Introduction

There were many reports of bacterial blight in crops of field pea (Pisum sativum) in south-eastern Australia during the late 1990s and early 2000s. Pseudomonas syringae pv. pisi has been regarded as the most important cause of bacterial blight in Australia and internationally (Hollaway et al. 2007).

However, P. syringae pv. syringae has been reported to occur in pea crops in Australia (Wimalajeewa and Nancarrow 1984; Clarke 1990) and overseas (Taylor and Dye 1972; Jindal and Bhardwaj 1989; Lawyer and Chun 2001), but has been regarded as less important than P. syringae pv. pisi, which is reported to cause disease over a wider range of environmental conditions (Taylor and Dye 1972; Lawyer and Chun 2001). The symptoms of bacterial blight (also known as brown spot) in field peas have been described previously (Taylor and Dye 1972; Lawyer and Chun 2001). All aerial plant parts are susceptible to attack throughout the growing season and affected stipules develop characteristic water-soaked, fan-shaped lesions which turn brown and papery. Disease caused by P. syringae pv. syringae or P. syringae pv. pisi cannot be distinguished between by field symptoms.

There is little reported information regarding yield loss caused by P. syringae pv. syringae in field pea. Jindal and Bhardwaj (1989) reported a severe outbreak in northern India attributed to P. syringae pv. syringae, but the extent of yield loss was not quantified. Likewise, there have been limited studies into the epidemiology of P. syringae pv. syringae in field peas. It is thought that P. syringae pv. syringae survives from one season to the next on seed and/or infected crop residues (Lawyer and Chun 2001) in a similar way to P. syringae pv. pisi (Hollaway et al. 2007). Studies of brown spot in common bean (Phaseolus vulgaris), caused by P. syringae pv. syringae indicate that disease development is favoured by cool, wet conditions with ideal temperatures between 12 and 25°C (Hirano and Upper 1990). Disease development is often favoured by frosts due to the ice nucleating properties of the pathogen (Maki et al. 1974).

Pathogenic variation of P. syringae pv. pisi is well documented with a race structure having been determined and resistance identified in cultivars (Taylor et al. 1989; Bevan et al. 1995; Hollaway and Bretag 1995b). There have been limited studies internationally into the presence or absence of host plant resistance or pathogenic variability toward field pea with regard to P. syringae pv. syringae. Butler and Fenwick (1970) reported the presence of pathogenic variability within limited isolates of P. syringae pv. syringae toward field peas.

There appears to have been an increased incidence of bacterial blight in field peas caused by P. syringae pv. syringae since the introduction of new semi-leafless cultivars in south-eastern Australia. New cultivars released during the late 1990s and early 2000s possessed many desirable agronomic traits which included more upright growth habitat and higher grain yields than the older cultivars. This has resulted in a significant increase in the area sown to new semi-leafless cultivars, such as cv. Kaspa.

These studies were undertaken to 1) determine the importance of P. syringae pv. syringae as a cause of bacterial blight of field peas in south east Australia; 2) quantify disease susceptibility and associated yield loss in current commercial cultivars; 3) evaluate a field screening method that could be adopted by field pea breeders, and 4) determine the survival period of P. syringae pv. syringae in infected stubble.

Materials and methods

Cause of bacterial blight in symptomatic field pea crops

Collection of symptomatic field pea plants

During 2005, 40 field pea crops exhibiting symptoms of bacterial blight were sampled from South Australia, Victoria and New South Wales at the 8th to 12th node growth stage (Knott 1987).

From each crop, two or three symptomatic plants were collected and placed in a plastic zip-lock bag and transported back to Horsham for testing. Plants were stored at 4°C until processed for testing.

Isolation of Pseudomonas species

Tests to determine the causal organism(s) were undertaken within two days of sampling the plant material. Leaf and stem tissue with symptoms of bacterial blight was finely chopped using sterile scissors. Ten gram of tissue was added to a 250 ml Schott bottle containing 100 ml of sterile distilled water and soaked overnight at 21°C ± 4°C . The resulting suspension was serially diluted (1:1, 1:10, 1:100 and 1:1000) with sterile distilled water. Using a glass rod, 100 μl of each of the dilutions was spread onto sucrose-nutrient agar supplemented with boric acid, cefuroxime, cycloheximide and cephalexin (SNAC) (Hollaway and Bretag 1995a). Colonies similar in appearance to P. syringae were subcultured after 48 h incubation at 21°C ± 2°C and maintained on King’s medium B (King et al. 1954) until identified. A set of four Pseudomonas spp. reference cultures obtained from the Biological and Chemical Research Institute, Rydalmere (DAR 69866 P. syringae pv. pisi, DAR 58721 P. viridiflava, DAR 35680 P. syringae pv. syringae and DAR 55534 P. cichorii) were also maintained on King’s medium B and used in all subsequent pathogenicity and biochemical tests for reference.

