Printer Friendly

Suppression of Rhizoctonia solani AG-8 induced disease on wheat by the interaction between Pantoea, Exiguobacterium, and Microbacteria.


The soil-borne pathogen Rhizoctonia solani AG-8 causes major yield losses in wheat worldwide and is a major constraint to the uptake of sustainable agricultural practices such as direct-drill or minimal tillage and stubble retention, as Rhizoctonia root rot is increased under these practices (Rovira 1986; Pumphrey et al. 1987). A field site at Avon, South Australia, was set up to investigate the effect of different cultivation-seeding methods, crop rotations, and other practices under stubble retention farming systems (Rovira 1986). This soil became suppressive to Rhizoctonia root rot (Roget 1995), following a similar trend to that observed during the development of take-all decline under continuous cereal monoculture (Werker et al. 1991; Hornby 1992). Disease initially increased and then decreased after 4-5 years to negligible levels after 8 years. This occurred under both wheat--wheat and wheat--pasture rotations, and under both direct drill and conventional cultivation. Wiseman et al. (1996) found that this soil was also suppressive to take-all disease and Fusarium crown rot of wheat, and that soil microorganisms were the causal agents of this suppression. Suppression of bare-patch disease of wheat caused by R. solani has also been reported by MacNish (1988) and Lucas et al. (1993). A wide range of microorganisms has been reported to be able to reduce diseases caused by various anastomosis groups of R. solani, including pseudomonads (Homma 1996), actinomycetes (Kulik 1996), Trichoderma and other fungi (Elad 1996), non-pathogenic Rhizoctonia (Sneh 1996), obligate mycoparasites (Van Den Boogert 1996), and soil fauna (Curl and Lartey 1996). There is, however, little information on the organisms involved in suppression of R. solani AG-8 induced disease in wheat cropping systems.

Elucidation of the microorganisms that contribute to suppression of plant diseases within a complex soil microflora is difficult. However, a knowledge of which microorganisms are significant contributors to the suppression of disease is of prime importance to manage and enhance the development of disease suppression in the field. Caldwell et al. (1997) proposes that complex microbial communities can be viewed at various levels of genetic organisation, with a decrease in the level of organisation and complexity of genetic information going from the ecosystem level to the DNA level of organisation. A step-wise approach through progressively simpler levels of organisation was used in this study to simplify a complex ecosystem expressing a disease-suppressive phenotype. The soil is a heterogeneous environment, and it is clear that microbial communities differ depending on their spatial relationship to roots (Lemanceau et al. 1995; Mahaffee and Kloepper 1997). It would seem logical then, that for a particular mechanism of disease suppression the microbes responsible are not evenly distributed in the root--soil environment, and may have particular niche preferences where their relative numbers or activity would be expected to be greater. As it is important to ensure that the information produced is relevant to what is occurring in the field (Madsen 1998), the relevance of each genetic organisation level (communities, populations, cells) to disease suppression was assessed in a simplified microcosm ecosystem to measure the expression of disease suppression.

This paper reports the isolation of 3 groups of bacteria that interact to suppress Rhizoctonia root rot on wheat seedlings using a step-wise approach to progressively narrow down organisms contributing to disease suppression. This stepwise approach involved 4 consecutive steps to (1) determine where suppressive communities were located by assessing microbial communities in soil and root samples from different locations within a field plot for disease suppression, (2) determine which populations of culturable bacteria on selective media within specific communities were correlated with disease suppression, (3) characterise populations which suppress disease by fatty acid methyl ester profiles (FAME) and by BOX-PCR fingerprints, and (4) assess the interaction between microbial groups for their ability to suppress disease.

Materials and methods

Avon soil characteristics and sampling

The development of disease suppression at Avon, South Australia (34[degrees] 14'S, 138[degrees]18'E), and soil characteristics were described by Rovira (1986) and Roget (1995). The soil is an alkaline calcareous sandy loam, pH 8.2, and Avon has predominantly winter rainfall with an annual average of 350 mm. Soil was collected after summer (May 1999) before the start of the 1999 growing-season rains. Avon experiences hot dry summers and there is no summer cropping. Soil was collected from a trial plot that had been under continuous wheat cropping and direct-drill management since 1979 and was suppressive to Rhizoctonia root rot. Collected soil was air dried, sieved to less than 2 mm and stored in sealed plastic bags at room temperature for up to 3 years. This will be referred to as 'soil' in this report and was used as the soil matrix in pot bioassays after sterilisation.

Research methodology.

The step-wise approach used to determine the microorganisms that contributed to disease suppression involved the following 4 consecutive phases.

(1) Determine the location of suppressive communities by comparing rhizosphere, rhizoplane, and root microflora for their ability to suppress disease in pot bioassays.

(2) Determine which populations of culturable bacteria on selective media within specific communities were correlated with disease suppression. This was carried out by comparing cultural populations of bacteria with reduced disease.

(3) Characterision of populations which suppress disease by fatty acid methyl ester profiles (FAME) and by BOX-PCR fingerprints.

(4) Assess the interaction between microbial groups for their ability to suppress disease in pot bioassays.

Pot bioassay to assess disease suppression

Disease suppression by microorganisms was determined by adding non-sterile soil, roots, or microbes into pots containing sterilised Avon soil, R. solani AG-8 (Wiseman et al. 1996), and wheat. The Avon soil was sterilised by autoclaving air-dry soil at 121[degrees]C for 1h, repeated twice, 24h apart each time. Pots were prepared by adding 150g air-dry soil to 300mL non-draining pots, one 8-mm agar plug from the edge of a R. solani AG-8 colony (1/4 strength Potato Dextrose Agar, 7 days growth), followed by an additional 150g air-dry soil and 45 mL sterile distilled water (15% w/v moisture, field capacity). Non-sterile soil and root samples, or microbes, to be assessed for disease suppression, were mixed with the sterilised Avon soil after autoclaving and prior to adding soil to pots. These pots were then preincubated for 2 weeks at 15[degrees]C to allow the pathogen to colonise the soil. Seven surface-sterilised (1% Na hypochlorite for 10min, followed by 3 washes in sterile distilled water) and pregerminated (24 h on moist filter paper) wheat seeds (Triticum aestivum cv. Excalibur) were then planted and thinned to 5 after emergence. Polyethylene beads (ca. 25 g) were added to the soil surface to reduce evaporation. Pots were watered to their initial weight as required, incubated in a controlled-environment room (15[degrees]C, 12 h day night cycle), and plants were harvested 4 weeks after sowing. Shoots were removed from roots just above the crown and dried (4 days at 60[degrees]C) to measure shoot dry weight (shoot DW) per pot. Roots were assessed for root infection given as the percentage of seminal roots (first 10 cm) truncated by R. solani (%RI). Roots infected with R. solani (thinning of the cortex) but not yet truncated counted as half a truncated root. Bioassay experiments were set up as randomised complete block designs (RCBD) and plant growth parameters (shoot DW and %RI) analysed using ANOVA unless otherwise indicated. Comparison of treatment means was by Fisher's protected least significant difference (l.s.d.) at P = 0.05. Control treatments with no added pathogen were not included in analysis of %RI. All data were statistically analysed using GENSTAT 5.4.

