|
|
|
|
|
|
Scientific
Publications - Work Done by Microbiology Reader
Plant and Soil, 232 (1-2): 181-193, May 2001
Monitoring temporal and spatial variation in rhizosphere bacterial
population diversity: A community approach for the improved selection of
rhizosphere competent bacteria
V.J. Goddard, M.J. Bailey, P. Darrah, A.K. Lilley and I.P. Thompson
ABSTRACT The potential for developing a reliable strategy for selecting rhizosphere competent bacteria, based on an improved understanding of the community diversity and population dynamics of fluorescent pseudomonads, was investigated. Isolates from a collection of over 690 fluorescent pseudomonads, obtained from sugar beet and wheat plants grown in field soils in laboratory microcosms, were genotypically and phenotypically characterised. RFLP rRNA analysis (ribotyping) revealed that the sampled population was composed of 385 related but distinct ribotypes. Most ribotypes were isolated only once and represented a transient colonising population. However, representatives of 26 ribotypes were detected more often, of which five were isolated from rhizosphere soils sampled 7 months after the first sampling. Comparative phenotypic analysis of isolates (motility, antibiotic resistance and production, adherence, fatty acid composition, substrate utilisation patterns) demonstrated that the ability to utilise organic acids as carbon sources correlated with rhizosphere competence. Single inoculum and competitive colonisation studies in planted microcosms confirmed rhizosphere competence, but also demonstrated synergistic interactions. The colonisation ability and population densities of transient strains were significantly increased when co-inoculated with rhizosphere competent isolates. These data demonstrate potential cross-feeding and combined niche exploitation, rather than direct competition, confirming the multi-factorial nature of rhizosphere competence in diverse fluorescent pseudomonad communities. They also highlight the need to consider the use of mixed inocula for plant growth promotion and the systematic selection of strains for effective biotechnological exploitation.
Introduction Rhizosphere bacteria activities play a vital role in ecosystem functioning and some of these have been exploited for promoting plant growth and enhancing soil bioremediation by bioaugmentation. Amongst the most prodigious colonisers of the rhizosphere are the fluorescent pseudomonads. This group is of particular interest in terms of exploitation since it contains a high proportion of strains demonstrated to be effective at suppressing soil-borne plant diseases and the degradation of xenobiotics in the rhizosphere (Crowley et al., 1996; Weller, 1988). However, the performance of inoculants, in field and scaled-up applications, have commonly been inconsistent or short-lived (Handelsman and Stabb, 1996; Tchelet et al., 1999) and this has been largely attributed to poor colonisation and survival (McClure et al., 1991; Weller, 1988). The ecological performance of bacteria colonising the rhizosphere is complex and is affected by many different phenotypic traits and environmental parameters. Considerable effort has been invested into determining the biotic and abiotic parameters, as well as phenotypic traits, that together determine successful rhizosphere colonisation. These include amino acid and thiamine synthesis (Simons et al., 1996), lipopolysaccharide production (De Weger et al., 1989), agglutination and motility (De Weger et al., 1987). Although these studies have helped identify those factors that may be important in selecting for rhizosphere competence, the selection of rhizosphere competent strains remains largely empirical (Lugtenberg and Dekker, 1999). One aspect of rhizosphere microbiology that must be improved in order to select competent strains more reliably, is knowledge of the natural diversity and dynamics of target source populations, such as the fluorescent pseudomonads. The rhizosphere is a reservoir of genetically diverse fluorescent pseudomonad populations the composition of which is greatly determined in part by the selective influence of the plant and soil (Lemanceau et al., 1995; Latour et al., 1996). The selectivity of plants is not surprising, and is generally attributed to the influence of root exudates. These are composed, predominantly, of amino acids and sugars (Simons et al., 1997). The selective influence of the root is also temporal and this is reflected by changes in exudate composition and root architecture as the plant matures (Di Cello et al., 1997). Although the characteristics of the rhizosphere are known to be highly selective, few data are available which allow predictions to be made for the improved selection of rhizosphere competent strains, or indeed determine the characters that are associated with rhizosphere competence (Rainey, 1999). Community analysis, based on an improved understanding of the temporal and spatial distribution and genetic composition of rhizosphere communities, should provide more informed criteria for the selection of suitable strains for exploiting rhizosphere populations (Pierson and Weller, 1994). Furthermore, the introduction of novel genes into rhizobacteria provides the opportunity for generating effective biocontrol agents or enhancing xenobiotic degradation (Bangera and Thomashow, 1996; Fenton et al., 1992; Ripp et al., 2000; Schnider et al., 1995). The combination of these approaches offers the opportunity of an alternative strategy for development of inocula: One based on the systematic selection of rhizosphere competent strains that, if required, can be made more effective by introduction of novel genes. The aim of this study was firstly to investigate the genetic diversity of fluorescent pseudomonads, and their distribution, both spatially and temporally, in the rhizosphere of two different plant species grown in the same soil using simple laboratory microcosms. A second aim was to investigate the phenotypic traits of competent rhizosphere colonisers and compare them to transient forms. The third aim was to track the colonisation potential and population dynamics in the rhizosphere of bacterial inocula derived from competent and transient isolates in planted laboratory microcosms.
