groups was by bootstrap test (1000 pseudoreplicates) (Felsenstein, 1985).

In order to allocate the various strains to known molecular groups, rpoB sequences for Pss and Psm were downloaded from the GenBank database and included in a phylogenetic tree with P. fluorescens (AB178888) as an outgroup. Sequences for the Australian strains were deposited in GenBank (accession numbers are included in Table 1).

Pathogenicity

Grapevine leaf assay

The two representative strains from Tumbarumba grape (DAR 77819 and DAR 77820), plus NSW I&I Pss strains DAR 73915, DAR 75527, DAR 75241 and DAR 72042 and Psm strain DAR 33417 were tested for the ability to cause necrotic leaf lesions on grapevine leaves. Attached young leaves on potted Chardonnay plants were inoculated by puncturing with a 26-gauge needle and applying aliquots (100 μL) of cell suspension (cells from 4-day-old nutrient agar (NA; Oxoid) culture suspended in SDW to a concentration of 105 cells mL−1). Control leaves were treated identically with SDW. The potted vines were maintained in a growth cabinet at 24°C day/18°C night with a 12-h fluorescent light/dark cycle. Each treatment was replicated three times. After 1 week, the leaves were surface-sterilized and plated onto PS agar for reisolation of P. syringae.

Pot experiments

Further pathogenicity testing of Pss strains DAR 77819 and DAR 77820 from infected grapevines at Tumbarumba was conducted at low and high humidity on 3-year-old potted V. vinifera cv. Semillon plants in 39-cm-diameter, 12-L pots containing coarse river sand, loam and compost (1:1:3). Each treatment was replicated three times.

Prior to inoculation, two leaves per plant and one flower from each inflorescence were removed and checked for the presence of P. syringae and B. cinerea. Botrytis cinerea was identified by colony appearance on PDA and by microscopy (shape, size and colour of conidiophores and conidia). Plant tissue was surface-disinfected with calcium hypochlorite (1% available Cl) for 1 min, macerated with a mortar and pestle in 5–10 mL sterile phosphate buffered saline (pH 7), serially diluted, spread (50 μL) over PS agar and DRBC and incubated at 25°C in darkness.

For testing at low humidity, leaves and rachises were inoculated at 20% cap-fall (early flowering) by spraying to runoff with bacterial suspension (108 cells mL−1 in SDW) prepared from 2-day-old NA cultures (incubated at 25°C in darkness). Control plants were sprayed with SDW. The plants were left open to the air and were arranged in randomized complete blocks in a glasshouse maintained at 15–25°C, 40–60% RH and watered to field capacity twice weekly. Pathogenicity testing at high humidity was similar to that in low humidity except that a soft paint brush was used to apply inoculum (108 bacterial cells mL−1) to the leaves and rachises only. Fifteen leaves per plant were inoculated abaxially and another 15 leaves were inoculated adaxially, by painting the inoculum within a 20-mm-diameter circle, drawn with a non-phytotoxic marker, close to the petiole. Relative humidity was maintained at > 99% by enclosing each plant in a clear plastic bag with limited ventilation.

Six weeks after inoculation (a.i.), leaves and bunches for bacterial reisolation and microscopic examination were aseptically cut from the plants and immediately placed in zip-lock plastic bags at 4°C. Discs (4 mm) were cut from both necrotic and healthy sections of leaves and suspended in SDW and observed for bacterial streaming. Subsamples of inoculated and control tissue were surface-sterilized, macerated, plated on PS and DRBC and incubated as described previously. Pseudomonas syringae growing on PS was identified by motility, colony and soluble pigment colour, fluorescence under UV light, gram stain, Biolog, LOPAT, gelatin liquefaction and aesculin hydrolysis. The number of leaves and rachises affected by sporulating B. cinerea lesions were also noted and the fungus was identified by colony appearance and microscopy as described previously. Following the final sampling, the plants were pruned, removed to low-humidity conditions outdoors and observed for symptoms of disease in following seasons.

Analysis of variance, using genstat for Windows, 8th Edition, was applied to determine the extent of reisolation of Pss from lesions; the size of necrotic leaf and petiole lesions; and the numbers of senesced and/or B. cinerea-infected leaves, bunches with black rachises, and detached bunches.

Microscopy

Light and SEM microscopy were performed on shoot and bunch tissues from the pot experiment, 6 weeks a.i., to observe whether the bacteria entered the host via stomata and to investigate subsequent tissue colonization.

For light microscopy, leaf pieces were processed within 4 h of sampling. Leaf tissue squares (10 × 10 mm) were cut with a razor blade from five positions within the inoculation site and five positions outside the inoculation site, stained with 0·05% aniline blue in lactophenol, and observed under UV (330–385 nm) and brightfield illumination with an Olympus Provis AX70 microscope.

Plant tissue for cryoSEM was frozen in liquid N2 (LN2) on sampling and stored at −80°C. Sections of leaf or rachis were mounted on stubs with carbon paste and frozen in LN2 slush, and then transferred to the cold stage of a cryoSEM (Cambridge Model S360; Cambridge Instrument Co.), etched for a few minutes at −85°C to show the cell outlines, cooled to −160°C, coated with gold and observed at 18 kV (McCully et al., 2000).