Bacterial identification

Bacterial isolates were identified using pathogenicity and biochemical tests. Isolates were first tested for pathogenicity on field pea seedlings at the 3 to 4 leaf stage using the methods described by Hollaway and Bretag (1995a). Bacteria were scraped from the surface of a 48 h King’s B culture plate using a toothpick. For each isolate, a field pea seedling (cv. Kaspa) was then stabbed with the toothpick in two distinct stem locations. Once inoculated the plants were placed in dew chambers at 25°C ± 3°C and 100% relative humidity for 48 h. Plants were transferred to a controlled environment with a 24 h photoperiod at 15°C ± 3°C for a further 48 h. They were then returned to the glasshouse for the disease to develop and were assessed three days later. A positive result was recorded if a water soaked lesion had developed around the inoculation site. Whether or not the lesion was localised or extensive was also recorded. P. syringae pv. syringae usually causes an extensive lesion that causes stem collapse, whilst P. syringae pv. pisi tends to be more localised (Mazarei and Kerr 1990). Isolates that were positive on pea stem were tested for oxidase reaction (Kovacs 1956), levan production (Lelliott et al. 1966), fluorescence (Hildebrand et al. 1988), pectolytic activity (Hildebrand et al. 1988) and pathogenicity of common bean (Phaseolus vulgaris) and lemon (Citrus limonium) fruits (Wimalajeewa and Nancarrow 1984). Isolates that were oxidase negative, levan positive and potato rot negative belong to the P. syringae group of bacteria (Fahy and Lloyd 1983; Hildebrand et al. 1988). The two pathovars are differentiated by their reaction on lemon and bean fruit as P. syringae pv. pisi does not cause a reaction on lemon and bean fruit whereas P. syringae pv. syringae does (Wimalajeewa and Nancarrow 1984).

Disease scores and yield loss in pea cultivars infected with P. syringae pv. syringae

A field experiment was conducted during 2006 near Horsham at the Department of Primary Industries’ Plant Breeding Centre in the Wimmera region of Victoria. The field site has an average annual rainfall of 450 mm and a friable grey soil type. Monthly rainfall data and the number of frost days observed for 2005–2007 at Horsham are shown in Table 1. The site was cropped to lupins (Lupinus spp.) the previous year and had not been cropped to field peas for at least 5 years. The site was flood irrigated with approximately 50 mm of water 4 weeks prior to sowing to allow for early sowing to encourage disease development.

Table 1 Monthly rainfall (mm) and the number of frost days observed at Horsham and Wagga Wagga during 2005–2007 (Bureau of Meteorology)
Full size table

Eleven pea cultivars (Table 2) were sown on 17 May 2006. Each cultivar was sown at 100 kg of seed per ha in 6-row plots, 6 m long, with a row spacing of 15 cm. Seed was obtained from Tony Leonforte (Department of Primary Industries, Horsham, Victoria). Double superphosphate (0% N, 9% P, 0% K) was applied in furrow at seeding at a rate of 75 kg/ha. Galant West® (Haloxyfop 130 g/L) was applied (19 May) at a rate of 200 ml/ha to control grass weeds and volunteer cereals and Select® (Clethodim 240 g/L) was applied (11 July) at a rate of 150 ml/ha to control broad-leaved weeds.

Table 2 Phenotypic traits and release dates of commercial cultivars used to estimate yield loss caused by P. syringae pv. syringae during 2006
Full size table

The experimental design was a randomised split block with three replicates. The main plot comprised the 11 pea cultivars which were split into three sub-plots. The first sub-plot had field pea stubble naturally infected with P. syringae pv. syringae spread at a rate of 2,500 kg/ha on 28 June 2006 following the sowing of uninoculated seed. The second sub-plot was sown with seed artificially infected with P. syringae pv. syringae prior to sowing in addition to receiving the stubble treatment as described above. Seed was artificially infected by soaking seed in a bacterial suspension (1 × 106 colony forming units/ml) for 20 min while under vacuum and then air dried. The third sub-plot was the nil treatment and was sown with uninoculated seed.

Disease severity within each plot was assessed on the 18 September 2006 using the 0 to 9 scale described in Table 3. Grain yield was determined at plot maturity by recording grain weight from each plot following harvest with a self-propelled Hege plot harvester on the 4 December 2006.