Materials and Methods The investigation comprised two studies: (1) Spatial diversity. Pseudomonad populations colonising the length of rhizosphere (upper, middle and growing tip) of two plant species, wheat, sugar beet and unplanted soil, were investigated. (2) Temporal diversity. Pseudomonad populations colonising the rhizosphere of wheat seedlings and soil, 5, 14 and 21 days after planting, were investigated.
Soils and plant The soil was classified as a heavy clay (pH = 7.4, sand = 24.9%, silt = 21.6% and clay = 53.5%), and was collected from the University of Oxford farm in Wytham (Oxfordshire). The field soil was mixed with a commercial sterile compost in a 2:1 (soil: compost) mixture, and placed in pots (diameter 12 cm, depth 15 cm) that were weighed and maintained at the same moisture level (30% w/v) by daily watering with sterile distilled water. Study 1 was established in September and Study 2 in April the following year using freshly collected field soil. All soils were potted on the day of collection. Three sugar beet (Beta vulgaris var. Amethyst) or wheat (Triticium aestivum var. Hussar) seeds were planted into separate pots and germinated in a growth cabinet (Fisons, internal volume 1 m3) in the dark at 24 °C and 60% relative humidity. After germination, the seedlings were exposed to a 16 h photo-period regime.
Bacterial sampling and enumeration For the spatial diversity study (1) 9 soil samples (approximately 1 g) were taken: 3 samples from each of 3 randomly selected pots. On the same day, 3 wheat and 3 sugarbeet plants, each taken from 6 randomly selected pots, were sampled. For the temporal diversity study (2) on each of 3 sampling days (5, 14 and 21 days after plantings), 3 plants were randomly selected from three separate pots. Similarly, 3 soil samples were taken from each of 3 randomly selected pots (9 soil samples in total). All samples were treated separately. The plants were shaken vigorously to remove excess soil and treated separately. Soil remaining adhered to the roots after shaking, was classified as rhizosphere soil and included in the sample. The roots were cut into three sections of equal lengths (upper, middle and growing tip) and placed in separate sterile polythene 50 ml centrifuge tubes, containing 10–15 sterile glass beads (5 mm in diameter) and shaken vigorously using a wrist action shaker for 3 min. Ten ml of quarter strength Ringer (QSR, Oxoid, UK) solution (BR25, Oxoid, UK) was added to the samples and then shaken for a further 3 min. Bacteria were isolated from unplanted soils using the same procedure. Bacteria were enumerated by spread plating 100 µl of 10-fold serial dilutions of root/soil suspension on to Tryptic Soy broth (TSB, Difco UK) solidified with 1.2% (w/v) agar (No.3 Oxoid, UK). Suspensions were also plated on Pseudomonas selective agar (PSA) (supplemented with 10 mg l-1 centrimide, 10 mg l-1 fucidin and 10 mg l-1 cephaloridin, PSA-CFC, Oxoid, UK, SR103E), which is a general selective supplement for pseudomonad isolation. The plates were inverted and incubated at 28 °C for 48 hours. Colonies were counted and expressed as colony forming units (cfu) g-1 fresh weight. Fluorescent pseudomonads were taken to be those colonies that fluoresced under UV light (wavelength 312 nm) when grown on PSACFC. For each sample, where possible, 15 colonies were picked at random from plates containing between 20 and 200 colonies; 45 colonies for each replicated root section sampled. The isolates were purified and sub-cultured on to Luria broth (LB) and stored frozen in 0.8 ml glycerol saline solution (70% glycerol, 30% NaCl solution [0.85%] vol./vol.) at -70 °C.
Genotypic characterisation Isolates were characterised by ribotyping, essentially as described by Ellis et al. (1999). Briefly, chromosomal DNA was extracted by the CTAB method (Bailey, 1995), purified and digested with Kpn1 to fragment complete rRNA operons (no Kpn1 sites were identified following ribosomal sequence data base comparisons). Genomic DNA fragments were separated by field inversion gel electrophoresis (FIGE, Bio-Rad, UK) using a 1% (w/v) pulsed field certified agarose (Bio-Rad, UK) in 0.5 × Tris-Borate buffer as recommended by the manufacturer. Lambda DNA molecular markers and a standard strain, P. fluorescens SBW25 (Rainey and Bailey, 1996) were included to allow comparison between gels. DNA was transferred overnight to nylon membranes Hybond-N+membrane (Amersham, UK) and hybridised with a fragment of DNA containing the 16/23S rRNA operon labelled with a32P dATP (Stratagene, UK). Ribosomal operon DNA hybridisation fragment patterns were identified and analysed by commercial pattern recognition software (Phoretix, UK)
Phenotypic traits. Five representative isolates of ribotypes that were detected more than once in wheat samples (ribotypes A, B , C, D, E), and thus taken to be rhizosphere competent, and five isolates only detected once colonising wheat and taken to be transient (ribotypes V, W, X, Y, Z), were selected for further analyse to determine whether rhizosphere competence was associated with specific phenotypic traits.