In 2008, five vineyard tissue samples for transmission electron microscopy (TEM) (viz. two Sauvignon Blanc inflorescences from separate vines, one Pinot Noir lignified shoot and two Riesling inflorescences from one vine) were surface-sterilized and plated as described previously for bacterial isolation. Pieces of tissue (2 × 1·5 mm) were also fixed in Karnovsky (1965) fixatives for 6 h, treated with 2% osmium tetroxide for 4 h, stained with 2% uranyl acetate in isobutanol saturated water for 1 h, dehydrated in an ethanol series, transferred to acetone and polymerized in Spurr (1969) epoxy resin. Sections (80 nm) were stained with uranyl acetate and lead citrate and examined under a Philips 208 Transmission Electron Microscope.

Results

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

Vineyard symptoms and bacterial isolation

Disease symptoms in vineyard-grown plants first appeared at flowering, about 60 days after budburst. They included brown, longitudinal striations on the shoots, rachises and leaf petioles and veins, accompanied by necrotic rachises and abscission of up to 60% of inflorescences. The leaves were initially covered in small dark spots with yellow halos. The leaf spots enlarged, became irregular in shape and were generally bounded by the small leaf veins. Affected tissue split as it became dry and brittle. As infection progressed, the spots tended to merge until much of the leaf was covered and the leaves senesced. Drops of bacterial ooze were sometimes visible near shoot lesions.

TEM of the five Pinot Noir, Sauvignon Blanc and Riesling vineyard plant tissue samples collected in 2008 revealed bacteria within decayed rachis phloem companion cells and decayed cortical cells surrounding the phloem of pedicels (flower stems) of Sauvignon Blanc. Bacteria were also revealed inside the cortical cells (data not shown) and between cortical cells of lignified Pinot Noir shoots (Fig. 1). TEM did not reveal visible bacteria from the Riesling samples, but Pss was the sole bacterium isolated from all five surface-sterilized samples.

Identification of bacteria from grapevines with symptoms

Motile gram-negative rods (1–2·5 μm long) were recovered from a surface-sterilized shrivelled berry and a necrotic rachis collected in 2006 (DAR 77820 and DAR 77819, respectively). The biolog microlog 1 Version 4.20 bacterial identification system and the results of LOPAT (group Ia, + −−− +) (Lelliott et al., 1966) identified P. syringae. The tobacco leaf HR tests were positive, with typical ‘water-soaked’ HR lesions (with translucent collapsed tissue) covering the entire infiltrated region of tobacco leaf (data not shown). On PS, the colonies produced yellow soluble pigment and fluoresced blue under UV light (365 nm) initially and yellow-green subsequently within 4 days; features consistent with Pss (Burkowicz & Rudolph, 1994).

FAME analysis did not provide a definite identification for the bacterial strains from Tumbarumba i.e. DAR 77819 was identified as either Pss (similarity index 0·919) or Psm (0·875); and DAR 77820 as either Psm (0·930) or Pss (0·919). However, results for both gelatin liquefaction and aesculin hydrolysis testing were positive, indicating the probable identity of the strains to be Pss. Coronatine, a phytotoxin not produced by Pss, was not detected by HPLC, providing further confirmation that the Tumbarumba strains DAR 77819 and DAR 77820 were Pss and not Psm. DAR 77820 and DAR 77819 were ice-nucleation-active, as all 10 inoculated droplets froze immediately at −6°C whereas the control droplets did not.

Eighteen additional bacterial strains, indistinguishable from DAR 77820 and DAR 77819 (by gram stain, fluorescence, motility, LOPAT, gelatin liquefaction and aesculin hydrolysis tests) were recovered from the diseased inflorescences sampled both in 2006 and 2007. Healthy inflorescences did not contain Pss.

Molecular characterization of Pseudomonas strains

Sequencing (rpoB) of the grape strains in this study revealed that they were most closely related to known Australian NSW I&I Pss strains (Table 1) and clustered with Pss sequences downloaded from GenBank. The rpoB sequences produced a phylogenetic tree which clearly separated the Pss and Psm sequences with strong bootstrap support (100%) (Fig. 2). This analysis confirmed the Tumbarumba grape strains to be Pss rather than Psm.

image

Figure 2. upgma tree derived from rpoB gene sequences from representative collected Australian grape strains (boxed); Australian apple and Prunus strains; and published sequences from GenBank. Pseudomonas syringae pv. syringae (Pss) and P.s. pv. morsprunorum (Psm) groups are clearly separated. Numbers on nodes are bootstrap values, the frequency (%) with which a cluster appeared in a bootstrap test of 1000 runs. Sequence for Pseudomonas fluorescens (AB178888) was included as an outgroup. rpoB sequence data from representative collected strains were deposited in GenBank with accession numbers HM370531–HM370538. *Probably synonyms of Pss (Gardan et al., 1991, 1999; Sawada et al., 1999; Young et al., 1992).