Table 3 Disease scale used for assessment of bacterial blight severity in field plots at Horsham during 2006
Full size table

To determine the cause of bacterial blight symptoms a representative, inoculated plot of each cultivar was tested for the causal agent present. Plants with symptoms were arbitrarily selected and infected material was removed and placed in a separate plastic zip lock bag for each cultivar and promptly returned to the laboratory for testing. Bacterial isolation and identification was then conducted as described above

Statistical analysis

All statistical analyses were conducted using Genstat 11th Edition. Analysis of variance was conducted on the grain yield data. For analysis of the disease score a Friedman Sum Rank test (Friedman 1937) was conducted. Regression analysis was conducted to relate yield to disease score for each plot.

Field screening of pea cultivars for their reaction to P. syringae pv. syringae

Sixty six advanced field pea breeding lines from Pulse Breeding Australia were screened at Horsham for resistance to bacterial blight caused by P. syringae pv. syringae using the same site as the field experiment described previously. Field pea stubble naturally infected with P. syringae pv. syringae was spread at a rate of 2,500 kg/ha on 28 June 2006 on all plots following the sowing of uninoculated seed on 17 May 2006. The experimental layout was a non-replicated design with checks. The checks consisted of 12 commercial cultivars, distributed randomly throughout the experimental design, and replicated at least twice. Agronomic details, disease and yield assessment were as described for the previous field experiment. The cause of disease symptoms was determined within a representative plot of each cultivar and breeding line, also as described above for the previous field experiment. Likewise, data were analysed as described above.

Survival of P. syringae pv. syringae on field pea stubble

The survival of P. syringae pv. syringae on field pea stubble was studied using the method described by Hollaway and Bretag (1997). Stubble naturally infected with bacterial blight, caused by P. syringae pv. syringae, was collected during January 2005 from a field pea crop (cv. Kaspa) near Rupanyup in the Wimmera region of Victoria. The crop was harvested during December 2004.

Infected pea stubble (5 g) was placed into plastic mesh bags (mesh size 3 × 2 mm; bag size 15 × 15 cm) which were stapled closed and placed in the field. The mesh bags were either pegged and left at the soil surface or buried at 10 cm below ground level.

The experiment was conducted twice at each of two sites using randomised block designs. The first site was at the Plant Breeding Centre of the Victorian Department of Primary Industries (friable grey clay, soil pH 8.6) near Horsham and was established on 2 March 2005 and 10 February 2006. The second site was at the Wagga Wagga Agricultural Institute, New South Wales Department of Primary Industries (red-brown earth, soil pH 5.0) near Wagga Wagga in the Riverina region of New South Wales and was established on 21 February 2005 and 13 February 2006.

Each treatment was replicated five times with ten bags of field pea stubble (5 buried, 5 surface) and removed from the field after 0, 10, 14, 18, 22, 26, 30, 34, 38, 66 and 118 weeks. The content of each bag was tested for the presence of viable P. syringae pv. syringae as described below. Monthly rainfall data and frost days for the Horsham and Wagga Wagga weather stations were obtained from the Bureau of Meteorology (Table 1).

Upon removal from the field excess soil was shaken from each bag and any plant matter attached was also removed. The bags were opened, stubble residues removed, placed into a 500 ml Schott bottle with 100 ml of sterile distilled water and left overnight to soak at 21°C ± 4°C . The resulting suspensions were serially diluted with sterile distilled water and using a glass rod, 100 μl of each of the dilutions 1:1, 1:10, 1:100 and 1:1000 were spread onto sucrose-nutrient agar supplemented with boric acid, cefuroxime, cycloheximide and cephalexin (SNAC) as used by Hollaway and Bretag (1995a). Isolation and identification methods were then conducted as described previously in the section, “Cause of bacterial blight in symptomatic field pea crops”.

Results

Cause of bacterial blight in symptomatic field pea crops

P. syringae pv. syringae and/or P. syringae pv. pisi were isolated from all 40 crops with symptoms of bacterial blight sampled from south-eastern Australia during 2005 (Table 4). P. syringae pv. syringae was the sole cause of bacterial blight identified in 40% of crops, P. syringae pv. pisi the sole cause in 47.5%, while both pathovars of P. syringae were detected in 12.5% of crops.

Table 4 Number of field pea crops with symptoms of bacterial blight from which P. syringae pv. pisi (Psp) and/or P. syringae pv. syringae (Pss) were detected during 2005 in New South Wales, Victoria and South Australia
Full size table

Disease scores and yield loss in pea cultivars infected with P. syringae pv. syringae

In the presence of P. syringae pv. syringae infected pea residues all eleven commercial cultivars evaluated in the field showed symptoms of bacterial blight and suffered grain yield loss relative to the uninoculated plots in which only minimal symptoms of bacterial blight developed (Table 5). Within the cultivars there were reductions in grain yield ranging from 13% in cv. Sturt to 94% in cv. Moonlight.