Motility This was assessed by measuring the diameter of the colony swarm on ‘sloppy agar’ using a method adapted from Grewal and Rainey (1991). Essentially single colonies were stabbed into the centre of tenth strength Pseudomonas agar F (PAF Difco, UK), solidified with 0.3% (w/v) agar. The distance swarmed through the sloppy agar was determined and the speed of motility, over the 48 h incubation period, calculated.
Antibiotic resistance Susceptibility to a range of antibiotics was tested using a testing kit (Oxoid HP053A) in which filter paper discs (6 mm diameter), impregnated with ampicillin (10 mg), carbenicillin (100 mg), chloramphenicol (50 mg), erythromycin (5 mg), gentamicin (10 mg), neomycin (30 mg), penicillin (5 mg), tetracycline (30 mg), were placed onto TSB agar, spread plated with each individual strain under investigation. Plates were incubated at 28 °C and examined for zones of inhibition, after for 48 h.
Phenazine production This was tested by streaking each isolate onto Potato Dextrose Agar (PDA, Oxoid, UK). A positive reaction was indicated by the presence of green crystals in the centre of the colonies after 2–3 days incubation at 28 °C (Ellis et al., 2000). A positive control, Pseudomonas fluorescens 2–79 was included (Mazzola et al., 1992).
Root adherence Wheat seeds (Triticium aestivum var. hussar) were surface sterilised by gentle shaking in 80% (v/v) ethanol for 30 s, and washing in sterile distilled water, before soaking in 0.5% Chloros, with 0.001% Tween-80, for 15 min. The seeds were washed 3 times in sterile distilled water for 2 min and dried on filter paper, and germinated in the dark, then allowed to grow for 4 days. The root tips (2 cm) were immersed for 20 min in bacterial suspension of strains grown overnight in LB, and washed in QSR. The root tips were washed gently in 10 ml sterile QSR for 15 s to remove loosely adhered bacteria, transferred into 10 ml of sterile QSR and shaken for 15 min, and the numbers of adhering cells determined by isolation on PSA-CFC agar incubated at 28 °C for 48 h.
Fatty acid methyl ester analysis Each isolate was analysed in triplicate, essentially as described by Thompson et al. (1993).
Nutrient assimilation The ability of isolates to utilise 95 sole carbon sources was determined using BioLOGb GN-Microplates, as described by Ellis et al. (1995).
Growth rate In order to determine whether growth rate (µ) and doubling time (tD) in specific sole carbon sources was a significant factor distinguishing the two population, these parameters were determined for each isolate in LB, and M9 (Sambrook et al., 1989) minimal media, containing either citric acid or mannitol. Analysis of the nutrient assimilation data (the extent of growth) (see above) indicated that organic acids, such as citric acid, were effective at distinguishing rhizosphere competent from transient forms, whereas utilisation of sugars such as mannitol showed no difference. Growth studies were performed in microtitre plates incubated at 28 °C for 22 h and examined every 15 min, using a plate reader (BioScreen C), at an absorbance of 600 nm.
Survival studies Three strains A, X and Y originating from wheat, were selected from the community composition studies, to be representative of ribotypes that were either rhizosphere competent (ribotype A was isolated more than once, See Tables 1 and 3) or transient (ribotype X and Y were only isolated once). Each strain was marked by the stable insertion of mini-Tn5 variants into their chromosome to provide novel selectable phenotypes suitable for competition studies. Variants of the disarmed miniTn5 transposon (Herrero et al., 1990), modified to carry the lacZY reporter and kanamycin resistance genes, were introduced from pUTTKZY into ribotype A, following the method of Timms-Wilson et al. (2000a). Similarly the genes for chloramphenicol and gentamicin resistance were introduced on pUT-mini-Tn5 variants into ribotypes X and Y, respectively. Transconjugants were enumerated by spread plating serial diluted samples onto PSA-CFC agar, amended with either 100 mg l-1 kanamycin and 25 X-gal mg l-1 (ribotype A), 50 mgl-1 chloramphenicol (ribotype X) or 5 mg l-1 gentamicin (ribotype Y). The survival of the marked strains was determined in unplanted and rhizosphere soils, using simple microcosms. These were composed of 15 cm diameter plants pots containing the field soil used for the original isolations described above, with or without wheat seeds. Broth cultures (24 h) of strains were spun down, washed in QSR and adjusted by dilution to ensure inocula densities (equivalent to 1.6×107 cfu g-1) were the same in all studies. The marked strains were inoculated into the microcosms as a soil drench, immediately prior to planting. Planted pots were germinated in the dark, then placed in a plant growth chamber (Fisons, internal volume 1m3), exposed to a 16 h photo-period, as described above. The relative densities and population dynamics of each of the inocula were determined, using the same sampling procedure described above. The samples were plated on PSA-CFC amended with the appropriate antibiotics.
Statistical analysis
Colony count data were log10 transformed and
tested for homogeneity of variance. Two-way analysis of variance (ANOVA) was
performed, and minimum significant differences calculated to identify
significant differences between mean counts, as described by Thompson et al.