Location of microbial communities that suppress disease

Rhizosphere soil, rhizoplane soil, and seedling roots were obtained from wheat seedlings planted in Avon soil that had been air dried for less than 1 year. To produce the seedlings, surface-sterilised and pregerminated wheat seeds (T. aestivum cv. Excalibur) were sown in 400g of soil (15% w/w moisture) in non-draining pots as a layer of seeds covering the soil surface, thereby ensuring all of the soil volume would be influenced by roots. Seeds were then covered with a 1-cm layer of soil. Plants were grown for 2 weeks in a controlled-environment room (15[degrees]C, 12 h day--night cycle). At harvest, loosely adhering soil was removed from roots by shaking and lightly brushing roots and air dried for 24 h before use. This loosely adhering soil will be referred to as rhizosphere soil. Soil tightly adhering to roots was collected by placing roots into 500-mL centrifuge bottles. Roots were covered with sterile water, vortexed (5 s), and then shaken for 15 min on a rotary shaker (200 rpm). Roots were removed and the soil collected by centrifugation and air dried (24 h) before use. This soil will be referred to as rhizoplane soil. The washed seedling roots were cut into ca. 5-mm segments before use.

Rhizosphere soil was added to bioassays at 10, 1.0, 0.1, and 0.01% (w/w). Root samples were added at 0.5, 0.1, and 0.01% (w/w) to bioassays. Rhizoplane soil was added to bioassays at 1% (w/w). Serial dilutions were made by adding 300 g rhizosphere soil, 3 g rhizoplane soil, or 1.5g of roots to sterilised soil to make 3.0 kg and mixed well. For rhizosphere soil, a 1 in 10 dilution series was carried out transferring 30 g to 2.7 kg sterile soil to give the required dilutions. For roots a 1 in 5 followed by 1 in 10 dilution series of root inoculum in 3.0 kg total weight was carried out to give the required dilutions.

Determining which populations within mixed microbial communities suppress disease

To determine which colony type populations were related to disease suppression, 2-pot bioassays were set up and inoculated with a dilution series of rhizosphere soil or root samples (described above). This was to provide a range of levels of infection suppression and plant growth promotion by soil and root samples, respectively. After 4 weeks of growth, bacteria were isolated from either rhizosphere soil or seedling roots in pot bioassays with a range of high to low disease suppression. Rhizosphere soil and root extracts from these pots were plated onto selective media and colony type populations (colonies with the same colony morphology) counted. Based on results from previous experiments, colony type populations isolated from rhizosphere soil were correlated to the level of root infection, and populations from roots were correlated to the corresponding shoot dry weight by regression analysis. Colony type populations associated with reduced root infection or increased plant growth were isolated from rhizosphere soil or roots, respectively, and assessed in bioassays. Colony type populations that reduced diseased in the first bioassay were then isolated from fresh rhizosphere soil or root samples and subject to a second assay. Colony type populations that reduced disease in both assays were then characterised.

Estimation and isolation of bacterial populations from rhizosphere soil and roots

For estimation of bacterial population densities and isolation of bacteria from rhizosphere soil, 5g of rhizosphere soil samples were placed into 10 mL of phosphate buffered saline (PBS, pH 7.3), vortexed briefly and shaken for 15 min on a rotary shaker (200 rpm), and allowed to settle (2-3 min). For estimation of bacterial populations and isolation of bacteria from roots, 0.5g of root material (either crowns plus 2 cm root, 2 cm of root hair zone, or 2 cm of root tips) was washed in PBS until clean of soil, blotted dry, and macerated in 3 mL of PBS.

Ten fold dilution series of rhizosphere soil and root extracts were made in PBS and 100-[micro]L aliquots spread onto 7 selective media to select for a range of bacterial groups and incubated at 25[degrees]C. Cycloheximide (100 [micro]g/mL) was added to all media to suppress fungal growth. The 2 media which proved useful in this work were 4-carbon-source medium (CM, Weaver et al. 1975) used to select for free-living, nitrogen-fixing bacteria, and 1/10-strength tryptic soy agar with 5 [micro]g/mL crystal violet (TCV, Elliott and Des Jardin 1999) was used to select for Gram-negative bacteria. Bacterial colonies which grew on CM medium were counted after 5 days, and colonies on TCV medium were counted after 2 days. Diluted extracts were also plated onto 1/10-strength tryptic soy agar (TSA/10) to select for total aerobic bacteria, TSA/10 after heating rhizosphere soil and root extracts to 60[degrees]C for 15 min (TSA/10-60) to select for heat-resistant bacteria, 1/100-strength tryptic soy agar (TSA/100) to select for oligotrophic bacteria, and King's B medium (KB, Simon and Ridge 1974) to select for fluorescent pseudomonads. Bacterial colonies with the same morphology (size, colour, shape, and texture) on the same medium were categorised as a colony type population. Counts were made of each colony type population as well as total bacterial populations on each medium. Populations were Log10 transformed for analysis. Colony type populations were labelled according to the medium they were isolated on, colony type population number, and isolate number, e.g. CM6.1 was CM6 colony type population, isolate 1.

Correlation analysis

Simple linear regression was used to correlate colony type populations with disease suppression. Colony type populations ([log.sub.10] cfu/g) and proportion of the colony type population (relative to the total population on the selective media) were fitted as independent variables, and percent root infection (rhizosphere soil populations) or shoot dry weight (root populations) were fitted as response variables. Colony type populations which were either positively or negatively correlated to percent root infection or shoot dry weight at P < 0.10 were included in back-regression models, which were simplified by excluding the least significant colony type populations one at a time. Colony type populations isolated from rhizosphere soils that were negatively correlated with root infection, and root populations positively associated with shoot weight, were selected as candidate populations for further evaluation of disease-suppressive ability in pot bioassays. Colony type populations which were present on wheat seedling roots but either not detectable or at a much reduced level in soil were also assessed for plant growth promoting ability.

Assessment of bacterial colony type populations for disease suppression

Colony type populations were assessed for disease suppression in a pot bioassay system as described above, except that soil moisture was adjusted to 15% (w/w) prior to addition to pots and wheat cv. Frame seeds were used, to reflect changes in cultivars used by growers. Bacterial populations were added to this system by mixing a bacterial suspension with sterilised Avon soil 24 h prior to potting up. In the first bioassay to assess disease suppression by colony type populations, individual colonies with the same morphology as the target colony type population from rhizosphere soil or root dilutions plated onto selective media were streaked onto fresh media (one streak per colony, 10 streaks of the same colony type per agar plate and used for inoculation without further treatment). In the second experiment, rhizosphere soil and root extracts were plated onto selective media. Individual colonies of CM6, CM7, CM18, and TCV 14 colony morphologies (9-16 colonies per colony type) were purified by subculturing twice and bacteria from plates containing only a single colony type were suspended in 3 mL of 15% glycerol and stored at -70[degrees]C. These isolates were used in further experiments to assess disease suppression by populations, individual isolates, and for isolate identification. For bioassays, bacterial suspensions were obtained by streaking bacterial isolates onto the medium they were isolated from (minus antibiotics), one isolate per plate, and incubating at 25[degrees]C for 2 days (CM 18 and TCV 14 populations) or 5 days (CM6 and CM7 populations). Cells were harvested in 3 mL of sterile distilled water per plate and isolates from the same colony type population were pooled. Colony type populations thus consisted of 9-16 isolates of the same colony type. Bacterial suspensions were then diluted to the required concentration based on absorbance at 550 nm. CM6, CM7, and TCV14 populations were inoculated into soil at the rate of [10.sup.6] cfu/g soil, and CM18 populations at [10.sup.4] cfu/g soil. For pots inoculated with root-associated bacteria, seeds were dipped into diluted bacterial inoculum (TCV14, [10.sup.6] cfu/mL; CM18, [10.sup.4] cfu/mL in 0.75% xantham gum to aid adhesion of bacteria to seeds) after surface sterilisation of seeds and prior to pre-germination. This same method was used to assess individual isolates from each bacterial group for disease suppression.

