Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Johansson, P. M. Articles by Wright, S. A. I. Search for Related Content PubMed PubMed Citation Articles by Johansson, P. M. Articles by Wright, S. A. I. Agricola Articles by Johansson, P. M. Articles by Wright, S. A. I.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, November 2003, p. 6464-6474, Vol. 69, No. 11
0099-2240/03/$08.00+0     DOI: 10.1128/AEM.69.11.6464-6474.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

Low-Temperature Isolation of Disease-Suppressive Bacteria and Characterization of a Distinctive Group of Pseudomonads

P. Maria Johansson^1* and Sandra A. I. Wright^2*

Plant Pathology and Biocontrol Unit, Sveriges Lantbruksuniversitet, SE-750 07 Uppsala,¹ Department of Cellular and Molecular Biology, Göteborg University, SE-405 30 Göteborg, Sweden²

Received 21 July 2003/ Accepted 6 August 2003

Top Abstract Introduction Materials and Methods Results Discussion References

 
The influence of environmental factors during isolation on the composition of potential biocontrol isolates is largely unknown. Bacterial isolates that efficiently suppressed wheat seedling blight caused by Fusarium culmorum were found by isolating psychrotrophic, root-associated bacteria and by screening them in a bioassay that mimicked field conditions. The impact of individual isolation factors on the disease-suppressive index (DSI) of almost 600 isolates was analyzed. The bacteria originated from 135 samples from 62 sites in Sweden and Switzerland. The isolation factors that increased the probability of finding isolates with high DSIs were sampling from arable land, Swiss origin of samples, and origination of isolates from plants belonging to the family Brassicaceae. The colony morphology of the isolates was characterized and compared to DSIs, which led to identification of a uniform morphological group containing 57 highly disease-suppressive isolates. Isolates in this group were identified as Pseudomonas sp.; they were fluorescent on King's medium B and had characteristic crystalline structures in their colonies. These isolates were morphologically similar to seven strains that had previously been selected for suppression of barley net blotch caused by Drechslera teres. Members of this morphological group grow at 1.5°C and produce an antifungal polyketide (2,3-deepoxy-2,3-didehydrorhizoxin [DDR]). They have similar two-dimensional polyacrylamide gel electrophoresis protein profiles, phenotypic characteristics, and in vitro inhibition spectra of pathogens. In summary, in this paper we describe some isolation factors that are important for obtaining disease-suppressive bacteria in our system, and we describe a novel group of biocontrol pseudomonads.

Top Abstract Introduction Materials and Methods Results Discussion References

 
In biological control, microorganisms isolated from plants and soils are employed to protect crop plants from diseases. In cereals, seed- and soilborne fungal diseases are often suppressed by fluorescent pseudomonads (4, 6, 16, 17, 25). These bacteria are well adapted to the rhizosphere environment (for a recent review see reference 29), and they produce several antimicrobial substances that are involved in suppression of plant diseases (7). Many Fusarium species, including Fusarium culmorum, and the related organism Microdochium nivale are well-known and widespread fungal plant pathogens which can cause different diseases at various developmental stages of a crop (36, 43). This fact and the risk of triggering increased mycotoxin production by use of fungicides make these fungi difficult to control (30, 36). However, since biological control often operates through several modes of action, it provides new possibilities to address these two issues (15, 19, 21, 25).

Screening systems that approximate field conditions are more likely to result in selection of effective biocontrol isolates (12, 24, 50). In the temperate regions of the world, one critical environmental factor is the low temperature during winter and early spring (12). Hökeberg and et al. (13) developed an isolation and screening method that is selective for psychrotrophic biocontrol bacteria. By using this low-temperature assay, isolates that suppress many seed-borne diseases of field-grown barley were obtained (13, 17). The low-temperature approach in combination with a high level of pathogen inoculum was used by Johansson et al. (16) in a greenhouse screening for disease-suppressive bacteria. In field experiments that were performed for five consecutive years, four strains of different species (three fluorescent pseudomonads and one Pantoea sp.) efficiently suppressed both F. culmorum causing wheat seedling blight and M. nivale causing snow mold of wheat (16). Other examples of psychrotrophic bacteria are the plant growth-promoting rhizobacterium strain Pseudomonas putida GR12-2, which was selected for its ability to fix nitrogen and colonize the roots of canola (Brassica campestris) at low temperatures (28), and the biocontrol agent (BCA) Bacillus strain L324-92, which suppressed three soilborne diseases in growth chamber experiments carried out at 15°C and in direct-drilled field trials of winter and spring wheat in the inland northwest of the United States (22). The authors believed that the ability of strain L324-92 to grow at 4°C contributed to its biocontrol activity.

Several factors, such as soil type, soil texture, presence of a pathogen inoculum, sampling techniques, medium used for isolation, and plant material used to isolate organisms, all contribute to selection during the isolation process. There is a dearth of descriptions of selective factors that are active during isolation of potential BCAs. The selective effect of the medium used for isolation has been described in terms of the composition of the isolates obtained (21, 51), but there have been few reports of how different media influence the proportion of potential BCAs obtained (33). The plant species used for isolation of potential BCAs can also influence this proportion (2). It is well documented that the microflora in the rhizosphere is different from that in bulk soil, and some groups of bacteria can only be found in close proximity to plant roots (20, 42). The bacterial community of the rhizosphere is often found to be plant species specific (10, 11, 31, 32, 37, 44). Consequently, the plant material used for isolation can have a significant influence on the composition of the microflora obtained, as well as on the probability of finding isolates with biocontrol traits. Isolates that are efficient in protecting a particular plant species do not necessarily have to originate from that species.

In a previous study, disease-suppressive indices (DSIs) were generated for 598 psychrotrophic isolates that were screened with a greenhouse bioassay and described briefly (16). Based on data for 44 of these isolates, representing high, moderate, and low disease suppressiveness, the authors concluded that greenhouse DSIs correlated well with the DSIs obtained in field tests (16).