(1999). Confidence limits on proportions were calculated using the ‘shortest
unbiased confidence limits for proportions’ tables (Table P) and interpolations
of Rohlf and Sokal (1994). Quantitative data obtained from FAME examination were
subjected to numerical analysis, as described by Thompson et al. (1993). The
relationship between nutrient utilisation profiles for each isolates was
determined by the Euc
Table 1. Frequencies (%) of pseudomonad ribotypes in the rhizospheres of wheat and sugarbeet and in unplanted soil. Relative abundances are given as the percentages that each ribotype represented in a sample of pseudomonads. These ribotypes were detected in more than one sample in our studies. The binomial 95% confidence limits are given for the percentages. Each sample was derived from three replicates. Field soil was collected in September
Results Viable bacterial counts In Study 1 (field soil collected September), counts on PSA-CFC of fluorescent pseudomonads detected in the rhizosphere of 14 day old plants ranged from 6×104 to 9×107 cfu g-1 for sugar beet, and 6×104 to 8×107 cfu g-1 for wheat, respectively. There was no significant difference in counts between wheat and sugar beet (P = 0.05) or along the root length (P = 0.05). Colony counts on TSBA (total culturable counts) were higher than those determined on PSA, ranging from 4×107 to 1×109 on sugar beet, and 8×107 to 4×108 on wheat. In Study 2 (field soil collected 7 months later in April), counts of bacteria were determined over three sampling occasions (5, 14 and 23 days) in the rhizosphere of wheat only. This revealed a rapid increase in densities between planting and the first sampling (day 5); the rate of increase then declined between day 5 and 14; eventually reaching population densities similar to those detected in Study 1 after 14 days. No significant changes in counts were detected between 14 and 23 days after planting. Fluorescent pseudomonad counts in the growing tip, 23 days after planting, were significantly lower (P = 0.05) than the middle and upper root sections. In unplanted soils, counts ranged from 5×104 to 3×105cfu g-1, and showed no significant difference over the three samplings. Variation in counts between replicate samples were typically within one log, however in a few cases counts varied by two log units.
Population diversity Study 1 – Spatial diversity
Genetic analysis revealed that the 370
isolates taken from sugar beet (102), wheat (121) and unplanted soil (147), were
composed of 54, 26 and 94 distinct ribotypes, respectively. Most ribotypes were
isolated only once, however 15 were isolated more often and these represented
49.2% of the sample population. The spatial distribution and relative abundances
of these 15 ribotypes in the rhizosphere of 14 day old wheat and sugar beet
plants and unplanted soil is represented in Table 1. Which gives the
percentages of each ribotype in the sample of pseudomonads. The fitness of the
strain is thus assessed and compared by these proportions. Of the 15 more
frequently isolated ribotypes, only 3, designated A, B and I, were detected in
all three habitats sampled. The remaining 12 ribotypes were unique to one
habitat. Considering the spatial distribution of ribotypes on the root, fewer
ribotypes were detected in the growing tips than the upper root. Table 1 gives
the 95% confidence limits for the proportions of each ribotype in the samples.
It is important to discern when the presence and absence of a ribotype, in
different samples, is significant. To
Table 2. The minimum proportions (%) a ribotype may appear in a sample and be detected with 95% confidence, for a range of sample sizes. Calculated from binomial distribution (Sokal and Rohlf, 1994).
Study 2 – Temporal diversity Three hundred and twenty-five isolates were randomly sampled on three sampling occasions (5,14 and 21 days from planting), 218 from the wheat rhizosphere and 107 from unplanted soil (including on the day of planting). This population was composed of 211 different ribotypes, including 5 (A, B, C, I and K) that were isolated in Study 1. As with Study 1, most ribotypes were isolated only once. However, 11 were isolated more than once, and these represented 44% of the sample taken from the rhizosphere and 54% of those isolated from unplanted soil, respectively. This data is presented, as above, as relative abundance, the proportion of each ribotype relative to the number of pseudomonads typed from a sample (Table 3). Ribotypes A, B, C, I and K were the most frequently isolated (36% of isolates taken over 21 days) and were identical to ribotypes isolated in Study 1, in soil taken from the same field site undertaken seven months previously. Only 5 ribotypes, collected more than once, were unique either to the rhizosphere (P, T and U) or to the soil (Q and S). Smaller sample sizes limit the testing of differences to ribotypes occurring in higher proportions. Some significant (P<0.05) changes with seedling development were been observed.
Phenotypic characteristics of ribotypes.
In order to compare which phenotypes might be correlated with rhizosphere competence, representatives of ribotypes isolated twice or more from wheat (sugar beet isolates were not studied) and unplanted soil, were selected for further study. The selected strains were representatives of ribotypes A–E. These were contrasted with a second group, consisting of ribotypes (V–Z) that were also isolated from wheat, but only once, and assumed to be poor root colonisers and transient in nature.
The two groups were assessed to determine
whether they could be distinguished on the basis of six phenotypic
characteristics. The rate of motility of each strain on agar ranged from 0.07 to
0.43 mm per, but overall no consistent difference could be detected between the
rhizosphere competent and transient strains. Similarly no consistent
differences in the resistances to seven antibiotics could be detected: all
strains were susceptible to ampicillin, erthyromycin and penicillin, and showed
a variable response to chloramphenicol, gentamicin, neomycin and tetracycline.