Characterisation of disease suppressive populations

Individual isolates from colony type populations that were shown to suppress root infection or to promote growth of infected plants in 2 out of 2 experiments were characterised by (a) their fatty acid methyl ester (FAME) profile, (b) genomic fingerprint using repetitive sequence based PCR (BOX-PCR), and (c) 16S rDNA of representative isolates. FAME profiles were determined using the Microbial Identification System version 4 (MIS) (MIDI Inc. Newark, Delaware). Isolates were subcultured twice on Tryptic Soy Broth Agar (TSBA) for 24 h at 28[degrees]C, bacterial cells (40 [+ or -] 2 mg) were harvested, and fatty acids saponified, methylated, and extracted. Fatty acid profiles were obtained by gas chromatography (Hewlett-Packard model 5890). Fatty acid peaks were identified by the MIS system, and compared with the Sherlock TSBA aerobe library version 4.1 for taxonomic identification. The unweighted pair-group method using arithmetic averages was used for hierarchal cluster analysis and generation of a dendrogram.

BOX-PCR fingerprints were obtained using BOXA1R primers by the method of Versalovic et al. (1994). Genomic DNA was obtained from pure isolates by the method of Ross et al. (2000) and used at 1/10 dilution. PCR products were separated on agarose gels (2% in 0.5% TBE buffer and run at 70 V for 2.75 h) and stained in ethidium bromide solution for visualisation of bands. One representative isolate of each BOXA1R genotype was selected for sequencing of the 16S rDNA gene. The 16S genes were amplified in 2 segments using the Eubacteria primer pairs fD1-Gr (8-800bp fragment) and Gf-rP2 (781-1512 bp fragment) (Weisburg et al. 1991). PCR products were purified using Concert Rapid PCR Purification System (Life Technologies) according to the manufacturer's protocol and amplification products were sequenced using dye-terminator chemistry. DNAMAN ( was used for manipulation and comparison of sequence data. Complete 16S rDNA gene sequences (8-1512 bp) were compared with a sequence database using BlastN search at Genome Net (

Assessment of interaction between bacterial groups for disease suppression

To assess the interaction between bacterial groups with respect to disease suppression, representative isolates were inoculated into pot bioassays as described above. Bacterial groups used were the plant growth promoting Pantoea agglomerans (TCV14 colony type) and Exiguobacterium acetylicum (CM18) and the infection suppressing Microbacteria (CM6-CM7) group. Representative isolates included 3 isolates from the P. agglomerans and E. acetylicum group. The Microbacteria group consisted of 3 isolates from each of the 4 Microbacteria BOXA1R genotype groups. Bacterial groups were inoculated into pots as a 3-way factorial RCBD, with each factor consisting of plus or minus one of the bacterial groups. Population densities added to soil were: P. agglomerans 106 cfu/g soil; E. acetylicum [10.sup.4] cfu/g soil; and Microbacteria [10.sup.6] cfu/g soil (equal densities of each genotype). The amount of R. solani in the soil at the end of bioassays was estimated using a commercially available DNA-based assay (Root Disease Testing Service, South Australian Research and Development Institute, South Australia). This method uses PCR followed by hybridisation to a radiolabelled DNA probe specific for R. solani AG-8 for quantification of pathogen DNA (Ophel-Keller et al. 1999).


Location of microbial communities that suppress disease

In the bioassay system, infestation of sterilised Avon soil with R. solani AG-8 typically resulted in truncation of 60-100% of seminal wheat roots after 4 weeks, with no or few secondary roots present. Inoculation of sterile rhizosphere soil or root inoculum at 0.1% (w/w) or greater significantly reduced the percentage of roots infected (%RI) by R. solani and increased shoot growth compared with control plants inoculated with pathogen only (P < 0.05, Fig. 1). Rhizosphere soil or root inoculum at 0.01% (w/w) had no effect on disease (Fig. 1). The effect on disease was different between rhizosphere soil and root inoculation. Rhizosphere soil inoculum at 0.1% (w/w) significantly reduced root infection to a greater extent than seedling roots (P < 0.05, Fig. 1). Seedling roots only produced a minor reduction in %RI to around 80% (P < 0.05); however, shoot dry weight was significantly greater (P < 0.05) with 0.5% (w/w) root inoculum than with 10% (w/w) rhizosphere soil (Fig. 1). Rhizoplane soil at 1% (w/w) inoculation reduced root infection to the same degree as rhizosphere soil and increased shoot dry weight to the same extent or slightly higher than with root inoculum washed free of soil. The effects of (a) soil inoculum reducing infection and (b) root inoculation increasing growth of infected plants were consistent in all 3 experiments where soil and root inoculum were compared.


Determining which populations within mixed microbial communities suppress disease

Identification of soil bacterial populations correlated with reduced infection and root bacterial populations correlated with extent of shoot growth was undertaken by comparing colony type populations on selective media with %RI or shoot weight, respectively, by regression analysis. In total, 77 colony type populations were counted and assessed by regression analysis, 19 populations from TSA/10, 14 from TSA/10-60, 11 from TSA/100, 14 from TCV, 18 from CM, and 1 from KB medium. Eight colony type populations isolated from rhizosphere soil on CM medium showed a negative correlation with root infection and were assessed in a bioassay. Two of these populations, CM6 (0.5-1 mm, translucent, circular, yellow colony morphology after 5 days) and CM7 (0.5-1 mm, translucent, circular, white colony morphology after 5 days), significantly reduced root infection in the first assay, and were therefore assessed in a second experiment. The other 6 populations did not produce a significant reduction in root infection in the first assay. In the second bioassay, CM6 and CM7 populations significantly reduced root infection by Rhizoctonia (Fig. 3).