Table 5 Means of disease scores and grain yields recorded from uninoculated (nil) and inoculated plots of eleven commercial cultivars from a field trial in Horsham, 2006
Full size table

With regard to disease score there were significant treatment, variety and treatment by variety interactions observed (Table 5). Within the varieties two distinct groups were identified; a resistant group, with a disease score less than 4 in the presence of inoculum and a susceptible group with a disease score of greater than 6 in the presence of inoculum. The median disease scores of the resistant and susceptible groups in the presence of inoculum were 3 and 7, respectively. Similarly, in the presence of inoculum the mean grain yield of the resistant and susceptible groups was 1.82 t/ha and 0.04 t/ha, respectively. The average yield loss of the resistant cultivars in the presence of infected stubble was 23% and within the susceptible group it was 75%.

A linear regression model relating yield to disease score was fitted to individual plot data. There was a significant negative linear effect of disease on yield (P < 0.001). The model for yield was:

$$ {\hbox{Yield}}\,\left( {{{\hbox{t}} \left/ {\hbox{ha}} \right.}} \right) = 2.35 - 0.283 \times {\hbox{disease}}\,{\hbox{score}} $$

The overall variance explained from this model was 65%.

Testing of symptomatic plants from the field trial confirmed in all cases that P. syringae pv. syringae was the cause of the bacterial blight epidemic in the field.

Evaluation of breeding lines for resistance to P. syringae pv. syringae

The 66 breeding lines evaluated varied significantly (p = 0.006) in disease severity and corresponding grain yields (p = <0.001) (data not shown). They could be grouped into resistant and susceptible categories (Table 6) based on disease score (0–5 and 6–9 respectively) and yield (>0.3 t/ha and < 0.29 t/ha respectively).

Table 6 Severity of bacterial blight in advanced breeding lines and commercial cultivars of field peas grown in the presence of stubble infected with P. syringae pv. syringae in the field at Horsham, 2006
Full size table

A linear regression model relating yield to disease score was fitted to individual plot data. There was a significant negative linear effect of disease on yield (P < 0.001). The model for yield was:

$$ {\hbox{Yield}}\left( {{{\hbox{t}} \left/ {\hbox{ha}} \right.}} \right) = 1.41 - 0.186 \times {\hbox{disease}}\,{\hbox{score}} $$

The overall variance explained from this model was 69%.

Survival of P. syringae pv. syringae on field pea stubble

P. syringae pv. syringae could not be detected on naturally infected field pea stubble monitored in the field after 30 weeks in either of the two studies each conducted at two locations (Horsham, Victoria or Wagga Wagga, New South Wales) regardless of whether the stubble was buried or remained on the soil surface (Table 7).

Table 7 Percentage of mesh bags (n = 5) containing field pea stubble naturally infected with P. syringae pv. syringae from which viable P. syringae pv. syringae could be isolated when placed on the soil surface or buried at Horsham or Wagga Wagga for 2 years with studies commencing in March/February 2005 and in February 2006
Full size table

Discussion

This study shows that P. syringae pv. syringae can be an important cause of bacterial blight in field pea and is the first to quantify yield losses due to this pathogen. Variation between field pea cultivars in their reactions to P. syringae pv. syringae was demonstrated which has implications for cultivar selection and plant breeding. Previously P. syringae pv. syringae was not considered a serious pathogen of field pea (Lawyer and Chun 2001), but results of these studies showed that P. syringae pv. syringae can cause significant crop loss in susceptible cultivars.

In 40% of crops with symptoms of bacterial blight, P. syringae pv. syringae was the causative agent during 2005. Although P. syringae pv. syringae was known to be widespread in field pea crops within Victoria during the 1980s (Wimalajeewa and Nancarrow 1984; Clarke 1990) it has not been considered an important cause of bacterial blight epidemics (Hollaway et al. 2007). The following possibilities exist for the increased importance of P. syringae pv. syringae as a cause of bacterial blight in field peas: 1) an increased prevalence of the pathogen in field pea crops; 2) the environment has become more conducive for development of bacterial blight caused by this pathogen; and/or 3) new field pea cultivars are more susceptible to this pathogen than older cultivars.