None of the strains produced phenazine or showed any difference in adherence to
the wheat
Table 3. Frequencies (%) of pseudomonad ribotypes in the rhizospheres of wheat and sugarbeet and in unplanted soil. Relative abundances are given as the percentages that each ribotype represented in a sample of pseudomonads. These ribotypes were detected in more than one sample in our studies. The binomial 95% confidence limits are given for the percentages. Each sample was derived from three replicates. Field soil was collected in April
The ability of all 10 isolates to assimilate 96 sole carbon sources was assessed using the BioLOGG system. These analyses indicate that the rhizosphere competent strains utilised a narrower range of carbon sources than transient forms. The five rhizosphere competent strains utilised 68% of the amino acids and 37% of the sugars, respectively, compared to 87% and 52%, respectively, for the transient forms. Further assessment, using cluster analysis of substrate utilisation patterns, demonstrated that the organic acids were particularly effective at differentiating the rhizosphere competent group from the transient colonisers (Figure 1). The organic acids that were effective at distinguishing rhizosphere competent and transient colonisers, in terms of presence or absence of growth, are shown in Table 4. When tested further, it was shown that the mean specific growth rates (µ), a function of doubling time (tD), of the competent strains in organic acid were significantly greater (P=0.01) than those of the transient strains. For example, the average growth rate of the competitive strains (ribotypes A–E) in citric acid was µ = 0.31 (mean tD = 3.22 h) compared to µ = 0.23 (mean tD = 4.35 h) for representatives of the transient ribotypes (ribotypes V–Z). No consistent difference in the growth rate of strains in mannitol was detected. Mannitol and citric acid were selected for further study, since the growth characteristics of the ribotypes in both compounds, was representative (data not shown) of those in the other and sugar and citric acids tested.
Figure 1. Dendrogram clustering Pseudomonas strains on the basis of their ability to utilise organic acids. Cluster 1 consists exclusively of rhizosphere competent strains (ribotypes A–E), whilst Cluster 2 were exclusively transient strains (ribotype V–Z).
Figure 2. Colony counts of genetically marked strains of ribotype A (E ) and Y (B ) recovered from unplanted soil microcosms. For each day, bars labelled with different letters are significantly different (P S 0.04) from the compared treatments. The initial inoculum density was approximately 1.6×107 cfu g-1 soil, for each ribotype.
Fate of introduced marked ribotypes
Three strains (A, X and Y), taken from ribotype populations that were demonstrated to have different persistence characteristics in the rhizosphere of wheat were selected and genetically marked in order to assess survival in soil and root colonisation, when introduced as an inoculum on wheat. The survival and colonisation ability of ribotype A, a rhizosphere competent ribotype and Y taken from the transient population, introduced separately into unplanted potted soils, were compared. Ribotype A survived significantly better (P<0.04) than ribotype Y. By 17 d after inoculation the population density of ribotype Y had declined, representing less than 1% of the density of ribotype A (Figure 2).
Similarly, when introduced as a seed dressing,
ribotype A survived at significantly (P<0.002) greater densities in the
seedling rhizosphere than either ribotypes X or Y. Fourteen days after
introduction, densities of ribotype A remained similar to the original
inoculum, whereas densities of ribotype X and Y declined to approximately 10%
and 1%, respectively (Figure 3). Furthermore, ribotype A had a much greater
impact on the composition of the indigenous microbial community than ribotype X
and Y. Colonies of ribotype A represented 80% of isolates developing on PSA-CFC
plated with root samples inoculated with ribotype A sampled 14 days after
introduction. In contrast, on plants inoculated with ribotypes X or Y alone,
isolates of these two strains represented 15% and 5%, respectively, of the
colonies isolated on PSA-CFC. The interaction of all three marked strains in the
rhizosphere, when co-inoculated on seed, was also investigated. When ribotypes X
and Y were coinoculated there was no significant synergistic interaction or
alteration of the survival (realised population density) of either strain.
However, co-inoculation of strains X or Y with ribotype A significantly improved
their survival (P=0.03), but neither strains X or Y had any detectable
influence on the survival of ribotype A (Figure 3).
Table 4. Organic acids effective at distinguishing rhizosphere competent (A–E) and transient strains isolated from wheat, determined using BioLOG® GN-MicroPlates
Figure 3. Counts of each marked ribotype recovered from the rhizosphere of 14 day old wheat seedlings, following inoculation alone, or in combination with other marked ribotypes. For each ribotype, bars labelled with different letters are significantly different from the compared treatments (P<_0.03). The initial inoculum density was 6.7×107 cfu g-1 soil.