Analysis of data obtained with root-associated colony type populations indicated that only one population on TCV media (TCV10) was correlated with changes in shoot growth. However, this population produced only a small and non-significant (P > 0.05) promotion of plant growth in a pot bioassay. It was noted, though, that 2 root-associated colony type populations, which occurred at relatively high numbers on wheat roots, were either not detected in rhizosphere soil (CM18) or present at much lower populations (TCV14) compared with the total population on selective media (Fig. 2). TCV14 colonies were 3 mm, flat, translucent, circular, and brown-yellow after 2 days. CM18 colonies were 1 mm, opaque, circular, and yellow after 5 days. These 2 populations substantially increased the growth of infected plants in both bioassays. Results from the second bioassay are shown in Fig. 3, with CM18 and TCV14 populations significantly (P < 0.05) increasing plant growth, yet producing only a small non-significant (P > 0.05) reduction in root infection.


Characterisation of disease suppressive populations

Each isolate from the CM6, CM7, CM18, and TCV14 populations (consisting of 16, 12, 9, and 9 isolates, respectively) was characterised using the MIDI GC-FAME system. The 2 plant growth promoting root-associated populations were identified as homogenous populations of Pantoea agglomerans (TCV14, Euclidean distance within the group of <3.5) or Exiguobacterium acetylicum (CM18, Euclidean distance within the group of <2), with high similarity (>0.8) to reference strains in the Microbial Identification System (MIS) TSBA aerobic library (Fig. 4). The infection-suppressing soil populations, CM6 and CM7, consisted of overlapping and more diverse groups of isolates that were divided into 4 main clusters (indicated as MB, Bs, KC, and KM in Fig. 4), with generally low similarity to reference strains in the MIS library. Cluster 1 contained mainly CM7 isolates with closest similarity (0.02-0.2) to Micrococcus spp. or Bacillus flexus (MB group, Euclidean distance < 19). Cluster 2 contained two CM6 and one CM7 isolate with closest similarity (0.02-0.2) to Bacillus subtilis (Bs group, Euclidean distance <12). Cluster 3 contained mainly CM6 isolates with closest similarity (0.6-0.9) to Kocuria kristinae or Curtobacterium spp. (KC group, Euclidean distance <16). Cluster 4 contained mainly CM6 isolates with closest similarity (0.6-0.9) to Kocuria kristinae or Microbacterium spp. (KM group, Euclidean distance <15). These clusters are referred to as the KM, KC, Bs, or MB groups based on their similarity to MIS library strains.


Isolates from the 4 beneficial populations, TCV14, CM18, CM6, and CM7, were characterised by BOX-PCR DNA fingerprinting using the BOXA1R primer. Only two CM7 isolates from the MB FAME group were characterised by BOX-PCR, as isolates from this group did not reduce infection in bioassays (data not shown). Results from PCR fingerprinting indicated the same groupings of isolates as revealed by fatty acid profiles, except for the KM FAME group for which there were 2 distinct BOX-PCR profiles, designated KMa and KMb (Fig. 5). BOXA1R profiles of 3 representative isolates from each BOXA1R group are shown in Fig. 5 and indicated in Fig. 4. CM6 and CM7 isolates within each of the BOXA1R groupings, KMa, KMb, KC, and Bs, had similar BOX-PCR profiles with many common bands; however, within each group there was some divergence in the presence or absence of some bands (Fig. 5). The KMb and KC groups were the most diverse in banding patterns. Isolates within each of the TCV14 and CM18 groups had highly similar or identical BOX-PCR profiles. There was little similarity in banding pattern between isolates from different BOXA1R groups. A representative isolate from each BOXA1R group (indicated in Fig. 4) was chosen and the 16S rDNA genes amplified, sequenced, and compared with sequences in GenBank. The 16S sequence from TCV14.2 was homologous to Pantoea agglomerans strains (Gram negative, gamma Proteobacteria, Enterobacteriaceae). The 16S sequence from CM18.1 was homologous to Exiguobacterium acetylicum strains (Gram positive, Firmicutes, Bacillus group). Data from the 4 sequenced CM6 isolates indicated that these were closely related to the Microbacteria group (Gram positive, high GC, Firmicutes, Actinobacteria, Actinobacteridae, Actinomycetales, Micrococcineae, Microbacteriaceae). There was 95% or greater 16S rDNA sequence similarity between the CM6 isolates. The similarity between the CM6 isolates and closely related species is shown in Fig. 6. The CM18 isolate CM18.1 had 80% 16S rDNA sequence similarity to the CM6 isolates. The TCV14 isolate TCV14.2 had only 77% sequence similarity to the other groups.


Assessment of interaction between bacterial groups for disease suppression

In initial experiments, single isolates inoculated into pot bioassays were ineffective in reproducing the infection suppressing or plant growth promoting responses observed with populations of 9-16 isolates such as shown in Fig. 3. Combinations of single isolates of different bacterial genotypes could produce a response; however, this was inconsistent. To overcome this problem it was decided to inoculate bioassays with a subset of 3 isolates from each BOX-PCR group (shown in Fig. 5) in a factorial design to assess the interaction between groups. Groups consisted of 3 Pantoea, 3 Exiguobacterium, and 12 Microbacteria (3 isolates from 4 BOX-PCR genotypes). Co-inoculation with all three groups produced a consistently higher degree of disease suppression than inoculation with a single group or pairs of groups (Fig. 7). Shoot growth of plants inoculated with all 3 groups was similar to plant growth when inoculated with non-sterile suppressive Avon soil at 10% w/w (644 and 678 mg/pot, respectively, P > 0.05), although root infection was significantly (P < 0.05) higher (35%RI) with tri-bacterial inoculation than with Avon rhizosphere soil inoculation (14%RI, Fig. 7). The interaction between the 3 bacterial groups was highly significant for both shoot dry weight (P < 0.001) and per cent root infection (P < 0.001).


R. solani AG-8 DNA levels measured in soil at the end of the bioassay in pots co-inoculated with all 3 bacterial groups were similar (P > 0.05) to R. solani DNA levels measured in pots with no added bacterial inoculum (232 and 266 pg DNA/g soil, respectively, Fig. 7). These are considered high levels of pathogen inoculum. However, inoculation of pots with non-sterile suppressive Avon soil significantly (P < 0.01) reduced pathogen DNA to moderate levels (66 pg DNA/g soil, Fig. 7).

In all experiments (>4), co-inoculation with all 3 groups consistently reduced root infection by R. solani and increased plant growth to a greater extent than inoculation of single or pairs of groups. When measured in soil at the end of 3 pot bioassays, levels of R. solani DNA in soil were not significantly different (P > 0.05) between disease controls with no added bacteria and co-inoculation with the 3 groups of bacteria.