The aim of the present study was to identify and evaluate the influence of some isolation factors on the proportion of disease-suppressive bacterial isolates obtained. The origins, the methods of isolation, and the colony morphologies of almost 600 potential BCAs were analyzed to identify key isolation factors. Among the isolates analyzed, a specific, uniform group of highly disease-suppressive isolates emerged, and its distinguishing features are described below.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Isolation and handling of bacterial isolates: general procedure.
Root-associated bacteria were isolated on six different occasions, designated isolations 1 to 6 (Table 1). The following protocol was used for all isolations unless indicated otherwise. Plants with adhering soil were collected from various biotopes, such as arable land, forest, and gravel ground, at latitudes ranging from 47 to 68°N and at longitudes ranging from 6 to 19°E (Table 1). The roots of each plant were separated from the rest of the plant and washed under tap water. Undamaged root pieces that were 1 to 5 mm in diameter and 1 to 2 cm long were used for isolation of bacteria. The root pieces were subsequently rinsed in sterile water and left to dry on autoclaved filter paper for 1 h. The roots pieces were cleaved longitudinally with a scalpel, and the surfaces were rubbed over the agar in petri dishes containing (i) King's medium B (KB) (23), (ii) wheat straw agar, (iii) soil extract agar, (iv) water agar (20 g of agar technical no. 3 [Oxoid Ltd., Basingstoke, England] in 1,000 ml of deionized water), and/or (v) 33% tryptic soy agar (10 g of tryptic soy broth [Difco, Becton Dickinson, Le Pont de Claix, France] and 20 g of agar technical in 1,000 ml of deionized water). Wheat straw agar was prepared by mixing 100 ml of dry wheat straw meal with 500 ml of nonsterile tap water. After overnight swelling, the mixture was filter sterilized, and then 50 ml of the filtrate was added to an autoclaved agar solution (20 g of agar in 950 ml of deionized water) at 60°C. Soil extract agar was prepared by mixing 100 ml of peat soil with 1,000 ml of nonsterile tap water; the mixture was boiled for 10 min and filtered. Then 100 ml of this extract, 5 g of malt extract (Difco), 20 g of agar technical, and 900 ml of deionized water were mixed and autoclaved. The inoculated agar plates were incubated in the dark at 1.5°C. Once or twice a week, the plates were examined, and one representative colony of each of the different types that appeared (Table 2) was picked and transferred onto 33% tryptic soy agar. The isolates were incubated for 3 days in the dark at 15°C, and the purity and colony morphology were checked. Subsequent growth and maintenance of the cultures have been described previously (16).

View this table: [in this window] [in a new window]

TABLE 1. Origins of isolates and conditions during isolation

View this table: [in this window] [in a new window]

TABLE 2. Matrix used to create the codes describing bacterial colony morphologya

Morphological characterization.
Each isolate was observed with a stereomicroscope at a magnification of x40 to x50. The morphological traits were divided into four categories (shape, color, texture, and other distinctive traits), and there were four alternative descriptions of the traits in each category, trivially numbered 1 to 4 (Table 2). By using this key, the colonial morphology was described with a four-digit code. When a trait was missing or not covered by the key, we placed a zero in the appropriate position. By assigning codes, we could readily group and identify isolates with similar morphological traits.

Specific conditions for the six isolations.
Table 1 shows complementary information, such as the sampling location(s) and details concerning the plant used, as well as the number of individual samples per isolation. Samples were kept separate from one another at all times. For isolation 1, samples from various biotopes were taken in April 1996. Only isolates showing fluorescence on KB under UV light (254 nm) were picked. For isolation 2, samples from a table land in the Swedish mountain area were taken in July 1994. The roots were separated from the rest of each plant, washed, and stored at minus 28°C for 3 years. After thawing and rehydration in sterile tap water, the roots were treated as described above. For isolation 3, samples from arable land were taken in November 1996. Root pieces that had been treated as described above were placed in a test tube containing 1 ml of a 0.01 M MgSO₄ solution and ground with a glass rod. The resulting suspension was diluted with MgSO₄ (100 µl in 5 ml of MgSO₄). Then 50 µl of this solution was spread on each of four different media (Table 1). For isolation 4, samples from arable land were taken in March 1997. Plants with 200 to 500 g of adhering soil were packed in bags made of composite material (paper, aluminum, and plastic; one bag per sample) and sent by mail to Sweden. Upon arrival, the material was heavily fermented and had a characteristic smell, reminiscent of hydrogen sulfide. For isolation 5, samples from arable land were taken in December 1997. For isolation 6, samples from arable land were taken in March 1999 and the conditions were similar to those of isolation 4.

Reference strains.
Seven strains of bacteria (Ab 17 [= NB 1], Ab 131, Ki 59 [= NB 3], Ki 84 [= NB 4], Ki 325 [= NB 5], Ki 341 [= NB 6], and MA 342) that had been isolated from roots of wild plants sampled in 1988, 1993, and 1994 from table land in the northern part of Sweden (Table 1) were used as reference strains. These strains had been isolated on three separate occasions by using a method similar to the one used in this work and were selected because of their ability to suppress barley net blotch caused by Drechslera teres and stinking smut caused by Tilletia caries in greenhouse experiments (13). The 16S rRNA sequences of three of these strains (Ab 17 [= NB 1], Ab 131, and Ki 341 [= NB 6]) were retrieved from the National Center for Biotechnology Information database (http://www.ncbi.nlm.nih.gov) (accession number AJ012712), in which the strains are referred to as Pseudomonas borealis (3).

Screening for suppression of wheat seedling blight caused by F. culmorum.
The greenhouse bioassay used has been described in detail previously (16). Wheat seeds artificially infested with F. culmorum were treated with 48-h bacterial cell cultures (in 50% tryptic soy broth [Difco]) (log 9 to log 10 cells ml^-1; 300 ml kg^-1). The treated seeds were dried and sown in pots by using six seeds per pot and a minimum of two pots per isolate. The pots were incubated in the dark at 6°C for 10 days before they were transferred to a greenhouse with a temperature of 15 ± 3°C and a minimum of 12 h of light per day. After 14 days, the degree of germination and symptom development were recorded. Disease suppression was graded by using the following scoring method: 2, healthy plants; 1, plants with symptoms; and 0, dead or nonemerged plants. A DSI was determined for each pot (1 pot = 1n in the analysis). Thus, a pot in which all six plants were healthy had a DSI of 12, and a pot in which all of the plants were dead or nonemerged had a DSI of 0. The DSIs correlated both with the fresh weight of the seedlings (r² = 0.89), which provided a quantitative measure of disease suppression, and with the levels of disease suppression observed under field conditions (r² = 0.72), as described previously (16).