Discussion
Many factors contribute to the performance of bacterial inocula and key amongst these is colonisation. Irrespective of the underlying mechanism controlling the desired activity, if the introduced strain does not survive and colonise, it will not be effective in disease suppression or bioremediation of contaminated environments. Poor survival and colonisation remains a major impediment to the wide-scale use of bioaugmentation. Many approaches have been used to improve the colonisation of introduced microbial strains, including increasing the dose introduced (Osburn et al., 1989) or the use of inocula composed of several genotypes (Pierson and Weller, 1994), which have had some success. The results of this study suggest that a major reason for the variable success of bioaugmentation in both biocontrol and bioremediation may be poor initial selection of strains. Traditionally, this has been achieved by screening large numbers of isolates for desired phenotypes, with little consideration of other essential traits, such as rhizosphere competence. An alternative approach, investigated in this study is to improve understanding of the diversity, distribution and dynamics of specific members of the community, and systematically select strains that are persistent and abundant. Our results certainly suggest this is an effective means of selecting strains, such as ribotype A, that was not only rhizosphere competent but also improved the performance of other introduced strains. Such competent strains may be further exploited as vectors of functional traits, introduced by genetic engineering. For example, we have demonstrated that introduction of the phzABCDEFG operon (encoding phenazine-1-carboxylic acid biosynthesis isolated from P. fluorescens 2–79), into a rhizosphere competent isolate, P. fluorescens 54/96, produced a transformant. This not only persisted in soils but also significantly enhanced the suppression of damping-off disease when compared to either wild type (Timms-Wilson et al., 2000b). The isolation of ribotypes with proven abilities to survive as part of the indigenous community, and as an inoculum, in microcosm studies at least, has provided an opportunity to examine some of the traits associated with rhizosphere competence and survival in the environment. Several phenotypic traits were examined including mobility, antibiotic production and adherence, since these have been suggested to be associated with successful root colonisation. However, only the ability to utilise a narrower range of organic acids faster, distinguished competent from transient strains. The importance of nutrients is not surprising, since these are know to be a major factor influencing rhizosphere populations, and amongst these, organic acids are significant constituents (Rovira, 1969). In a recent study using in vitro expression technology, it has been demonstrated that a high percentage of bacterial genes inducible by the rhizosphere were associated with nutrient acquisition and included amongst these were genes for organic acid utilisation (Rainey, 1999). Efficient nutrient scavenging systems are likely to be crucial for increased competitive ability in the rhizosphere. However, it is surprising that the range of organic acids utilised by ribotype A was narrower than the less competitive strains. The ability to utilise a diverse range of nutrients has previously been correlated to competitive ability in the rhizosphere (Bakker et al., 1993; Oresnik et al., 1998). It may be that, in some cases, the ability to grow rapidly on a narrower range of carbon sources, as we observed with ribotype A on citric acid, confers a competitive advantage if that carbon is abundant in the habitat. By recording the abundance of each ribotype in a sample, relative to the number of pseudomonads typed from that sample, the fitness (rhizosphere competence) of each strain has been assessed and compared with each other. This also avoids complications where differences may arise from changes in the population densities of bacteria or pseudomonads in general. Ideally, pseudomonad inocula will remain prominent in the indigenous ’pseudomonad community’ during their use. However, the complex selective pressures confronting an inoculum are similar to those for indigenous bacteria, where fitness fluctuates with time and habitat. We have seen that root type and position on the root may select for different ribotypes and in other studies have shown evidence for succession in pseudomonad populations associated with plant development and season (Rainey et al., 1994). The degree to which succession can be detected is of course determined by sample size, as the appearance and disappearance of strains from samples may be a function of the size of the samples. Table 2 indicates the limits of sensitivity of the range of sample sizes we have used. Given the relatively small numbers that are typed in most experiments, only changes to those populations occurring at high proportions will be reliably detected. The results of this study have several implications for biotechnological exploitation of bacterial inocula in field applications. Firstly, it is not possible to undertake extensive community analysis of every potential target habitat, and this may be seen as a real limitation for the approach described. Plant and soil habitats are known to be highly selective, supporting distinctive populations (Glandorf et al., 1993) and have been shown to be plant specific (Lemanceau et al., 1995). Indeed, most ribotypes in this study were detected in one habitat only. However, the detection of a few ribotypes (A, B and I) in all three habitats (sugar beet, wheat and unplanted soil) examined and also in field soils taken after a seven month interval, is strong evidence that a few ribotypes had an exceptional ability to persist. The initial observation from this study suggests that these ribotype populations may provide a source of inocula, which not only survive well, but can also persist in a broad range of habitats. Secondly, on introduction the ribotype A inoculum not only persisted, but also significantly improved the survival of inocula composed of less persistent ribotypes (Figure 3). Previous workers have demonstrated the greater effectiveness of inocula composed of mixed genotypes (Pierson and Weller, 1994; Weller and Cook, 1983). This has been attributed to the ability of the greater range of phenotypes to cope with continually fluctuating conditions or the more diverse ‘arsenal’ of mechanisms available for suppressing pathogens. The results of our study provide an alternative explanation – strains in mixed inocula can have a synergistic affect on the survival of other community members. This suggests that inocula composed of ribotype A for instance, may not only prove to be effective as a vector for biocontrol traits but can also be used effectively to enhance the persistence of less competitive, but desirable strains. The third important observation in terms of exploitation relates to the ability to differentiate rhizosphere competent forms, according to the ability to utilise a narrow range of organic acids (Table 4). This may offer a rapid diagnostic test for rhizosphere competence and reduce the dependence on expensive and labour intensive screens. For instance, in this study, it was possible to differentiate rhizosphere competent from transient forms according to the utilisation of just four organic acids (acetic, formic, proprionic and sebacic). The last significant observation relates more to risk than exploitation, was the observation that ribotype A, the rhizosphere competent type, had a much greater impact on the composition of the indigenous bacterial community than inocula composed of transient strains. This clearly deserves further investigation, as do the other observations made in this study. The reservoir of genetic diversity is clearly extensive, however, this can only be effectively exploited, in combination with applied molecular genetics, by improving our knowledge of the ecology and population dynamics of microbial populations that harbour potentially useful strains.
Acknowledgements
VJG was supported by a BBSRC Special Initiative (Plant, Soil, Microbial Interactions) Studentship.
References Bakker P A H M, Raaymakers J M and Schippers B 1993 Role of iron in the suppression of bacterial plant pathogens by fluorescent pseudomonads. In Iron Chelation in Plants and Soil Microorganisms. Eds. LL Barton and BC Hemmings. pp 269–281. San Diego: Academic Press. Bailey M J, Lilley A K, Thompson I P, Rainey R B and Ellis R J 1995 Site-directed chromosomal marking of a fluorescent pseudomonad isolated from the phytosphere of sugar beet: Stability and potential for marker gene transfer. Mol. Ecol. 4, 755–763. Bangera M G and Thomashow L S 1996 Characterization of a genomic locus required for synthesis of the antibiotic 2,4-diacetylphloroglucinol by the biological control agent Pseudomonas fluoresces Q2-87. Mol. Plant-Microbe Interact. 9, 83–90. Crowley D E, Brennerova M V, Irwin C, Brenner V and Focht D D 1996 Rhizosphere effects on biodegradation of 2,5-dichlorobenzoate by a bioluminescent strain of root-colonising Pseudomonas fluorescens. FEMS Microbiol. Ecol. 20, 79–89. De Weger L A, van der Vlugt C I M, Wijfjes A H M, Baker P A H M, Schipper B, Lugtenberg B J J 1987 Flagella of a plantgrowth-stimulating Pseudomonas fluorescens strain are required for colonization of potato roots. J. Bacteriol. 169, 2769–2773. De Weger L A, Bakker P A H M, Schippers B, Van Loosddrecht M C M and Lugtenberg B J J 1989 Pseudomonas spp. with mutational changes in the O-antigenic side chain of their lipopolysaccharide are affected in their ability to colonise potato roots. NATO ASI series 36. Ed. BJJ Lugtenberg. pp 197–202. Springer Verlag, Berlin. Di Cello F, Bevivino A, Chiarini L, Fani R, Paffetti D, Tabacchioni M and Dalmastri C 1997 Biodiversity of a Burkholderia cepacia population isolated from the maize rhizosphere at different plant growth stages. Appl. Environ. Microbiol. 11, 4485–4493. Ellis R J, Thompson I P, Bailey M J 1995 Metabolic profiling as a means of characterising plant-associated microbial communities. FEMS Microbiol. Ecol. 16, 9–18. Ellis R J, Thompson I P and Bailey M J 1999 Temporal fluctuations in the pseudomonad population associated with sugar beet leaves. FEMS Microbiol. Ecol. 28, 345–356. Ellis R J, Timms-Wilson T M and Bailey M J 2000 Identification of conserved traits in fluorescent pseudomonads with anti-fungal activity. Environ. Microbiol. 2, 274–284. Fenton A M, Stephens P M, Crowley J, O’Callaghan M and O’Gara F 1992 Exploitation of gene(s) involved in 2,4-diacetylphloroglucinol biosynthesis to confer a new biocontrol capability to a Pseudomonas strain. Appl. Environ. Microbiol. 58, 3873–3878. Glandorf D C M, Peters L G I, Van Der Sluis I, Bakker P A H M and Schippers B 1993 Crop specificity of rhizosphere pseudomonads and the involvement of root agglutins. Soil Biol. Biochem. 25, 981–989. Grewal S I S and Rainey P B 1991 Phenotypic variation of Pseudomonas putida and P. tolaasii affects the chemotactic response to Agaricus bisporus mycelial exudate. J. Gen. Micro. 137, 2761–2768. Handelsmann J and Stabb E V 1996 Biocontrol of soil borne plant pathogens. Plant Cell. 8, 1855–1869. Herrero M, De Lorenzo V and Timmis K N 1990 Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in Gram-negative bacteria. J. Bacteriol. 172, 6557–6567. Latour X, Corberand T, Laguerre G and Lemanceau P 1996 The composition of fluorescent pseudomonad populations associated with roots is influenced by plant and soil type. Appl. Environ. Microbiol. 62, 2449–2456. Lemanceau P, Corberand T, Gardan L, Latour X, Laguerre G, Boeufgras J-M and Alabouvette C 1995 Effect of two plant species, flax (Linum usitatissinum L.) and tomato (Lycopersicon esculentum Mill.) on the diversity of soil borne populations of fluorescent pseudomonads. Appl. Environ. Microbiol. 61, 1004–1012.
Lugtenberg B J and Dekker L C 1999 What makes
Pseudomonas Mazzola M, Cook R J, Thomashow L S, Weller D M and Pierson L S 1992 Contribution of phenazine antibiotic biosynthesis to the ecological competence of fluorescent pseudomonads in soil habitats. Appl. Environ. Microbiol. 58, 2616–2624. McClure N C, Fry J C and Weightman A J 1991 Survival and catabolic activity of natural and genetically engineered bacteria in a laboratory-scale activated-sludge unit. Appl. Environ. Microbiol. 57, 366–373. Oresnik I J, Pacarynuk L A, O’Brien S A P, Yost C K and Hynes M F 1998 Plasmid-encoded catabolic genes in Rhizobium leguminosarum bv. Trifolii: evidence for a plant-inducible rhamnose locus involves in competition for nodulation. Mol Plant Microbe Interact 11, 1175–1185. Osburn R M, Schroth M N, Hancock J G and Hendson M 1989 Dynamics of sugar beet seed colonization by Pythium ultimum and Pseudomonas species: effects on seed rot and damping-off. Phytopath. 79, 709–716. Pierson E A and Weller D M 1994 Use of mixtures of fluorescent pseudomonads to suppress take-all and improve the growth of wheat. Phytopathology 84, 841–947. Rainey P B, Bailey M J and Thompson I P 1994 Phenotypic and genotypic diversity of fluorescent pseudomonads isolated from the phyllosphere of field-grown sugar beet. Microbiology (UK) 140, 2315–2331. Rainey P B and Bailey M J 1996 Physical and genetic map of the Pseudomonas fluorescens SBW25 chromosome. Mol. Microbiol. 19, 521–533. Rainey P B 1999 Adaptation of Pseudomonas fluorescens to the plant rhizosphere. Environ. Microbiol. 1, 243–257. Ripp S, Nivens D E, Ahn Y, Werner C, Jarrell J, Easter J P, Cox C D, Burlage R S and Sayler G S 2000 Controlled field release of a bioluminescent genetically engineered micro-organism for bioremediation process monitoring and control. Environ. Sci. Technol. 34, 846–853. Rohlf and Sokal 1994 Statistical Tables F. James Rohlf, Robert R. Sokal 3rd edition. NY: W H Freeman & Co. Rovira A D 1969 Plant root exudates. Botan. Rev. 35, 35–57. Sambrook J, Fritsch E F and Manniatis T 1989 Molecular Cloning: A Laboratory Manual. Cold Spring Harbour, NY: Cold Spring Harbour Laboratory Press. Schnider U, Keel C, Blumer C, Troxler J, Defago G and Haas D 1995 Amplification of the housekeeping sigma factor in Pseudomonas fluorescens CHA0 enhances antibiotic production and improves biocontrol abilities. J. Bacteriol. 177, 5387–5392. Simons M, Van Der Bij A J, Brand I, De Weger L A, Wijffelman C A and Lugtenberg B J J 1996 Gnotobiotic systems for rhizosphere colonisation by plant-promoting Pseudomonas bacteria. Mol. Plant-Microbe. Interact. 9, 600–607. Simons M, Permentier H J, De Weger L A, Wijffelman C A and Lugtenberg B J J 1997 Amino acid synthesis is necessary for tomato root colonisation by Pseudomonas fluorescens strain WCS365. Mol. Plant-Microbe. Interact. 10, 102–106. Tchelet R, Meckenstock P, Steinle P and Van Der Meer J R 1999 Population dynamics of an introduced bacterium degrading chlorinated benezenes in a soil column and in sewage sludge. Biodegrad. 10, 13–125. Thompson I P, Bailey M J, Ellis R J and Purdy K J 1993 Subgrouping of bacterial populations by cellular fatty acid composition. FEMS Microbiol. Ecol. 102: 75–84. Thompson I P, Bailey M J, Ellis R J, Maguire N and Meharg A A 1999 Response of soil microbial communities to single and multiple doses of an organic pollutant. Soil Biol. Biochem. 31, 95–105. Timms-Wilson T M, Ellis R J and Bailey M J 2000a Immunocapture differential display method (IDDM) for the detection of environmentally induced promoters in rhizobacteria. J. Microbiol. Methods. 41, 77–84. Timms-Wilson T M, Ellis R J, Renwick A, Rhodes D J, Weller D M, Mavrodi D V, Thomashow L S and Bailey M J 2000b Chromosomal insertion of the phenazine biosynthetic pathway (phzABCDEFG) enhances the efficacy of damping off disease control by Pseudomonas fluorescens. Mol. Plant-Microbe. Interact. 13, 1293–1300. Weller D M and Cook R J 1983 Suppression of take-all of wheat by fluorescent pseudomonads. Phytopathology 73, 463–469. Weller D M 1988 Biological control of soil borne plant pathogens in the rhizosphere with bacteria. Ann. Rev. Phytopath. 26, 379–407.
(order Full Text from publisher)
|
© 2005
Transgalactic Ltd (manufacturer of Bioscreen C software) |
Privacy Statement | P.O. Box
1393, 00101 Helsinki, Finland,
Last modified: May 25, 2005
| ||||||