Download figure to PowerPoint

The Pss cluster contained Pss sequences downloaded from the GenBank data base (pv. aceris, pv. aptata, pv. atrofaciens, pv. dysoxyli, pv. japonica, pv. lapsa, pv. panici, pv. papulans and pv. pisi) that were probably mislabelled as it has been suggested that these are all synonyms of pv. syringae (Gardan et al., 1991, 1999; Young et al., 1992; Sawada et al., 1999).

The rpoB sequences for representative grape vineyard strains DAR 77819 and DAR 77820 (from Tumbarumba, 2006) were identical but differed by 1/728 nt from the sequence for MW844 (diseased rachis, 2007, Tumbarumba). Significantly, the sequence for MW844 was identical to that of Australian Pss strain DAR 73915. The protein sequences, coded by the rpoB gene sequences, for these three Tumbarumba strains were identical to those of the bacteria isolated from South Australia grape. Pss strain DAR 72042 differed from the Tumbarumba grape sequences by 4/728 nt, whereas Psm DAR 33419 differed by 28/728 nt and P. fluorescens differed by 77/728 nt (Table 1).

The 16S rRNA sequences were not able to separate Pss and Psm. Sequences of Pss grape strains DAR 77819 and DAR 77820 were either identical to, or differed by 1/666 nt from, all the other Psm and Pss sequences (Table 1).

Pathogenicity

Grapevine leaf assay

The two representative Tumbarumba grape strains (DAR 77819 and DAR 77820), plus NSW I&I Pss strains DAR 73915, DAR 75527, DAR 75241 and DAR 72042 (apple, Batlow); and Psm strain DAR 33417 caused necrotic lesions on attached leaves of Chardonnay potted plants (both at the points of inoculation and leaf extremities). Two days a.i. brown spots appeared along the veins, progressing from the point of inoculation to the (necrotic) leaf extremities. Seven days a.i. the leaves were covered in dry angular necrotic leaf lesions which tended to split, leaving holes. At 13 days a.i. all the inoculated leaves and rachises had sporulating B. cinerea lesions and had senesced, whereas leaves treated with water showed no disease symptoms (Table 1).

Pot experiments

No natural P. syringae was isolated from the potted vine tissue (surface-sterilized or not) prior to inoculation, but B. cinerea was found to be a latent asymptomatic endophyte. Regardless of humidity, application of representative Tumbarumba Pss strains DAR 77819 and DAR 77820 to potted Semillon vines elicited angular lesions at the site of inoculation. The leaf inoculation position (abaxial or adaxial) did not significantly affect the symptoms. The leaf lesions developed in the same way as those observed in the vineyard (Figs 1b,c and 3a). Similarly to vineyard-infected material, dark, necrotic spots appeared along the veins, initially within the inoculation site but progressing to the non-inoculated regions of the leaves (Fig. 3a), and brown longitudinal lesions developed on petioles (Fig. 3b) and shoots. Stereomicroscopy also showed that the smallest leaf veins were dark and necrotic (data not shown). At low humidity, inoculated rachises did not appear necrotic and did not abscise (data not shown).

image

Figure 3.  Visible symptoms 2 weeks after Semillon grapevines were inoculated with Pseudomonas syringae pv. syringae in pot experiment 2 (high humidity); (a) necrotic split HR lesion within inoculation site and necrotic spots along vein (arrows); (b) brown longitudinal striations on leaf petioles.

Download figure to PowerPoint

Six weeks a.i. inoculated plants at high humidity (> 99% RH) had lesions on leaves and petioles 46–58 and 47–58 mm long, respectively; 63–83% of leaves were covered in sporulating lesions of B. cinerea and 32–80% of all leaves had senesced. All inoculated rachises appeared dark, moist and necrotic, and 56–67% of the inflorescences had abscised. In contrast, non-inoculated plants at high humidity remained symptomless (Table 2).

Table 2.   Disease symptoms observed on Vitis vinifera cv. Semillon plants after inoculation with Pseudomonas syringae pv. syringae (Pss) (pathogenesis pot experiment, high humidity)a
Pseudomonas syringae pv. syringae strainsExtent of necrotic lesion on leaves 6 weeks a.i. (mm)Length of necrotic lesion on petioles 6 weeks a.i. (mm)Number of senesced leaves 6 weeks a.i. (%)Number of leaves covered with sporulating Botrytis cinerea 6 weeks a.i. (%)Inflorescences with black rachises 6 weeks a.i. (%)Bunches detached 6 weeks a.i. (%)Reisolation of Pss from lesions on shoots, rachises, petioles 6 weeks a.i.bPss bacterial streaming from leaf discs of inoculated leaves 6 weeks a.i.cReisolation of Pss from lesions on shoots, rachises, petioles 2 years a.i.Caterpillars on leaves 2 years a.i.
  1. aBacterial inoculum (108 cells mL−1) or sterile deionized water (SDW; non-inoculated control) was applied to potted V. vinifera cv. Semillon plants with a soft paint brush. Leaves were inoculated within 2-cm-diameter circles marked adjacent to the petiole; 15 leaves per pot were treated abaxially and 15 adaxially. Inflorescences were inoculated along the rachis. The vines were enclosed in clear plastic bags at RH > 99%. Means of three replicate determinations are shown. Values within a column followed by the same letter are not significantly different. Leaf inoculation position (abaxial or adaxial) did not significantly affect symptoms.