Suppression of soil-borne diseases depends on complex interactions between the host plant, pathogen, environment, and microorganisms. Past research on bare patch disease of wheat caused by R. solani has concentrated on the plant, pathogen, and environment, but the ecology of the root-soil microflora and its role in disease suppression is not well understood. At the Avon field site, it was known that disease was being suppressed by soil microorganisms (Wiseman et al. 1996) but it was not known which organisms contributed to this suppression. Traditional methods of screening for disease-suppressive microorganisms by isolation of target groups of microbes and individually subjecting them to in vitro inhibition assays against the pathogen were not thought to be appropriate. This is because (1) there often is no good relationship between in vitro inhibition and disease control (Elsherif and Grossmann 1994; Reddy et al. 1994; Castejon-Munoz and Oyarzun 1995), (2) interactions between microbes may be important (Simon and Sivasithamparam 1988; Becker et al. 1997; Duijff et al. 1999), and (3) a prior selection of a target group of organisms for study may preclude other important microbial components. For these reasons it was decided to take an approach based on complex adaptive ecosystem ecology (Caldwell et al. 1997; Molin and Molin 2000), i.e. to progressively move from the ecosystem level of organisation to the microbial isolate level as a means to determine which organisms contribute to disease suppression without presupposing a target microbial group. By physically partitioning the root-soil environment it was possible to determine that the soil microflora suppressed root infection and root-associated microflora promoted the growth of infected plants. Populations of microbes grouped by colony morphology from these 2 communities were then assessed for either infection suppression or plant growth promotion. By using a range of selective media it was possible to identify populations (classified by colony type) that were associated with each of these traits. Because these populations were culturable, it was possible to isolate and introduce them into a bioassay to determine which of the correlated populations could actually produce the response (infection suppression or plant growth promotion) that they were originally correlated with. The colony type populations that produced a response upon inoculation were then characterised to determine the identity of the individual isolates involved. This process resulted in the isolation of 3 groups of bacteria, Pantoea agglomerans, Exiguobacterium acetylicum, and Microbacteria (collectively referred to as PEM), that had not previously been associated with the suppression of Rhizoctonia root rot disease. These groups interact to provide significant reduction in disease but do not reduce pathogen levels in the soil.

The first stage in this step-wise process was to determine which spatially different microbial communities within the root-soil environment were able to suppress disease. It is known that microbial communities vary depending on their location in relation to roots (Lemanceau et al. 1995; Mahaffee and Kloepper 1997; Barnett et al. 1999a), and the suppressiveness of the microflora in soil or root samples can be measured in pot bioassays (Shipton et al. 1973; Cook and Rovira 1976). When soil and root samples from the Avon suppressive soil were compared for their ability to suppress R. solani induced disease it was evident that microbial communities from different locations produced different responses. An initial experiment indicated that suppression was greater (i.e. decreasing %RI and increasing shoot growth) the closer the soil sample tested was to wheat roots, with rhizosphere soil having the highest suppression. However, inoculation of bioassays with wheat seedling roots had a reduced effect on %RI compared with soil inoculation but produced more shoot growth. A second experiment, with dilutions of soil and root inoculum, confirmed this response. That very dilute (0.1% w/w) live soil samples were able to suppress disease provides good evidence that suppression is biologically based (Shipton et al. 1973; Cook and Rovira 1976). From these results it was concluded that there are at least 2 spatially separated communities combining to suppress the expression of disease in Avon soil by 2 different mechanisms. That is, soil microorganisms reduced root infection by R. solani and root associated microorganisms promoted the growth of infected plants. Rhizoplane soil (soil tightly adhering to roots) appeared to be a zone of overlapping root and rhizosphere soil communities, reducing root infection to levels similar to rhizosphere soil but with significantly higher shoot dry weight, indicating the presence of both soil and root associated microbes. Recognising that rhizosphere soil and root communities were functionally different, with respect to disease suppression, meant that rhizosphere soil populations needed to be assessed for their relationship to reduced pathogen infection, and root populations needed to be assessed for their relationship to plant growth.

Statistical approaches correlating soil biota and abiotic factors with disease expression have been used previously to relate biological components to functions of interest (Oyarzun et al. 1998). James and McCulloch (1990), however, point out many of the limitations of using statistical approaches, including the difficulty in interpreting cause and effect based on correlative evidence. In this work, correlating bacterial colony type populations from Avon soil to disease suppression was used only to give an indication of which populations may be related to suppression. A bioassay system was then used to provide sound evidence that the colony type populations correlated with reduced infection or increased plant growth could reduce disease. The results with microbial communities from Avon soil indicated that most of the bacterial populations that correlated with increased disease suppression had little effect when inoculated into pot bioassays, and no root populations were identified by regression analysis that consistently promoted plant growth. Despite these failings, 2 bacterial populations, CM6 and CM7, were identified by the regression analysis method and the 2 root populations, TCV14 and CM18, were identified by visual observation of these colony types on root extracts plated onto selective media.

The use of colony morphology on solid agar media to group and select bacteria reflects only a small proportion of the total microflora. However, the use of a range of different selective media greatly increases the number of different bacteria that can be isolated (Joseph et al. 2003) and allows the ready isolation of purified individual bacterial cultures, an essential requirement for assessing the ability of these cells to suppress disease and to enable further characterisation. Colony type populations shown to reduce disease (TCV14, CM18, CM6, and CM7) were characterised by GC-FAME and BOX-PCR. Isolates within each of the TCV14 and CM18 populations produced near identical GC-FAME and BOX-PCR profiles, indicating homogeneous populations. GC-FAME and 16S rDNA sequence comparison produced identical and unambiguous taxonomic identification, i.e. Pantoea agglomerans (TCV14 isolates) and Exiguobacterium acetylicum (CM18 isolates). GC-FAME was less effective in identifying the CM6 and CM7 isolates, with these isolates having only a very low similarity to isolates within the Microbial Identification System library. The 16S sequence data clearly placed the KMa, KMb, and BS isolates within the Microbacterium genus, and the KC isolate in the closely related Detolaasinbacter. These isolates are collectively referred to as Microbacteria in this paper. It should be noted that this is not a comprehensive taxonomic study and further characterisation is required to determine their exact taxonomic position.

The similarities between groupings of isolates using FAME and PCR profiles based on repetitive sequences has been reported by other researchers (Lemanceau et al. 1995; Tonso et al. 1995). Repetitive sequence PCR profiles may have an advantage in being more discriminatory than FAME (Hollaway et al. 1997). For example, BOXA1R PCR divided the KM group by FAME analysis into 2 distinct genotypic groups. Phenotype and genotype characterisation should be considered complementary, as isolates with the same genotypic profiles can have different phenotypic characteristics (Rainey et al. 1994; Barnett et al. 1999b). By combining both genetic and phenotypic techniques, isolates can be selected for further study with confidence that they are indeed representative of a broader group of isolates.