Results of these studies implies that the increased prevalence of bacterial blight caused by P. syringae pv. syringae may be associated with the adoption of cultivars susceptible to this pathogen. Prior to 2000 the dominant field pea cultivar in south-eastern Australia was Dundale, which was shown to be moderately resistant to P. syringae pv. syringae in our studies. Since 2000, growers have adopted new higher yielding cultivars such as cv. Kaspa, Excell and Snowpeak, which have been shown in this study to be more susceptible to bacterial blight caused by P. syringae pv. syringae than the conventional cultivar Dundale. The five most P. syringae pv. syringae susceptible cultivars identified in this field study were all released after 1999. As bacterial blight caused by P. syringae pv. syringae has been regarded as a minor disease of field peas, there has been no selection applied within breeding programs during the development of these cultivars.

Results from the Horsham field experiments showed that yield losses due to P. syringae pv. syringae can be as high as 94% in cv. Moonlight. Eleven commercial cultivars could be separated into two groups: resistant cultivars with a disease score <5 and susceptible varieties with a score >5. The yield loss in the resistant group was approximately 23% whereas in the susceptible group the loss was approximately 75% and this is the first report quantifying losses due to bacterial blight caused by P. syringae pv. syringae in field pea. The susceptible group consists of cvs. Snowpeak, Kaspa, Bundi, Excell and Moonlight. These cultivars are all semi-leafless and semi-dwarf types except for cv. Moonlight. These phenotypic traits may be linked to P. syringae pv. syringae susceptibility; however further studies are required to test this hypothesis.

The variability between field pea cultivars and advanced breeding lines in their reactions to P. syringae pv. syringae suggests that it should be possible for pea breeders to develop resistant cultivars and avoid the release of more susceptible cultivars. For the purpose of screening early generation breeding lines, when limited seed is available, this study has indicated that it is possible to screen for bacterial blight in the field. Of the 66 lines evaluated, 45 were rated as susceptible and 21 as resistant based on the disease scores applied relative to known control lines. Based on the results of this study the use of either a disease score or grain yield assessment would be appropriate to categorise breeding lines. However, the use of a disease score may be more suitable as it would allow for the use of un-replicated small plots and require fewer resources than replicated plots for yield assessment.

Within this study only a single isolate of P. syringae pv. syringae was used and consequently the varieties were categorised based on their reaction to this single bacterial isolate. Further work needs to be undertaken to determine if pathogenic variability exists within the P. syringae pv. syringae population as it does within the P. syringae pv. pisi population (Bevan et al. 1995). Therefore, screening of field pea breeding material should be undertaken with some caution until the pathogenic variability of P. syringae pv. syringae populations are established.

For a disease screening nursery to be successful, we believe that early sowing is critical for good disease development. There are many reports in the literature of the association of early sowing with the development of bacterial blight caused by P. syringae pv. pisi (see Hollaway et al. 2007). The field experiments reported in this study were sown early in the season relative to sowing dates recommended for the Horsham area, increasing the likelihood of an epidemic developing.

It is also likely that the high level of disease development within the field was assisted by the above average number of frosts that occurred during the 2006 growing season at Horsham (Table 1). P. syringae pv. syringae, like P. syringae pv. pisi, is an ice nucleating bacterium and there are many reports linking frost with increased occurrence and severity of bacterial blight caused by P. syringae pv. pisi (see Hollaway et al. 2007).

Results of this study have shown that infected stubble can be an important source of inoculum of P. syringae pv. syringae but only poses a significant risk to the following year’s pea crop and not crops in later years. This is in contrast to the findings of Hollaway and Bretag (1997) who reported that P. syringae pv. pisi poses a risk to subsequent field pea crops for 2 years as P. syringae pv. pisi survived for 78 weeks on stubble on the soil surface. This suggests that P. syringae pv. syringae does not withstand environmental conditions as well as P. syringae pv. pisi.

Although artificially infected seed was used in this study, there is limited knowledge of the importance of seed infection in the epidemiology of bacterial blight caused by P. syringae pv. syringae in field pea. This knowledge gap should be addressed.

In light of this study’s findings Australian field pea breeding programs should take this pathogen into consideration for future cultivar development to decrease the risk of commercialising new cultivars susceptible to bacterial blight. A glasshouse test should also be developed to help identify both P. syringae pv. syringae resistant and susceptible breeding lines to further improve the resistance of field pea to bacterial blight. Glasshouse tests could also be used to determine the nature and extent of pathogenic variability in different isolates of P. syringae pv. syringae. This information will enable pea breeders to develop new cultivars with improved resistance to bacterial blight caused by P. syringae pv. syringae. Furthermore, growers need to be reminded of the importance of crop rotation to reduce inoculum levels and avoid growing very susceptible cultivars in bacterial blight-prone areas.