  2. bFulfilling Koch’s postulates.

  3. cBacteria streaming from leaf discs in SDsW were identified as Pss.

  4. a.i. = after inoculation.

Pss (DAR 77819)58·3 a58·3 a80 a83 a100 a67 a+++4·3 a
Pss (DAR 77820)46·4 b46·6 b32 b63 a100 a56 a+++4·0 a
Not-inoculated control0 c0 c11 c0 b0 b0 b---1·7 b
P< 0·001< 0·001< 0·001< 0·001NA0·034NANANA0·014
l.s.d.7·410·228·327·7NA56·1NANANA1·7

Six weeks a.i. 10 Pss samples were recovered from the leaves, petioles and rachises of the inoculated high-humidity plants. One of these (MW953 recovered from a shoot lesion) was sequenced and proved to have identical rpoB sequence to the Pss inoculant (Fig. 2). No Pss was isolated from the non-inoculated plants at high humidity; thus confirming Koch’s postulates. Bacterial streaming was observed from discs (4 mm), cut from both necrotic and healthy sections of inoculated leaves, suspended in SDW, and these bacteria were identified as Pss (Table 2). Interestingly, following removal to low-humidity conditions outdoors, the inoculated plants proved to be more susceptible than the non-inoculated plants to attack by grapevine moth caterpillar in the next two seasons. Pss was also recovered from shoot lesions on the inoculated vines 2 years after inoculation followed by high humidity, indicating a long-lasting systemic infection (Table 2).

Microscopy

Bacterial cells on leaf surface

CryoSEM analysis of leaves from the high-humidity pot experiment conditions, 6 weeks a.i., showed that the bacteria had spread from the site of inoculation to sparsely cover the entire leaf surface, including symptomless sites (Fig. 4a). There was also evidence of limited spread, on inoculated plants, from the 30 inoculated leaves to the non-inoculated leaves on the same plants (Fig. 6d). However, leaves of the non-inoculated control plants remained free of the bacteria (data not shown).

image

Figure 4.  Stomata and trichomes on leaf surfaces and rachis of potted Semillon grapevines in high humidity, 6 weeks after inoculation with Pseudomonas syringae pv. syringae DAR 77819 (b, d, e, f) or DAR 77820 (a, c). (a) CryoSEM of symptomless part of inoculated leaf with bacteria (0·9 μm long) aggregating on normal stoma and some small (0·4 μm long) bacterial cells away from stoma. (b) CryoSEM of deformed stoma on a symptomless site of an inoculated leaf with bacteria (0·9 μm long) on deformed stoma (pore width = 4·4 μm) with radiating folds in adjacent tissue. (c) CryoSEM of aggregates of bacteria (1 μm long) at the base of a trichome in symptomless leaf tissue furthest from inoculation site; inset 1: same trichome at lower magnification. (d) CryoSEM of necrotic lesion on inoculated leaf, aggregates of bacteria (0·8–2 μm long) on and inside deformed stoma (pore width = 6·3 μm) with radiating folds in adjacent flattened tissue. (e) Brightfield illumination, aniline blue stain; brown deformed stoma (pore width = 3 μm) with radiating folds of tissue on a necrotic part of an inoculated leaf. (f) UV illumination, same leaf as (e); callose fluorescing on stomata with radiating folds of tissue. Arrows indicate bacteria; scale bars: a, c, d = 10 μm; b, e, f = 20 μm; c (inset 1) = 100 μm.

Download figure to PowerPoint

image

Figure 6.  Colonization by Botrytis cinerea of bunch rachises and leaves of cv. Semillon grapevines 6 weeks after inoculation with Pseudomonas syringae pv. syringae (Pss) strains DAR77820 (a, b, c) or DAR 77819 (d) under high-humidity conditions (cryoSEM). (a) Transverse section of a rachis, covered with B. cinerea conidiophores. (b) Closer view of rachis surface in (a) with B. cinerea and bacteria (1·8–2·7 μm long), showing chain of three bacterial cells (x) on surface of hyphae. (c) Non-inoculated leaf showing B. cinerea hyphae approaching open stoma and a few small bacteria (0·5–0·7 μm long). (d) B. cinerea hyphae penetrating open stoma, on symptomless region close to inoculated site. Key: b = Pss bacterial cells; c = B. cinerea conidiophore; h = hyphae; ph = penetrating hyphae; s = B. cinerea conidia (spores); x = chain of bacteria on hyphae. Scale bars: a = 500 μm; b, c, d = 10 μm.