Pantoea, Exiguobacterium, and Microbacteria have not previously been associated with the suppresison of Rhizoctonia diseases; however, they have been reported to have beneficial properties. The beneficial properties of P. agglomerans include nitrogen fixation (Haahtela et al. 1988), formation of symplasmata on rice (Achouak et al. 1996), phosphate solubilisation (Kim et al. 1997), increasing rhizosphere soil aggregation with extracellular polysaccharides (Amellal et al. 1999), as well as control of post-harvest moulds on oranges (Teixido et al. 2001) and control of fire blight on apples and pears (Stockwell et al. 2002), indicating that this is a very versatile and beneficial group of bacteria. Although Exiguobacterium has not previously been reported to have beneficial effects on plants, they are closely related to the Bacillus group (Farrow et al. 1994) that contains a number of species shown to have biocontrol effects on plant pathogens. An Exiguobacterium sp. has been reported to have a probiotic effect on shrimp larvae as a mixture with a Microbacterium sp. (Orozco-Medina et al. 2002). Microbacterium spp. appear to be commonly associated with plants, and have been isolated from the phyllosphere of grasses (Behrendt et al. 2001) and wheat (Legard et al. 1994), as an endophyte of stems (Zinniel et al. 2002) and roots (Sturz and Kimpinski 2004), and from seed spermosphere (McKellar and Nelson 2003). Regarding suppression of disease, Microbacterium spp. have been reported to control Pythium damping-off (McKellar and Nelson 2003) and Anthurium blight (Fukui et al. 1999) as part of bacterial consortia and also having activity against root-lesion nematodes (Sturz and Kimpinski 2004). Microbacterium sp. were also enriched in the wheat rhizosphere in the presence of 2,4-diacetylphloroglucinol-producing Pseudomonas fluorescens that are associated with take-all decline (Landa et al. 2003). P. agglomerans and Microbacterium spp. are clearly associated with beneficial effects on plants, and strains of Exiguobacterium and Microbacterium have beneficial effects as part of consortia. The results reported in this paper clearly show that the interaction between Pantoea, Exiguobacterium, and Microbacteria is important in suppressing Rhizoctonia disease in the Avon suppressive soil. This suggests that Pantoea, Exiguobacterium, and Microbacteria may have a greater effect on suppressing plant diseases than has been previously recognised. If phylogeneticaly diverse bacteria act as consortia this would be impossible to detect by studying organisms in isolation. Further work still needs to be done to determine if Pantoea, Exiguobacterium, and Microbacteria contribute to disease suppression in other soils and cropping systems.

The suppression of disease by the inoculated Pantoea, Exiguobacterium, and Microbacteria bacterial groups was not due to reduction of pathogen inoculum. R. solani AG-8 can be detected (Whisson et al. 1995) and quantified (Ophel-Keller et al. 1999) in DNA extracted from soil samples using R. solani AG-8 specific DNA probes. When this was done at the end of 3 experiments, there was no significant difference between levels of pathogen DNA in the soil between soil infected with R. solani only and soil inoculated with R. solani and Pantoea, Exiguobacterium, or Microbacteria, either singly or combined. This indicates that gross reduction of the pathogen in the soil is not a mechanism of action by the PEM bacteria, at least not during the short term of the 4-week seedling bioassay. Pathogen DNA was substantially reduced when soil was inoculated with non-sterile Avon soil, indicating that there are organisms unaccounted for that do reduce pathogen inoculum in the soil. Trichoderma, Bacillus, and Actinomycete strains that inhibit R. solani have been isolated from Avon soil based on in vitro pathogen inhibition assays (Yang et al. 2005a, 2005b) and may account for the reduction of the pathogen DNA in pots inoculated with the suppressive soil.

Management of R. solani disease suppression has important implications in dryland cropping systems as the pathogen appears to be ubiquitous in these systems, both in Australia and overseas, and has a wide crop host range. To understand the microbial ecology involved in this process we need to define the structure or spatial arrangement of the organisms involved and how their populations and activities are regulated (Molin and Molin 2000). The take-all prediction model (Roget 2001) is an example of how knowledge of these factors can be used in a practical sense to manage a specific component of the soil microflora to reduce disease and increase yield. In this model, the levels of take-all pathogen are quantified in soil by a DNA-based assay, and the effect of seasonal rainfall and how it affects the pathogen population are known and included in the predictive model. The first step in being able to predict and manage soil suppressiveness is to understand which organisms are involved and to be able to detect and quantify these organisms. In this study, a target group of organisms that contribute to suppression has been established. Further work is being undertaken to determine if the PEM bacteria are suitable indicator organisms to monitor and predict disease suppression and whether they can be used to modify disease prediction models based on pathogen DNA levels.


The use of a step-wise approach to progress through decreasingly complex organisation levels proved to be a successful mechanism to determine which organisms are involved in the suppression of R. solani induced disease. As a result of this process, it was concluded that there are at least 3 groups of bacteria that interact to suppress R. solani induced disease, but these do not reduce pathogen inoculum density. At least 2 mechanisms are involved in this process: soil-associated Microbacteria suppress root infection by the pathogen, and root-associated Pantoea agglomerans and Exiguobacterium acetylicum promote the growth of infected plants. These 3 groups have not previously been reported to be involved with suppression of Rhizoctonia root rot on wheat. The next stage of this work is to determine the distribution of these organisms in wheat ecosystems and how farm management practices affect on their populations and function, to aid in the development and management of disease-suppressive soils.


The authors thank the Grains Research and Development Corporation, Australia, for financial support for S.J.B. through a Post Doctoral Fellowship (BAR01), B. Hawke for assistance with FAME profiles, P. Harvey for advice on the molecular component, and K. Ophel-Keller and A. Mckay for estimation of R. solani DNA.

Manuscript received 12 August 2005, accepted 19 April 2006


Achouak W, Villemin G, Balandreau J, Heulin T (1996) Specificity of root colonization by symplasmata-forming Pantoea agglomerans. In 'Biological nitrogen fixation associated with rice production'. (Ed. M Rahman) pp. 191-201. (Kluwer Academic Publishers: Great Britain)

Amellal N, Bartoli F, Villemin G, Talouzte A, Heulin T (1999) Effects of inoculation of EPS-producing Pantoea agglomerans on wheat rhizosphere aggregation. Plant and Soil 211, 93-101. doi: 10.1023/A:1004403009353

Barnett S J, Alami Y, Singleton I, Ryder MH (1999b) Diversification of Pseudomonas corrugata 2140 produces new phenotypes altered in GC-FAME, BIOLOG, and in vitro inhibition profiles and taxonomic identification. Canadian Journal of Microbiology 45, 287-298. doi: 10.1139/cjm-45-4-287

Barnett SJ, Singleton I, Ryder MH (1999a) Spatial variation in populations of Pseudomonas corrugata 2140 and pseudomonads on take-all diseased and healthy root systems of wheat. Soil Biology and Biochemistry 31,633-636. doi: 10.1016/S0038-0717(98)00160-6

Becker DM, Kinkel LL, Schottel JL (1997) Evidence for interspecies communication and its potential role in pathogen suppression in a naturally occurring disease suppressive soil. Canadian Journal of Microbiology 43, 985-990.

Behrendt U, Ulrich A, Schumann P (2001) Description of Microbacterium foliorum sp. nov., and Microbacterium phyllosphaerae sp. nov. isolated from the phyllosphere of grasses and the surface litter after mulching the sward, and reclassification of Aureobacterium resistens (Funke et al. 1998) as Microbacterium resistens comb. nov. International Journal of Systematic and Evolutionary Microbiology 51, 1267-1276.

Caldwell DE, Wolfaardt GM, Korber DR, Lawrence JR (1997) Do bacterial communities transcend Darwinism. Advances in Microbial Ecology 15, 105-191.