Influence of isolation factors on DSI.
The isolates screened with the bioassay were grouped according to the material and methods used for isolation and the specific traits of the bacterial colonies. DSIs for individual pots were used as random samples for the defined groups, which in a previous analysis had exhibited a normal distribution (Fig. 1). Within the categories (categories i, ii, iii, and iv [see below]) the subgroups were subjected to one-way analysis of variance by using MiniTAB (release 13.2). The probability of finding an isolate that had a mean DSI of 6 was calculated for each group. The following categories and groups were defined for comparisons of mean DSIs: (i) geographic influence (all Swiss isolates versus Swedish isolates obtained on KB) (two groups); (ii) morphological traits of colonies (Table 2) (shape and edges, color, texture, and other traits) (11 groups were defined and analyzed); (iii) plant species influence (isolates from wild grasses [including members of the families Juncaceae and Cyperaceae], isolates from cultivated grasses [cereals], isolates from members of the family Brassicaceae [both wild and cultivated], and isolates from other plant species) (four groups); (iv) influence of media (Swedish isolates [Table 1] were analyzed) (five groups); and (v) biotope influence (isolates from arable land and isolates from other [natural] biotopes) (two groups).

View larger version (17K): [in this window] [in a new window]

FIG. 1. Distribution of 595 bacterial isolates based on the level of suppression of wheat seedling blight caused by F. culmorum. The suppression was tested by a greenhouse bioassay, and each isolate was graded by using a DSI, where 0 reflected no disease suppression (all seedlings dead) and 12 reflected total disease suppression (all seedlings healthy). UT, untreated (healthy) control (mean DSI, 10.5; n = 69); FC, F. culmorum control (mean DSI, 1.6; n = 84).

Production of DDR.
The following protocol was adapted from a method used by M. Svensson and L. Lundgren (unpublished data) and from the method used by Hökeberg (14), which was modified by AgriVir AB (Uppsala, Sweden). Briefly, bacterial cultures grown as described above for seed treatment were centrifuged for 20 min at 20,000 x g (rotor RC5B; DuPont) at 4°C. The supernatants were separated by solid-phase extraction. The material was separated by elution of the columns, first with 2% ethanol and then with 95% ethanol. The fractions were then dried by evaporation of the solvents and stored at -20°C for subsequent analysis. The hydrophobic fractions were injected into a preparative high-performance liquid chromatography column (Kromasil 100-5C18, HICHROM) by using 70% CH₃CN (5 ml m^-1) as the mobile phase. For some of the strains, crude supernatants were applied directly (without solid-phase extraction) to the high-performance liquid chromatography column. The presence of 2,3-deepoxy-2,3-didehydro-rhizoxin (DDR) was examined at a wavelength of 310.4 nm. Strains that produced DDR showed a peak at approximately 6.5 min (14).

In vitro inhibition of pathogenic fungi and bacteria.
The antimicrobial activities of a selected group of isolates were tested by using a protocol developed by AgriVir AB (46). Antifungal activity was tested by using six different filamentous fungi representing diverse taxonomic groups and one yeast, Candida albicans. Antibacterial activity was tested by using one plant pathogen, Pseudomonas savastanoi strain CCUG 2171, and one human pathogen, Staphylococcus aureus. The six fungi were Aspergillus fumigatus, Drechslera sorokiniana, F. culmorum No. 11, Heterobasidion annosum, Microsporum canis, and Pythium ultimum. Bacterial culture supernatants were fractionated, dried, and stored as described above. The dried material originating from 100 ml of culture supernatant was dissolved in 5 ml of 50% acetonitrile, and 50-, 25-, 10-, 5-, and 2-µl aliquots were transferred to 96-well microtiter plates, which was followed by evaporation of the solvent overnight. One hundred microliters of a test organism spore or cell suspension (approximately 10⁴ spores ml^-1 or 10⁴cells ml^-1) in the appropriate medium (46) was added to each well of the microtiter plates, which resulted in test concentrations that were 10, 5, 2, 1, and 0.5 times the concentration of the original supernatant. For most of the microorganisms tested, spore germination and cell growth were monitored after between 24 and 144 h of incubation at 28°C; the exceptions were S. aureus and C. albicans, which were incubated at 37°C. The antimicrobial activity was determined based on the absorbance at 620 nm as determined with a plate reader (Multiscan Ascent, version 2.4.2).

Two-dimensional electrophoresis of bacterial proteins.
Bacterial pellets consisting of 10⁸ to 10⁹ late-log-phase cells were agitated for 50 min at room temperature in 360 µl of a solution consisting of 8 M urea, 2% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), 0.5% immobilized pH gradient (IPG) buffer (for the nonlinear pH range from pH 3 to 10), 20 mM dithiothreitol (DTT), and 0.001% bromophenol blue. After 10 min of centrifugation, the supernatants were applied to 18-cm Immobiline dry strips with an immobilized nonlinear pH gradient (pH 3 to 10; Amersham Biosciences, Uppsala, Sweden). The first dimension was electrophoresed by using an IPGphor isoelectric focusing unit (Amersham Biosciences), starting with a rehydration step at 50 V for 12 h at 20°C. The following steps were 500 V for 1 h, 1,000 V for 1 h, and 8,000 V until a total of 38,000 V · h was reached after about 6 h.

The strips were then equilibrated for 15 min in 10 ml of a solution containing 50 mM Tris-HCl (pH 8.8), 6 M urea, 30% glycerol, 2% sodium dodecyl sulfate (SDS), 0.001% bromophenol blue, and 65 mM DTT, and this was followed by a 15-min alkylation step in 10 ml of the same solution except that the DTT was replaced by 135 mM iodoacetamide. The second-dimension electrophoresis was performed in SDS-12.5% polyacrylamide gels (1 mm by 18 cm by 18 cm) at constant current of 45 mA until the dye front reached the bottom of each gel. The markers used had molecular masses of 97, 66, 45, 31, 21, and 14 kDa. The gels were stained with silver as described by Shevchenko et al. (40).