Download figure to PowerPoint

On inoculated leaves, aggregates of bacterial cells were commonly observed near open stomata on both symptomless (Fig. 4a,b) and necrotic leaf parts (Fig. 4d) and at the base of leaf trichomes (Fig. 4c). CryoSEM of leaves, sampled both 1 and 6 weeks a.i., revealed that some stomata in both symptomless (Fig. 4b) and necrotic (Fig. 4d) parts of inoculated leaves were deformed. The deformed stomata were open, even in darkness, and protruded above the surrounding epidermis that was sometimes distorted by folds of tissue extending radially from them (Fig. 4b,d). No deformed stomata were observed on leaves of non-inoculated control plants (data not shown). Epidermal cells surrounding deformed stomata on necrotic or non-necrotic leaf tissue appeared to have unusually low turgidity (Fig. 4b,d). When leaf tissue with deformed stomata was stained with aniline blue and observed under bright field illumination the radial folds gave the deformed stomata a ‘star-like’ appearance (Fig. 4e), and when observed under UV the whole complex fluoresced, revealing deposition of callose (Fig. 4f).

Bacterial cells in rachis and petiole

CryoSEM of inoculated plants revealed large aggregates of bacteria inside petiole parenchyma cells (Fig. 5a,b) and smaller aggregates inside the rachis xylem vessels (Fig. 5c). Some bacteria were near the pits in xylem walls (Fig. 5d). The absence of bacteria inside petioles or rachises of non-inoculated control plants, the fact that only Pss was isolated from the surface-sterilized tissues, and the fact that bacteria were also observed by TEM in tissue from the vineyard, indicate that the bacteria observed within these tissues were likely to be Pss. However, future research using microscopy of fluorescently labelled Pss bacteria is needed to clarify this point.

image

Figure 5.  Bacteria within cells and xylem of grapevine cv. Semillon 6 weeks after external inoculation of leaves with Pseudomonas syringae pv. syringae DAR 77820. CryoSEM micrographs, high humidity. (a) Unusually large bacterial cells (up to 4 μm long) inside water-filled parenchyma cells of leaf petiole. (b) Large bacterial cells (up to 4·3 μm long) inside water-filled parenchyma cells of another leaf petiole. (c) Smaller bacteria (2 μm long) inside air-filled (shown by absence of ice crystals) xylem vessels in bunch rachis with black necrotic exterior; some bacteria adjacent to xylem vessel pits (p). (d) Higher magnification of portion of Fig. 4c showing a bacterium adjacent to a pit in xylem wall. Block arrows indicate bacterial cells; p = bacteria associated with pits; scale bars: a = 20 μm; b = 50 μm, c = 10 μm, d = 5 μm.

Download figure to PowerPoint

Bacterial cell size

The bacteria varied greatly in length; viz. 0·4 μm on leaf surfaces (Fig. 4a); 1–1·5 μm long at the base of leaf trichomes (Fig. 4c); 2 μm inside stomatal pores (Fig. 4d); 2 μm inside air-filled rachis xylem vessels in the rachis (Fig. 5c); 4 μm inside petiole parenchyma cells (Fig. 5a,b).

Fungal interactions: Botrytis cinerea

Prior to the commencement of the pathogenicity pot experiment, the plants were found to contain latent, asymptomatic B. cinerea in leaves and flowers. At both high and low humidity, sporulating B. cinerea lesions were prevalent on leaves and bunches of inoculated plants 6 weeks a.i. with Pss. However, the fungus remained latent and asymptomatic in the plants not inoculated with Pss and there were no grey mould symptoms. Leaf inoculation position (abaxial or adaxial) did not significantly affect the symptoms.

CryoSEM also revealed the presence of B. cinerea hyphae, conidiophores and conidia on the surface of rachises and leaves of plants inoculated with Pss (Fig. 6). Interestingly, short chains of bacteria were observed on the surface of some B. cinerea hyphae (Fig. 6b). Fungal hyphae tended to grow towards leaf stomata (Fig. 6c) and penetrated both the stomata pores and guard cells (Fig. 6d).

Discussion

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

This study has identified, by completion of Koch’s postulates, biochemical testing and sequencing of rpoB genes, that P. syringae pv. syringae (Pss) is the cause of a condition described here as ‘bacterial inflorescence rot’ in grapevines. Future work will include a greater number of bacterial strains from geographical regions in which the disease occurs and use of several housekeeping genes including the gyrB gene region in addition to the rpoB gene.

Other reports have described various pathologies of grapevines caused by P. syringae, including necrotic lesions in leaf tissues, shoots, tendrils and rachises (Klingner et al., 1976; Hall et al., 2002), bark necrosis (Cugusi et al., 1986) and bacteriosis (Samedov et al., 1988). Furthermore, one of the strains isolated was indistinguishable from those isolated by Hall et al. (2002). However, the present study is the first report of pathogenesis in grapevine inflorescences by P. syringae as the cause of significant fruit losses.

Pss is a widely distributed motile inhabitant of plant surfaces, soil and water. This study indicates a grapevine epidemiology in which the bacterium overwinters in wood. This is consistent with overwintering of Pss in wood of fruit trees (Wimalajeewa & Flett, 1985; Roos & Hattingh, 1986; Sundin et al., 1988; Mansvelt, 1989). Following grapevine dormancy, the bacterium spreads across wet surfaces of emerging shoots, leaves and inflorescences. Microscopy indicated that Pss gains entry to those tissues through open stomata. Thereafter, cellular infection is followed by reproduction and systemic movement of bacteria. Future investigations will assess inoculated tissues at a series of time intervals following inoculation in order to provide more useful information on the processes of colonization and host infection by the pathogen.

Although Pss strains from Tumbarumba grape are ice-nucleating, in the glasshouse study significant infection occurred in humid conditions unaccompanied by freezing. Thus, whether or not ice nucleation normally plays a significant role in the infection process is unclear. Although frost damage is often associated with Pss-induced pear blossom blast, that disease can also occur in frost-free cool, wet weather (Whitesides & Spotts, 1991; Latorre et al., 2002).

Infection was promoted by high humidity. This is in accord with the previous finding that South Australian Pss strain DAR 73915, very closely related to the Tumbarumba samples (Table 1, Fig. 2), caused leaf and shoot infections in the Adelaide Hills and Southern Vales regions of Australia after heavy rains in spring in 2001 and 2002 (Hall et al., 2002). It was found that loss of inflorescences occurs only at particularly high and sustained humidity. The Tumbarumba vineyards tend to be particularly susceptible to such losses because of the frequent use of foliar water sprays to protect plants from spring frost damage. The promotion of Pss incidence and infection by conditions (e.g. rain, dew and low temperature) favouring leaf wetness has been demonstrated in other fruit crop species, including pear (Latorre et al., 2002), stone fruits (Foulkes & Lloyd, 1980; Wimalajeewa & Flett, 1985) and mango (Young, 2008). The present findings indicate that the use of foliar wetting to ameliorate risks of frost damage imposes a concomitant risk of bacterial infection, particularly if wetness is prolonged.

Although solitary Pss bacteria were found all over the leaf surface, as expected from movement of motile bacteria through the water film, the bacteria clustered near trichomes and open stomata. Similar selective clustering of Pss was reported previously around tomato stomata (Melotto et al., 2006) and bean trichomes (Monier & Lindow, 2005). The observation of clusters of Pss within parenchyma cells of necrotic leaf petioles accords with similar reports of Psm aggregates within collapsed parenchyma cells in cherry fruit, and the cortical parenchyma of cherry stalks (Roos & Hattingh, 1988).

Pss has not been reported previously in grapevine xylem vessels but, other bacterial grapevine pathogens, viz. Xylella fastidiosa (Chatelet et al., 2006) and Burkholderia phytofirmans (Compant et al., 2007) are known xylem inhabitants. In the present study the lack of water in the xylem vessels occupied by Pss may be have been caused by the release of xylem tension upon severance for sampling or by cavitations induced by bacterial blockage. Cavitations are usually refilled but permanent inactivation of xylem might occur when bacteria accumulate on vessel end walls (McCully, 2001).

The bacteria observed varied greatly in length. Solitary bacteria on the leaf surface were small (from 0·4 μm long), whereas those inside the petiole parenchyma cells were up to 4·3 μm long This observation accords with that of Björklöf et al. (2000) who reported a similar range in length of P. syringae 4 days after inoculation on bean leaves. It also accords with that of Wilson & Lindow (1992), who showed that up to 75% of P. syringae on bean leaf surfaces for over 80 h were small and non-culturable as a result of a starvation-survival state induced by low nutrient availability.

The observation of distorted permanently open stomata may be significant in terms, not only of Pss infection, but for other microbial infection of grapevine. Recent studies suggest that, although stomata act to restrict bacterial invasion by closing within hours of infection, some plant pathogens have evolved pathogenicity factors to overcome this defence by re-opening the closed stomata (Underwood et al., 2007). As found in the present study, Allègre et al. (2007) observed that stomata on grape leaves infected with Plasmopara viticola (downy mildew) were permanently open, unable to close in darkness. It was reported that coronatine, a phytotoxin produced by some P. syringae pathovars, caused ‘lock-open’ in Arabidopsis stomata (Melotto et al., 2006), but as it was established that the Pss strains in this study did not produce coronatine, this is not a possible mechanism in this case. Other non-coronatine-producing pathovars of P. syringae, such as P. syringae pv. tabaci, are known to cause tobacco stomata to open (Melotto et al., 2006).

Most previous studies involving inoculation of plant leaves with P. syringae have used the ‘infiltration’ method which bypasses stomata. The inoculation method used in the present study allowed Pss to enter through the stomata as it normally would. The folds of tissue extending radially from the stomata through the adjacent, collapsed epidermal cells may result from degradation and desiccation of epidermal tissue. The resultant ‘star-like’ distortion of infected stomata has not apparently been reported previously.

The induction of growth and sporulation of B. cinerea, a necrotrophic fungus, from asymptomatic latency following infection by Pss, a biotrophic bacterium, is an important new finding. Notably, a similar phenomenon was reported in Arabidopsis leaves, i.e. B. cinerea grew much faster on leaves pretreated with biotrophic P. syringae pv. tomato (Govrin & Levine, 2000). Asymptomatic B. cinerea infections of grapevines are not unusual and, in parts of Australia and South Africa, many plants in commercial vineyards carry them (Nair, 1985; Holz et al., 2003). In cool, wet conditions that favour B. cinerea the fungus rots damaged grapevine parts, including leaves, inflorescences and fruit. The results suggest that Pss-induced cell damage can be a precursor to overt infection invasion by the necrotroph and further cellular decay. Plant defence against biotrophic pathogens such as Pss is regulated by the salicylic acid-dependent pathway and involves HR, whereas jasmonic acid-related defences control resistance against necrotrophic pathogens such as B. cinerea. Although plant HRs reduce the growth of biotrophs, they are detrimental for defence against necrotrophic pathogens which feed on necrotic tissue (Kliebenstein & Rowe, 2008).

Acknowledgements

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

This project was funded by NSW Department of Industry and Investment (NSW I&I). TEM was performed by Mukesh Srivistava, and FAME analysis was performed by Dorothy Noble, NSW I&I. Cryo-scanning electron microscopy was conducted at the Australian National University Electron Microscopy Unit. We thank Professor Martin Canny, Research School of Biological Sciences, Australian National University, Canberra; Dr Ian Dry (CSIRO, Adelaide, Australia) and Professor Jim Hardie, National Wine and Grape Industry Centre, Wagga Wagga, NSW for their helpful advice regarding the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Allègre M, Daire X, Héloir M et al. , 2007. Stomatal deregulation in Plasmopara uvicola-infected grapevine leaves. New Phytologist 173, 83240.
  • Björklöf K, Nurmiaho-Lassila EL, Klinger N, Haahtela K, Romantschuk M, 2000. Colonization strategies and conjugal gene transfer of inoculated Pseudomonas syringae on the leaf surface. Journal of Applied Microbiology 89, 42332.
  • Burkowicz A, Rudolph K, 1994. Evaluation of pathogenicity and of cultural and biochemical tests for identification of Pseudomonas syringae pathovars syringae, morsprunorum, and persicae from fruit trees. Journal of Phytopathology 141, 5976.
  • Chatelet DS, Matthews MA, Rost TL, 2006. Xylem structure and connectivity in grapevine (Vitis vinifera) shoots provides a passive mechanism for the spread of bacteria in grape plants. Annals of Botany 98, 48394.
  • Compant S, Kaplan H, Sessitsch A, Nowak J, Barka EA, 2007. Endophytic colonization of Vitis vinifera L. by Burkholderia phytofirmans strain PsJN: from the rhizosphere to inflorescence tissues. FEMS Microbiology Ecology 63, 8493.
  • Cugusi M, Garau R, Prota U, Dore M, 1986. A bark necrosis of grapevine caused by Pseudomonas syringae V. Hall, in Sardinia. Journal of Phytopathology 116, 17685.
  • Felsenstein J, 1985. Confidence limits in phylogenies: an approach using the bootstrap. Evolution 39, 78391.
  • Foulkes JA, Lloyd AB, 1980. Epiphytic populations of Pseudomonas syringae pv. syringae and P. syringae pv. morsprunorum on cherry leaves. Australasian Plant Pathology 9, 1145.
  • Gardan LS, Cottin S, Bollet C, Hunault G, 1991. Phenotypic heterogeneity of Pseudomonas syringae. Research in Microbiology 142, 9951003.
  • Gardan L, Shaftik H, Belouin S, Broch R, Grimont F, Grimont PAD, 1999. DNA relatedness among the pathovars of Pseudomonas syringae and description of Pseudomonas tremae sp. nov. and Pseudomonas cannabina sp. nov. International Journal of Systematic Bacteriology 49, 46978.
  • Govrin EM, Levine A, 2000. The hypersensitive response facilitates plant infection by the necrotrophic pathogen Botrytis cinerea. Current Biology 10, 7517.
  • Hall BH, McMahon RL, Noble D, Cother EJ, McLintock D, 2002. First report of Pseudomonas syringae on grapevines (Vitis vinifera) in South Australia. Australasian Plant Pathology 31, 4212.
  • Holz G, Gütschow M, Coertze S, Calitz FJ, 2003. Occurrence of Botrytis cinerea and subsequent disease expression at different positions on leaves and bunches of grape. Plant Disease 87, 3518.
  • Karnovsky MJ, 1965. A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy. Journal of Cell Biology 27, 137A.
  • Kliebenstein DJ, Rowe HC, 2008. Ecological costs of biotrophic versus necrotrophic pathogen resistance, the hypersensitive response and signal transduction. Plant Science 174, 5516.
  • Klingner AE, Palleroni NJ, Pontis RE, 1976. Isolation of Pseudomonas syringae from lesions on Vitis vinifera. Phytopathologische Zeitschrift 86, 10716.
  • Kumar S, Tamura K, Nei M, 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics 5, 15063.
  • Lane DJ, 1991. 16S/23S rRNA sequencing. In: StackebrandtE, GoodfellowM, eds. Nucleic Acid Techniques in Bacterial Systematics. New York, NY, USA: John Wiley and Sons, 11575.
  • Latorre BA, Jones AL, 1979. Pseudomonas morsprunorum, the cause of bacterial canker of sour cherry in Michigan, and its epiphytic association with P. syringae. Phytopathology 69, 3359.
  • Latorre BA, Rioja ME, Lillo C, 2002. The effect of temperature on infection and a warning system for pear blossom blast caused by Pseudomonas syringae pv. syringae. Crop Protection 21, 339.
  • Lelliott RA, Stead DA, 1987. Methods for the Diagnosis of Bacterial Diseases of Plants. Oxford, UK: Blackwell Publishing Ltd/British Society for Plant Pathology.
  • Lelliott RA, Billing E, Hayward AC, 1966. A determinative scheme for the fluorescent plant pathogenic pseudomonads. Journal of Applied Bacteriology 29, 47089.
  • Lindow SE, Arny DC, Upper CD, 1978. Distribution of ice nucleation-active bacteria on plants in nature. Applied and Environmental Microbiology 36, 8318.
  • Mansvelt EL, 1989. Infection and systemic invasion of deciduous fruit trees by Pseudomonas syringae in South Africa. Plant Disease 73, 7849.
  • McCully ME, 2001. Niches for bacterial endophytes in crop plants: a plant biologist’s view. Australian Journal of Plant Physiology 28, 98390.
  • McCully ME, Shane MW, Baker AN, Huang CX, Ling LEC, Canny MJ, 2000. The reliability of cryoSEM for the observation and quantification of xylem embolisms and quantitative analysis of xylem sap in situ. Journal of Microscopy 198, 2433.
  • Melotto M, Underwood W, Koczan J, Nomura K, He SY, 2006. Plant stomata function in innate immunity against bacterial invasion. Cell 126, 96980.
  • Monier JM, Lindow SE, 2005. Aggregates of resident bacteria facilitate survival of immigrant bacteria on leaf surfaces. Microbial Ecology 49, 34352.
  • Nair NG, 1985. Fungi associated with bunch rot of grapes in the Hunter Valley. Australian Journal of Agricultural Research 36, 43542.
  • Palmer DA, Bender CL, 1993. Effects of environmental and nutritional factors on production of the polyketide phytotoxin coronatine by Pseudomonas syringae pv. glycinea. Applied and Environmental Microbiology 59, 161926.
  • Roos IMM, Hattingh MJ, 1986. Pathogenic Pseudomonas spp. in stone fruit buds. Phytophylactica 18, 79.
  • Roos IMM, Hattingh MJ, 1988. Systemic invasion of immature sweet cherry fruit by Pseudomonas syringae pv. morsprunorum through blossoms. Journal of Phytopathology 121, 2632.
  • Samedov AN, Mogilevskaya MI, Karagezov TG, Khudaverdieva SR, Aliev LA, 1988. Detection of the causal agents of grape bacteriosis in Aspheron (Azerbaijan SSR, USSR). Izvestiya Akademii Nauk Azerbaidzhanskoi SSR Seriya Biologicheskikh Nauki 6, 1823.
  • Sawada H, Suzuki F, Matsuda I, Saitou N, 1999. Phylogenetic analysis of Pseudomonas syringae pathovars suggests the horizontal gene transfer of argK and the evolutionary stability of hrp gene cluster. Journal of Molecular Evolution 49, 62744.
  • Spurr AR, 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. Journal of Ultrastructure Research 26, 3143.
  • Sundin GW, Olson BD, Jones AL, 1988. Overwintering and population dynamics of Pseudomonas syringae pv. syringae and P.s. pv. morsprunorum on sweet and sour cherry trees. Canadian Journal of Plant Pathology 10, 2818.
  • Tayeb LA, Ageron E, Grimont Y, Grimont PAD, 2005. Molecular phylogeny of the genus Pseudomonas based on rpoB sequences and application for the identification of isolates. Research in Microbiology 156, 76373.
  • Underwood W, Melotto M, He SY, 2007. Role of plant stomata in bacterial invasion. Cellular Microbiology 9, 16219.
  • Whitesides SK, Spotts RA, 1991. Induction of pear blossom blast caused by Pseudomonas syringae pv. syringae. Plant Pathology 40, 11827.
  • Wilson M, Lindow SE, 1992. Relationship of total viable and culturable cells in epiphytic populations of Pseudomonas syringae. Applied and Environmental Microbiology 58, 390813.
  • Wimalajeewa DLS, Flett JD, 1985. A study of populations of Pseudomonas syringae pv. syringae on stonefruits in Victoria. Plant Pathology 34, 24854.
  • Young A, 2008. Notes on Pseudomonas syringae pv. syringae bacterial necrosis of mango (Mangifera indica) in Australia. Australasian Plant Disease Notes 3, 13840.
  • Young JM, Takikawa Y, Gardan L, Stead DE, 1992. Changing concepts in the taxonomy of plant pathogenic bacteria. Annual Review of Phytopathology 30, 67105.