Castejon-Munoz M, Oyarzun PJ (1995) Soil receptivity to Fusarium solani f. sp. pisi and biocontrol of root rot of pea. European Journal of Plant Pathology 101, 35-49. doi: 10.1007/BF01876092

Cook RJ, Rovira AD (1976) The role of bacteria in the biological control of Gaeumannomyces graminis by suppressive soils. Soil Biology and Biochemistry 8, 269-273. doi: 10.1016/0038-0717(76)90056-0

Curl EA, Lartey RT (1996) Role of soil fauna in biological control of Rhizoctonia. In 'Rhizoctonia species: Taxonomy, molecular biology, ecology, pathology and disease control'. (Eds B Sneh, S Jabaji-Hare, S Neate, G Dijst) pp. 217-226. (Kluwer Academic Publishers: The Netherlands)

Duijff BJ, Recorbet G, Bakker PAHM, Loper JE, Lemanceau P (1999) Microbial antagonism at the root level is involved in the suppression of Fusarium will by the combination of non-pathogenic Fusarium oxysporum Fo47 and Pseudomonas Putida WCS358. Phytopathology 89, 1073-1079.

Elad Y (1996) Bacterial and fungal cell-wall hydrolytic enzymes in relation to biological control of Rhizoctonia solani. In 'Rhizoctonia species: Taxonomy, molecular biology, ecology, pathology and disease control'. (Eds B Sneh, S Jabaji-Hare, S Neate, G Dijst) pp. 455-462. (Kluwer Academic Publishers: The Netherlands)

Elliott ML, Des Jardin EA (1999) Comparison of media and diluents for enumeration of aerobic bacteria from bermuda grass golf course putting greens. Journal of Microbiological Methods 34, 193-202. doi: 10.1016/S0167-7012(98)00088-8

Elsherif M, Grossmann F (1994) Comparative investigations on the antagonistic activity of flourescent pseudomonads against Gaeumannomyces graminis var. tritici in vitro and in vivo. Microbiological Research 149, 371-377.

Farrow JAE, Wallbanks S, Collins MD (1994) Phylogenetic interrelationships of the round-spore-forming Bacilli containing cell walls based on lysine and the non-spore-forming genera Caryophanon, Exiguobacterium, Kurthia and Planococcus. International Journal of Systematic Bacteriology 44, 74-82.

Fukui R, Fukui H, Alvarez AM (1999) Comparisons of single versus multiple bacterial species on biological control of anthurium blight. Phytopathology 89, 366-373.

Haahtela K, Laakso T, Nurmiaho EL, Korhonen TK (1988) Effects of inoculation of Poa pratensis and Triticum aestivum with root-associated, N2-fixing Klebsiella, Enterobacter and Azospirillum. Plant and Soil 106, 239-248. doi: 10.1007/BF02371219

Hollaway GJ, Gillings MR, Fahy PC (1997) Use of fatty acid profiles and repetitive element polymerase chain reaction (PCR) to assess the genetic diversity of Pseudomonas syringae pv. pisi and Pseudomonas syringae pv. syringae isolated from field peas in Australia. Australasian Plant Patholology 26, 98-108. doi: 10.1071/AP97015

Homma Y (1996) Antibiotic and siderophore producing bacteria. In 'Rhizoctonia species: Taxonomy, molecular biology, ecology, pathology and disease control'. (Eds B Sneh, S Jabaji-Hare, S Neate, G Dijst) pp. 445-453. (Kluwer Academic Publishers: The Netherlands)

Hornby D (1992) New information about take-all decline and its relevance to research on the control of take-all by biological control agents. In 'Biological control of plant diseases'. (Eds EC Tjamos, GC Papavizas, RJ Cook) pp. 95-98. (Plenum Press: New York)

James FC, McCulloch CE (1990) Multivariate analysis in ecology and systematics: panacea or Pandora's box. Annual Review of Ecology and Systematics 21, 129-166.

Joseph SJ, Hugenholtz P, Sangwan CA, Janssen PH (2003) Laboratory cultivation of widespread and previously uncultured soil bacteria. Applied and Environmental Microbiology 69, 7210-7215. doi: 10.1128/AEM.69.12.7210-7215.2003

Kim KY, McDonald GA, Jordan D (1997) Solubilization of hydroxyapatite by Enterobacter agglomerans and cloned Escherichia coli in culture medium. Biology and Fertility of Soils 24, 347-352. doi: 10.1007/s003740050256

Kulik MM (1996) Actinomycetes, cyanobacteria and algae. In 'Rhizoctonia species: Taxonomy, molecular biology, ecology, pathology and disease control'. (Eds B Sneh, S Jabaji-Hare, S Neate, G Dijst) pp. 463-471. (Kluwer Academic Publishers: The Netherlands)

Landa BB, Mavrodi DM, Thomashow LS, Weller DM (2003) Interactions between strains of 2,4-diacetylphloroglucinol-producing Pseudomonas fluorescens in the rhizosphere of wheat. Phytopathology 93, 982-994.

Legard DE, McQuilken MP, Whipps JM, Fenlon JS, Fermor TR, Thompson IP, Bailey MJ, Lynch JM (1994) Studies of the seasonal changes in the wheat microbial populations on the phyllosphere of spring wheat as a prelude to the release of a genetically modified microorganism. Agriculture, Ecosystems & Environment 50, 87-101. doi: 10.1016/0167-8809(94)90128-7

Lemanceau P, Corberand T, Gardan L, Latour X, Laguerre G, Boeufgras JM, Alabouvette C (1995) Effect of two plant species, flax (Linum usitatissinum L.) and tomato (Lycopersicon esculentum Mill.), on the diversity of soilborne populations of fluorescent pseudomonads. Applied and Environmental Microbiology 61, 1004-1012.

Lucas P, Smiley RW, Collins HP (1993) Decline of Rhizoctonia root rot on wheat in soils infested with Rhizoctonia solani AG 8. Phytopathology 83, 260-265.

MacNish GC (1988) Changes in take-all (Gaeumannomyces graminis var. tritici), Rhizoctonia root rot (Rhizoctonia solani) and soil pH in continuous wheat with annual applications of nitrogenous fertiliser in Western Australia. Australian Journal of Experimental Agriculture 28, 333-341. doi: 10.1071/EA9880333

Madsen EL (1998) Epistemology of environmental microbiology. Environmental Science & Technology 32, 429-439. doi: 10.1021/ es970551y

Mahaffee WF, Kloepper JW (1997) Temporal changes in the bacterial communities of soil, rhizosphere, and endorhiza associated with field-grown cucumber (Cucumis sativus L.). Microbial Ecology 34, 210-223. doi: 10.1007/s002489900050

McKellar ME, Nelson EB (2003) Compost-induced suppression of Pythium damping-off is mediated by fatty-acid-metabolizing seed-colonising microbial communities. Applied and Environmental Microbiology 69, 452-460. doi: 10.1128/AEM. 69.1.452-460.2003

Molin J, Molin S (2000) Complex adaptive systems ecology. Advances in Microbial Ecology 16, 233-275.