Phenotypic characterization of bacterial isolates.
For preliminary identification, an API 20NE kit (Bio Mérieux, Lyon, France) was used. The major goal, however, was to obtain information for the 21 biochemical characteristics that are tested with this kit. Additional carbon source utilization data were obtained with the PhenePlate-48 microtiter plate system (PhP-48 plates; BioSys Inova, Stockholm, Sweden). This kit contains a set of 48 dehydrated reagents (mostly amino acids and sugars) and was used according to the recommendations of the manufacturer. Three strains (MF 30, MF 588, and MF 626) were previously identified (16) at the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (Braunschweig, Germany) by using 40 biochemical tests, fatty acid methyl ester analysis, and analysis of partial sequences of 16S rRNA; for this study strain MF 381 was submitted to the Deutsche Sammlung von Mikroorganismen und Zellkulturen for identification.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Disease suppression as determined by greenhouse screening.
Overall, 33 isolates (5.5%) totally suppressed disease (mean DSI, 10.50 [the DSI for the untreated control]), and 96 isolates (16.0%) enhanced disease development (mean DSI, 1.60 [the DSI for the F. culmorum control]) (Fig. 1). The deleterious bacteria originated mainly from two of the six isolations; 46 isolates from isolation 2, 48 isolates from isolation 5, and 2 isolates from isolation 6. The 33 isolates with high disease levels of suppressiveness were also unevenly distributed over the six isolations (Table 3).

View this table: [in this window] [in a new window]

TABLE 3. Characteristics of the 33 isolates with full disease suppressive activity against F. culmorum causing seedling blight of wheat in greenhouse tests

The diversity of the bacterial isolates obtained from the isolation 4 material was low compared to the diversity obtained on previous isolation occasions. From 50 plant samples of 10 plant species, collected at 22 locations in Switzerland (Table 1), only 76 isolates were obtained. From each of the samples, which were kept separate throughout the isolation process, a certain type of bacteria with a characteristic colony morphology was isolated (Fig. 2). When the colony morphology was analyzed, these isolates formed one of the groups (code 1130). In Table 2, the code number 3 in the third position indicates the presence of crystal or optically dense spots; hence, isolates that obtained that code are referred to below as IODS. In isolation 4, 57 IODS were found, and these isolates efficiently suppressed seedling blight in greenhouse (Table 4) and field (Table 3) experiments (16).

View larger version (161K): [in this window] [in a new window]

FIG. 2. Single colony of a Pseudomonas sp. displaying the morphology of the IODS. Bar = 1 mm. The inset shows the crystal-like structure of a spot at a magnification of x400. A single crystal is approximately 20 µm in diameter.

View this table: [in this window] [in a new window]

TABLE 4. Analysis of isolation factors influencing the proportion of bacterial isolates that suppressed wheat seedling blight

Influence of isolation factors on DSI. (i) Geographic origin.
When the DSIs of bacteria isolated on KB (isolations 1, 4, 5, and 6) were compared, the 142 Swiss isolates (originating from 78 samples of 11 species) were more suppressive than the 216 Swedish isolates (originating from 24 samples of 13 species)

(ii) Colony morphology.
The IODS morphology was strongly correlated with the ability to suppress disease (mean DSI, 9.8; n = 57). Isolates with the code 1120 (Table 2) were also significantly more suppressive than the rest of the isolates. Their mean DSI (5.3; n = 151), however, was not as high as that of the IODS.

(iii) Plant species origin.
The isolates from cruciferous plants (family Brassicaceae) had the highest mean DSI, followed by the isolates from cultivated grasses. When the data were analyzed without the IODS, the plant species effect that favored the isolates originating from crucifers was even more pronounced. The members of this group were significantly separated from all the other isolates (P > 0.001) (Table 4). Of the 33 isolates that exhibited 100% disease suppression (DSI, 10.5), 10 were isolated from plants belonging to the family Brassicaceae and 20 were isolated from grasses (13 from wild grasses and 7 from cultivated grasses). However, the proportion of grasses in the whole material that was used for isolation was much higher (61%) than the proportion of plants belonging to the Brassicaceae (18%); hence, the 10 isolates from members of the Brassicaceae constitutes a relatively high proportion of the 33 isolates (Table 3).

(iv) Media.
Only Swedish isolates were included in the analysis of media. The bacterial colony types isolated did not differ in any obvious way for the different media. Isolates grown on soil extract agar were significantly more suppressive (P < 0.001) than isolates grown on other media, as shown in Table 4.

(v) Biotope.
The isolates that originated from arable land had a higher mean DSI than the isolates that originated from other environments (Table 4).

Production of DDR.
Sixteen IODS and 16 other fluorescent pseudomonads (MV isolates) were tested for DDR production. All IODS produced DDR or (as in the case of Ki 341) traces of DDR (3), while the other strains did not produce this compound (Table 5).

View this table: [in this window] [in a new window]

TABLE 5. Identities and properties of bacterial isolates with similar colony morphology (IODS)

In vitro inhibition of pathogenic fungi and bacteria.
Nine IODS and one non-IODS reference strain (Pantoea agglomerans MF626) from this collection, as well as seven IODS reference strains (3, 14), were tested for antimicrobial activity. There was significantly more in vitro antifungal and antibacterial activity in the hydrophobic fraction (Table 5) than in the hydrophilic fraction (data not shown).

Two-dimensional electrophoresis of bacterial proteins.
The two-dimensional SDS-polyacrylamide gels for 12 IODS (9 strains from the present work and 3 reference strains) had quite a uniform pattern of proteins (Fig. 3). There were eight areas (designated areas A to H) where differences among the strains tested were observed. Letters were assigned to the presence of dots in certain areas (one letter per dot). In this manner, seven unique patterns were identified for seven of the IODS, while the eighth pattern was shared by five of the strains (Table 5).

View larger version (27K): [in this window] [in a new window]

FIG. 3. Two-dimensional SDS-polyacrylamide gel electrophoresis analysis of 12 IODS (see Fig. 1). The left panel is a simplification of the representative two-dimensional gel on the right (of strain MF 381). The black dots represent proteins or clusters of proteins that were found in all 12 strains. The shaded areas marked A, B, C, D, E, F, G, and H are regions where there are differences among the strains. Finally, the shaded squares represent areas where additional dissimilarities were detected, but in these cases there were minor shifts in the positions of the proteins. The scale on the left indicates the approximate molecular masses of the proteins, based on the sizes of reference proteins with known molecular masses.

Phenotypic and genotypic characteristics of bacterial strains.
A summary of the results obtained by using the API 20NE and PhP-48 test systems and previously published data for other closely related pseudomonads is presented in Table 6. Of the isolates that were tested with the API 20NE system, 80% belonged to the genus Pseudomonas. The API 20NE system was not very discriminatory for the highly disease-suppressive isolates. Of the 33 isolates that exhibited full suppression (Table 3), 23, including the IODS, had the API 20NE code 0157555, which according to the kit gives very good identification for the genus Pseudomonas but only 79.3% certainty of identification for the most similar species, Pseudomonas fluorescens. A few strains were subjected to further identification (16). One of the IODS (strain MF 381) was determined to be most closely related to Pseudomonas frederiksbergensis (1) based on a partial sequence of the 16S rRNA. BLAST searches performed with the whole16S rRNA sequence of the reference IODS Ab 131 showed that the IODS are most closely related to Pseudomonas mandelii (49) and P. frederiksbergensis (99% identity in both cases). There were eight sequences from isolates of Pseudomonas sp. that had similarity scores in the BLASTN analysis which were higher than that of P. mandelii; six of the strains were isolated from woodland soils in the southwestern United States (26), and two uncultured strains were isolated from soils from California grassland (M. G. La Montagne, J. P. Schimel, and P. A. Holden, unpublished data). The best alignment was obtained with strain Pseudomonas sp. strain PsB (accession number AF105380) (26).

View this table: [in this window] [in a new window]

TABLE 6. Comparison of general and nutritional characteristics of IODS and reference strains to those of related Pseudomonas spp.a

Top Abstract Introduction Materials and Methods Results Discussion References

 
When bacteria from natural environments are sampled, there are a number of factors that influence the subset of bacteria that are actually isolated. The type and structure of the soil (15, 32), the season (42), the plant species (2, 11, 31, 32, 37, 44) (Table 4), the medium or substrate (21, 33, 51) (Table 4), and the temperature (14, 45) all have an impact on the composition of the microflora in the rhizosphere. We try to minimize the influence of factors outside the scope of a study, but there are always some factors that are beyond our control and whose presence becomes apparent with unbalanced sampling and small sample sizes. If the sample size is sufficiently large, a randomly sampled, natural population of root-associated bacteria will approximate a normal distribution with regard to the ability to suppress disease. Hökeberg et al. (13) found that 400 isolates that suppressed D. teres in barley and 309 isolates that suppressed M. nivale in wheat had such a distribution (13). The DSIs of the 595 isolates in the present study also had this pattern (Fig. 1), suggesting that the population tested was representative and sufficiently large to perform the analysis. In a study in which the aim is to compare multiple factors that are not always entirely independent, a high number of random samples (e.g., >100 samples) ensures that the factors analyzed also are the factors responsible for an increase in the number of suppressive isolates.

In Table 4, the increased probability of finding a disease-suppressive isolate is reflected both by the high proportion of isolates with a DSI of 6 or more and by the high mean DSI. Disease-suppressive bacteria were isolated more frequently from plants belonging to the family Brassicaceae than from the other plants. This was also true when the highly suppressive IODS were removed from the analysis (Table 4). Berg et al. (2) found plant species-dependent variation (both qualitative and quantitative variation) among isolates antagonistic in vitro to four soilborne pathogens. They also found a lot of antagonistic isolates originating from oilseed rape (Brassica napus var. oleifera), demonstrating that this family of plants may possess certain features that encourage the presence of a suppressive microflora. The diversity of the microflora of cruciferous plants is high (2, 11, 18), which in turn might increase the probability of finding disease-suppressive isolates. Moreover, of the 33 isolates that totally suppressed development of wheat seedling blight, 10 originated from plants belonging to the Brassicaceae (Table 3). Since the proportion of cruciferous plants in the entire material was only 18%, this finding further strengthens the conclusion that plants belonging to the Brassicaceae are a favored niche of potential BCAs.

An alternative approach to ensure a high proportion of isolates with desired traits is to use methods that target characteristics known to be related to biocontrol activity. Examples of this include media that detect the production of hydrolytic enzymes (3, 21, 34, 51) or antifungal metabolites (27, 33) and molecular markers that can detect certain traits (2, 27). The distinctive colony morphology of the IODS (Fig. 2) makes it possible to select potentially disease-suppressive isolates by observing the colonies.

Of the 595 isolates analyzed in this work, almost 40% resulted in suppression in which 50% or more of the disease symptoms were absent (i.e., the DSI was 6 or more) (Fig. 1 and Table 4). These isolates are facultative psychrotrophs, and in addition, the low temperature used during screening may have induced the expression of specific genes responsible for disease suppression, such as the production of antimicrobial compounds. The production of 2,4-diacetylphloroglucinol was greatest at 12°C (39). The psychrotrophic P. fluorescens strain ANP15 produces more pyoverdine at 12°C than at 19°C, the optimal growth temperature of this strain (38). Likewise, maximum production of phenazine occurred at a temperature below the temperature that is optimal for growth (41). We believe that the low-temperature approach that was used during isolation and screening increased the probability of recovering and identifying disease-suppressive isolates.

The IODS have emerged as a distinct group of pseudomonads with excellent disease suppressiveness. Seven IODS have also exhibited high levels of suppressiveness under field conditions (Table 3) (13, 16, 17; Johansson, unpublished results). The strain that has been tested most extensively is MA 342, which has been developed into a commercial product, Cedomon (BioAgri AB, Uppsala, Sweden) and is used as a seed dressing to combat seed-borne pathogens of barley (14). The modes of action of strain MA 342 include the ability to colonize the glumes and pericarp of barley seed (47) and DDR production (14). Of the 64 IODS isolated so far, 12 were chosen for further characterization. The strains chosen represent 12 distinct samples that were collected at nine geographically separate locations in Sweden and Switzerland, and they were isolated from five different species of plants at different times (Tables 1 and Table 3). Although most of these isolates reacted similarly in biochemical tests (Table 6), they could be differentiated into eight groups based on two-dimensional protein profiles (Fig. 3 and Table 5) and into two additional groups based on inhibition of M. canis (Table 5). Of the five strains (MF 376, MF 382, MF 402, MF 411, and MF 416) that were indistinguishable on the basis of their protein profiles, strains MF 411 and MF 416 were unique in that they did not inhibit M. canis in vitro (Table 5).

The identity of the IODS was difficult to establish firmly. The API 20NE kit identified all of these organisms as Pseudomonas sp. strains that were most closely related to P. fluorescens, but the level of identification was low (79.3%). The fatty acid methyl ester profile of one IODS reference strain (MA 342) suggested the identity Pseudomonas chlororaphis (13). The 16S rRNA sequence data for three of the reference strains (Ab 17, Ab 131, and Ki 341) and for MF 381 were most similar to the data for P. mandelii and P. frederiksbergensis (see above). The phenotypic characteristics of the IODS, however, are not identical to those of any of the species that are listed in Table 6. P. frederiksbergensis and P. mandelii are distinguished from the IODS by not producing a fluorescent pigment on KB (P. frederiksbergensis), by their denitrifying ability, by their inability to liquefy gelatin (P. mandelii), and by their growth on lactose (P. mandelii), maltose (P. mandelii), and sucrose (P. frederiksbergensis) but not on D-xylose (P. frederiksbergensis), gentobiose (P. frederiksbergensis), L-arabinose (P. mandelii), and melibiose (P. frederiksbergensis) (Table 6). The color of the colonies of P. frederiksbergensis is distinct from that of the IODS, since they are pale yellow (1), and the green or orange pigments produced by P. chlororaphis on King's medim A and KB are not produced by the IODS. All IODS that have been tested so far (except Ki 341 [3]) produce the antifungal metabolite DDR (Table 5) and have white-grey colonies containing optically dense spots. Attempts to analyze the contents of the spots (Fig. 2) have been made, but it is not clear yet what they are. A close-up of their appearance (at a magnification of x400) is shown in Fig. 2. The spots consist of yellow, transparent crystals which dissolve in ethanol but not in water. In the case of the IODS, neither 16S rRNA homology nor phenotypic characteristics provide the necessary information for unambiguous phylogenetic placement. The closest relative of these strains, based on all these characteristics, is P. mandelii. The data obtained thus far suggest that they form a novel taxonomic entity.

In summary, this work demonstrated that plants belonging to the family Brassicaceae seem to provide a favored niche for bacteria with biocontrol potential and that the colonial morphology of bacteria can be a valuable tool when potential BCAs are selected. Our results also suggest that there is a high yield of disease-suppressive bacteria when a screening procedure that mimics field conditions is used. Finally, in this study we characterized the IODS, a new group of pseudomonads, which are excellent candidates for development of commercial BCAs.

 
This work was funded by the MISTRA Foundation, by the JLT faculty at SLU, and by the Swedish Farmers' Foundation for Agricultural Research.

We are grateful to AgriVir AB and especially to Jolanta Levenfors, who opened their labs and shared their techniques and also contributed work and results. We thank Janna Krovacek, Zahra Saad Omer, Juliana Larsson, Valentin Carballo, Fredrik Heyman, and Celina Fällbom, who provided technical assistance and fruitful companionship. We are grateful to Lennart Lundgren and Anton Pohanka for help with identification of DDR and to Håkan Larsson for help with running two-dimensional protein gels. We thank Mattias Refardt at Siegfrid Agro AG, Zofingen, Switzerland, who provided plant samples from all over Switzerland, and Ulf Granhall for critical reading of the manuscript. Last but not least, we thank David Wright for language editing.

 
* Corresponding author. Mailing address for P. Maria Johansson: Plant Pathology and Biocontrol Unit, SLU, Box 7035, 75007 Uppsala, Sweden. Phone: 46-18-303158. Fax: 46-18-671690. E-mail: Maria.Johansson{at}vpat.slu.se. Mailing address for Sandra A. I. Wright: Department of Cellular and Molecular Biology, Göteborg University, Box 462, 405 30 Göteborg, Sweden. Phone: 46-18-671605. Fax: 46-18-671690. E-mail: Sandra.Wright{at}molbio.gu.se.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Andersen, S. M., K. Johnsen, J. Sørensen, P. Nielsen, and C. S. Jacobsen. 2000. Pseudomonas frederiksbergensis sp. nov., isolated from soil a coal gasification site. Int. J. Syst. Evol. Biol. 50:1957-1965.
2
2. Berg, G., N. Roskot, A. Steidle, L. Eberl, A. Zockand, and K. Smalla. 2002. Plant-dependent genotypic and phenotypic diversity of antagonistic rhizobacteria isolated from different Verticillium host plants. Appl. Environ. Microbiol. 68:3328-3338.[Abstract/Free Full Text]
3
3. Borowicz, J. J. 1998. Traits of biologically interacting pseudomonads. Ph.D. thesis. Swedish University of Agricultural Sciences, Uppsala, Sweden.
4
4. Bull, C. T., D. M. Weller, and L. S. Thomashow. 1991. Relationship between root colonization and suppression of Gaeumannomyces graminis var. tritici by Pseudomonas fluorescens strain 2-79. Phytopathology 81:954-959.[CrossRef]
5
5. Coroler, L., M. Elomari, B. Hoste, M. Gillis, D. Izard, and H. Leclerc. 1996. Pseudomonas rhodesiae sp. nov., a new species isolated from natural mineral waters. Syst. Appl. Microbiol. 19:600-607.
6
6. Defago, G., and D. Haas. 1990. Pseudomonads as antagonists against soilborne plant pathogens: mode of action and genetic analysis, p. 249-291. In J. M. Bollag and G. Stotzky (ed.), Soil biochemistry, vol. 6. Dekker, New York, N.Y.
7
7. Dowling, D. N., and F. O'Gara. 1994. Metabolites of Pseudomonas involved in the biocontrol of plant disease. Trends Biotechnol. 12:133-141.
8
8. Elomari, M., L. Coroler, D. Izard, and H. Leclerc. 1995. A numerical taxonomic study of fluorescent Pseudomonas strains isolated from natural mineral waters. J. Appl. Bacteriol. 78:71-81.[Medline]
9
9. Elomari, M., L. Coroler, B. Hoste, M. Gillis, D. Izard, and H. Leclerc. 1996. DNA relatedness among Pseudomonas strains isolated from natural mineral waters and proposal of Pseudomonas veronii sp. nov. Int. J. Syst. Bacteriol. 46:1138-1144.
10
10. Georgakopoulos, D. G., M. Hendson, N. J. Panopoulos, and M. N. Schroth. 1994. Analysis of expression of a phenazine biosynthesis locus of Pseudomonas aureofaciens PGS 12 on seeds with a mutant carrying a phenazine biosynthesis locus-ice nucleation reporter gene fusion. Appl. Environ. Microbiol. 60:4573-4579.[Abstract/Free Full Text]
11
11. Germida, J. J., S. D. Siciliano, J. Renato de Frietas, and A. M. Seib. 1998. Diversity of root-associated bacteria associated with field-grown canola (Brassica napus L.) and wheat (Triticum aestivum L.). FEMS Microbiol. Ecol. 26:43-50.
12
12. Glick, B. R., C. L. Patten, G. Holguin, and D. M. Penrose. 1999. Biochemical and genetic mechanisms used by plant growth promoting bacteria, p. 180-248. Imperial College Press, London, United Kingdom.
13
13. Hökeberg, M., B. Gerhardson, and L. Johnsson. 1997. Biological control of cereal seed-borne diseases by seed bacterisation with greenhouse-selected bacteria. Eur. J. Plant Pathol. 103:25-33.
14
14. Hökeberg, M. 1998. Seed bacterisation for control of fungal seed-borne diseases in cereals. Ph.D. thesis. Swedish University of Agricultural Sciences, Uppsala, Sweden.
15
15. Huang-Y-Wong, P. T. W. 1998. Effect of Burkholderia (Pseudomonas) cepacia and soil type on the control of crown rot in wheat. Plant Soil 203:103-108.[CrossRef]
16
16. Johansson, P. M., L. Johnson, and B. Gerhardson. 2003. Suppression of wheat-seedling diseases caused by Fusarium culmorum and Microdochium nivale using bacterial seed treatment. Plant Pathol. 52:219-227.[CrossRef]
17
17. Johnsson, L., M. Hökeberg, and B. Gerhardson. 1998. Performance of the Pseudomonas chlororaphis biocontrol agent MA 342 against cereal seed-borne diseases in field experiments. Eur. J. Plant Pathol. 104:701-711.[CrossRef]
18
18. Kaiser, O., A. Pühler, and W. Selbitschka. 2001. Phylogenetic analysis of microbial diversity in the rhizoplane of oilseed rape (Brassica napus cv. Westar) employing cultivation-dependent and cultivation-independent approaches. Microb. Ecol. 42:136-149.[Medline]
19
19. Kempf, H. F., and G. Wolf. 1989. Erwinia herbicola as a biocontrol agent of Fusarium culmorum and Puccinia recondita f. sp. tritici on wheat. Phytopathology 79:990-994.
20
20. Kent, A. D., and E. W. Triplett. 2002. Microbial communities and their interactions in soil and rhizosphere ecosystems. Annu. Rev. Microbiol. 56:211-236.[CrossRef][Medline]
21
21. Khan, N. I., D. A. Schisler, M. J. Boehm, P. J. Slininger, and R. J. Bothast. 2001. Selection and evaluation of microorganisms for biocontrol of Fusarium head blight of wheat incited by Gibberella zeae. Plant Dis. 12:1253-1258.
22
22. Kim, D.-S., R. J. Cook, and D. M. Weller. 1997. Bacillus sp. L324-92 for biological control of three root diseases of wheat grown with reduced tillage. Phytopathology 87:551-558.[Medline]
23
23. King, E. O., M. K. Ward, and D. E. Raney. 1954. Two single media for the demonstration of pyocyanin and fluorescein. J. Lab. Clin. Med. 44:301-307.[Medline]
24
24. Knudsen, I. M. B., J. Hockenhull, D. Funck Jensen, B. Gerhardson, M. Hökeberg, R. Thavonen, E. Teperi, L. Sundheim, and B. Henriksen. 1997. Selection of biological control agents for controlling soil and seed-borne diseases in the field. Eur. J. Plant Pathol. 103:775-784.[CrossRef]
25
25. Kropp, B. R., E. Thomas, J. I. Pounder, and A. J. Anderson. 1996. Increased emergence of spring wheat after inoculation with Pseudomonas chlororaphis isolate 2E3 under field and laboratory conditions. Biol. Fertil. Soils 23:200-206.[CrossRef]
26
26. Kuske, C. R., J. D. Busch, D. L. Adorada, J. M. Dunbar, and S. M. Barns. 1999. Phylogeny, ribosomal RNA gene typing and relative abundance of new Pseudomonas species (sensu stricto) isolated from two pinyon-juniper woodland soils of the arid southwest U. S. Syst. Appl. Microbiol. 22:300-311.[Medline]
27
27. Landa, B. B., O. V. Mavrodi, J. M. Raaijmakers, B. B. McSpadden Gardener, L. S. Thomashow, and D. M. Weller. 2002. Differential ability of genotypes of 2,4-diacetylphloroglucinol-producing Pseudomonas fluorescens strains to colonize the roots of pea plants. Appl. Environ. Microbiol. 68:3226-3237.[Abstract/Free Full Text]
28
28. Lifshitz, R., J. W. Kloepper, F. Scher, E. M. Tipping, and M. Laliberte. 1986. Nirogen-fixing Pseudomonas isolated from roots of plants grown in the Canadian high arctic. Appl. Environ. Microbiol. 51:251-255.[Abstract/Free Full Text]
29
29. Lugtenberg, B. J. J., L. C. Dekkers, and G. V. Bloemberg. 2001. Molecular determinants of rhizosphere colonization by pseudomonads. Annu. Rev. Phytopathol. 39:461-490.[CrossRef][Medline]
30
30. Magan, N., R. Hope, A. Colleate, and E. S. Baxter. 2002. Relationship between growth and mycotoxin production by Fusarium species, biocides and environment. Eur. J. Plant Pathol. 108:685-690.[CrossRef]
31
31. Marschner, P., C.-H. Yang, R. Lieberei, and D. E. Crowley. 2001. Soil and plant specific effects on bacterial community composition in the rhizosphere. Soil Biol. Biochem. 33:1437-1445.[CrossRef]
32
32. Miethling, R., G. Wieland, H. Backhaus, and C. C. Tebbe. 2000. Variation of microbial communities in response to crop species, soil origin, and inoculation with Sinorhizobium meliloti L 33. Microb. Ecol. 41:43-56.
33
33. Neiendam Nielsen, M., J. Sørensen, J. Fels, and H. C. Pedersen. 1998. Secondary metabolite- and endochitinase-dependent antagonism toward plant-pathogenic microfungi of Pseudomonas fluorescens isolates from sugar beet rhizosphere. Appl. Environ. Microbiol. 64:3563-3569.[Abstract/Free Full Text]
34
34. O'Sullivan, M., P. M. Stephens, and F. O'Gara. 1991. Extracellular protease production by fluorescent Pseudomonas spp. and the colonization of sugar beet roots and soil. Soil Biol. Biochem. 23:623-627.[CrossRef]
35
35. Palleroni, N. J. 1984. Family I. Pseudomonadaceae Migula 1894, 237^AL, p. 141-199. In N. R. Krieg and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 1 Williams & Wilkins, Baltimore, Md.
36
36. Parry, D. W., P. Jenkinson, and L. Mc Leod. 1995. Fusarium ear blight (scab) in small grain cereals—a review. Plant Pathol. 44:207-238.[CrossRef]
37
37. Rainey, P. B. 1999. Adaptation of Pseudomonas fluorescens to the plant rhizosphere. Environ. Microbiol. 1:243-247.[CrossRef][Medline]
38
38. Seong, K. Y., M. Hofte, J. Boelens, and W. Verstraete. 1991. Growth, survival, and root colonisation of plant growth beneficial Pseudomonas fluorescens ANP15 and Pseudomonas aeruginosa 7NSK2 at different temperatures. Soil Biol. Biochem. 23:485-486.[CrossRef]
39
39. Shanahan, P., D. J. O'Sullivan, P. Simpson, J. D. Glennon, and F. O'Gara. 1992. Isolation of 2,4-diacetylphloroglucinol from a fluorescent pseudomonad and investigation of physiological parameters influencing its production Appl. Environ. Microbiol. 58:353-358.
40
40. Shevchenko, A., M. Wilm, O. Vorm, and M. Mann. 1996. Mass spectrometric sequencing of proteins from silver stained polyacrylamide gels. Anal. Chem. 68:850-858.[Medline]
41
41. Slininger, P. J., J. E. van Cauwenberge, R. J. Bothast, D. M. Weller, L. S. Thomashow, and R. J. Cook. 1996. Effect of growth culture physiological state, metabolites, and formulation on the viability, phytotoxicity, and efficacy of the take-all biocontrol agent Pseudomonas fluorescens 2-79 stored encapsulated on wheat seeds. Appl. Microbiol. Biotechnol. 45:391-398.[CrossRef]
42
42. Smalla, K., G. Wieland, A. Buchner, A. Zock, J. Parzy, S. Kaiser, N. Roscot, H. Heuer, and G. Berg. 2001. Bulk and rhizosphere soil bacterial communities studied by denaturing gradient gel electrophoresis: plant-dependent enrichment and seasonal shifts revealed. Appl. Environ. Microbiol. 67:4742-4751.[Abstract/Free Full Text]
43
43. Smiley, R. W., and L.-M. Patterson. 1996. Pathogenic fungi associated with Fusarium foot rot of winter wheat in the semiarid Pacific Northwest. Plant Dis. 80:944-949.
44
44. Smith, K. P., and R. M. Goodman. 1999. Host variation for interactions with beneficial plant-associated microbes. Annu. Rev. Phytopathol. 37:473-491.[CrossRef][Medline]
45
45. Sun, X., M. Griffith, J. J. Pasternak, and B. R. Glick. 1995. Low temperature growth, freezing survival, and production of antifreeze protein by the plant growth promoting rhizobacterium Pseudomonas putida GR12-2. Can. J. Microbiol. 41:776-784.[Medline]
46
46. Thaning, C., C. J. Welch, J. J. Borowicz, R. Hedman, and B. Gerhardson. 2001. Suppression of Sclerotinia sclerotiorum apothecial formation by the soil bacterium Serratia plymuthica: identification of a chlorinated macrolide as one of the causal agents. Soil Biol. Biochem. 33:1817-1826.[CrossRef]
47
47. Tombolini R., D.-J. Van der Gaag, B. Gerhardson, and J. K. Jansson. 1999. Colonization pattern of the biocontrol strain Pseudomonas chlororaphis MA 342 on barley seeds visualized by using green fluorescent protein. Appl. Environ. Microbiol. 65:3674-3680.[Abstract/Free Full Text]
48
48. Verhille, S., M. Elomari, L. Coroler, D. Izard, and H. Leclerc. 1997. Phenotypically based taxonomy of fluorescent Pseudomonas strains isolated from four natural mineral waters. Syst. Appl. Microbiol. 20:137-149.
49
49. Verhille, S., N. Baida, F. Dabboussi, D. Izard, and H. Leclerc. 1999. Taxonomic study of bacteria isolated from natural mineral waters: proposal of Pseudomonas jessenii sp. nov. and Pseudomonas mandelii sp. nov. Syst. Appl. Microbiol. 22:45-58.
50
50. Weller, D. M. 1988. Biological control of soilborne plant pathogens in the rhizospere with bacteria. Annu. Rev. Phytopathol. 26:379-407.[CrossRef]
51
51. Williams, G. E., and M. J. C. Asher. 1996. Selection of rhizobacteria for the control of Pythium ultimum and Aphanomyces cochlioides on sugarbeet seedlings. Crop Prot. 15:479-486.[CrossRef]

---

Applied and Environmental Microbiology, November 2003, p. 6464-6474, Vol. 69, No. 11
0099-2240/03/$08.00+0     DOI: 10.1128/AEM.69.11.6464-6474.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Loper, J. E., Henkels, M. D., Shaffer, B. T., Valeriote, F. A., Gross, H. (2008). Isolation and Identification of Rhizoxin Analogs from Pseudomonas fluorescens Pf-5 by Using a Genomic Mining Strategy. Appl. Environ. Microbiol. 74: 3085-3093 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Johansson, P. M. Articles by Wright, S. A. I. Search for Related Content PubMed PubMed Citation Articles by Johansson, P. M. Articles by Wright, S. A. I. Agricola Articles by Johansson, P. M. Articles by Wright, S. A. I.