Ophel-Keller K, McKay A, Driver F, Curren J (1999) The cereal root disease testing service. In 'Proceedings of the 1st Australasian Soilborne Disease Symposium'. (Ed. RC Magarey) pp. 63-64. (Bureau of Sugar Experiment Stations: Brisbane, Qld)

Orozco-Medina C, Meada MAM, Lopez CA (2002) Effect of aerobic Gram-positive heterotrophic bacteria associated with Artemia franciscana cysts on the survival and development of its larvae. Aquaculture 213, 15-29. doi: 10.1016/S0044-8486(02)00026-1

Oyarzun PJ, Gerlagh M, Zadoks JC (1998) Factors associated with soil receptivity to some fungal root rot pathogens of peas. Applied Soil Ecology 10, 151-169. doi: 10.1016/S0929-1393(98)00042-0

Pumphrey FV, Wilkins DE, Hane DC, Smiley RW (1987) Influence of tillage and nitrogen fertiliser on Rhizoctonia root rot (bare patch) of winter wheat. Plant Disease 71, 125-127.

Rainey PB, Bailey MJ, Thompson IP (1994) Phenotypic and genotypic diversity of fluorescent pseudomonads isolated from field-grown sugar beet. Microbiology 140, 2315-2331.

Reddy MS, Hynes RK, Lazarovits G (1994) Relationship between in vitro growth inhibition of pathogens and suppression of preemergence damping-off and postemergence root rot of white bean seedlings in the greenhouse by bacteria. Canadian Journal of Microbiology 40, 113-119.

Roget DK (1995) Decline in root rot (Rhizoctonia solani AG-8) in wheat in a tillage and rotation experiment at Avon, South Australia. Australian Journal of Experimental Agriculture 35, 1009-1013. doi: 10.1071/EA9951009

Roget DK (2001) Prediction modelling of soilborne plant diseases. Australasian Plant Pathology 30, 85-89. doi: 10.1071/AP01005

Ross IL, Alami Y, Harvey PR, Achouak W, Ryder MH (2000) Genetic diversity and biological control activity of novel species of closely related pseudomonads isolated from wheat field soils in South Australia. Applied and Environmental Microbiology 66, 1609-1616. doi: 10.1128/AEM.66.4.1609-1616.2000

Rovira AD (1986) Influence of crop rotation and tillage on Rhizoctonia bare patch of wheat. Phytopathology 76, 669-673.

Shipton PJ, Cook RJ, Sitton JW (1973) Occurrence and transfer of a biological factor in soil that suppresses take-all of wheat in eastern Washington. Phytopathology 63, 511-517.

Simon A, Ridge EH (1974) The use of ampicillin in a simplified selective media for the reisolation of fluorescent pseudomonads. Journal of Applied Bacteriology 37, 459-460.

Simon A, Sivasithamparam K (1988) Interactions among Gaeumannomyces graminis var. tritici, Trichoderma koningii, and soil bacteria. Canadian Journal of Microbiology 34, 871-876.

Sneh B (1996) Non pathogenic isolates of Rhizoctonia spp. (np-R) and their role in biological control. In 'Rhizoetonia species: Taxonomy, molecular biology, ecology, pathology and disease control'. (Eds B Sneh, S Jabaji-Hare, S Neate, G Dijst) pp. 473-483. (Kluwer Academic Publishers: The Netherlands)

Stockwell VO, Hohnson KB, Sugar D, Loper JE (2002) Antibiosis contributes to biological control of fire blight by Pantoea agglomerans strain Eh252 in orchards. Phytopathology 92, 1202-1209.

Sturz AV, Kimpinski J (2004) Endoroot bacteria derived from marigolds (Tagetes spp.) can decrease soil population densities of root-lesion nematodes in the potato root zone. Plant and Soil 262, 241-249. doi: 10.1023/B:PLSO.0000037046.86670.a3

Teixido N, Usall J, Palou L, Asensio A, Numes C, Vinas I (2001) Improved control of green and blue moulds of oranges by combining Pantoea agglomerans (CPA-2) and sodium bicarbonate. European Journal of Plant Pathology 107, 685-694.

Tonso NL, Matheson VG, Holben WE (1995) Polyphasic characterisation of a suite of bacterial isolates capable of degrading 2,4-D. Microbial Ecology 30, 3-24. doi: 10.1007/BF00184510

Van Den Boogert PHJF (1996) Mycoparasitism and biocontrol. In 'Rhizoctonia species: Taxonomy, molecular biology, ecology, pathology and disease control'. (Eds B Sneh, S Jabaji-Hare, S Neate, G Dijst) pp. 485-493. (Kluwer Academic Publishers: The Netherlands)

Versalovic J, Schneider M, de Bruijn F, Lupski R (1994) Genomic fingerprinting of bacteria using repetitive sequence-based polymerase chain reaction. Methods in Molecular and Cellular Biology 5, 25-40.

Weaver PK, Wall JD, Gest H (1975) Characterisation of Rhodopseudomonas capsulata. Archives of Microbiology 105, 207-216. doi: 10.1007/BF00447139

Weisburg WG, Barns SM, Pelletier DA, Lane DJ (1991) 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173, 697-703.

Werker AR, Gilligan CA, Hornby D (1991) Analysis of disease-progress curves for take-all in consecutive crops of winter wheat. Plant Pathology 40, 8-24.

Whisson DL, Herdina PH, Francis L (1995) Detection of Rhizoctonia solani AG-8 in soil using a specific DNA probe. Mycological Research 99, 1299-1302.

Wiseman BM, Neate SM, Ophel Keller K, Smith SE (1996) Suppression of Rhizoctonia solani anastomosis group 8 in Australia and its biological nature. Soil Biology and Biochemistry 28, 727-732. doi: 10.1016/0038-0717(95)00178-6

Yang H, Ryder MH, Tang W (2005a) Characterisation and identification of Trichoderma isolates from a South Australian soil suppressive to Rhizoctonia solani on wheat. Shandong Science 18, 36-49.

Yang H, Ryder MH, Tang W (2005b) Isolation and biocontrol potential of bacteria and aetinomycetes from soils suppressive to Rhizoctonia bare-patch disease in South Australia. Shandong Science 18, 68-77.

Zinniel DK, Lambrecht P, Harris NB, Feng Z, Kuczmarski D, Higley P, Ishimaru CA, Arunakumari A, Barletta RG, Vidaver AK (2002) Isolation and characterisation of endophytic colonizing bacteria from agronomic crops and prairie plants. Applied and Environmental Microbiology 68, 2198-2208. doi: 10.1128/AEM.68.5.2198-2208.2002

Stephen J. Barnett (A,B,C,) David K. Roget (A), and Maarten H. Ryder (A)

(A) CSIRO Land and Water, PMB 2, Glen Osmond, SA 5064, Australia.

(B) South Australian Research and Development Institute, Field Crops Pathology Unit, GPO Box 397, Adelaide, SA 5001, Australia.

(C) Corresponding author. Email:
COPYRIGHT 2006 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Barnett, Stephen J.; Roget, David K.; Ryder, Maarten H.
Publication:Australian Journal of Soil Research
Geographic Code:8AUST
Date:Nov 1, 2006
Previous Article:The impact of crop residue amendments and lime on microbial community structure and nitrogen-fixing bacteria in the wheat rhizosphere.
Next Article:Potential for non-symbiotic [N.sub.2]-fixation in different agroecological zones of southern Australia.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters