Previous Article | Next Article 
Applied and Environmental Microbiology, April 2001, p. 1851-1864, Vol. 67, No. 4
0099-2240/01/$04.00+0 DOI: 10.1128/AEM.67.4.1851-1864.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.
Impact of Biocontrol Pseudomonas
fluorescens CHA0 and a Genetically Modified Derivative on the
Diversity of Culturable Fungi in the Cucumber Rhizosphere
M.
Girlanda,1,*
S.
Perotto,1
Y.
Moenne-Loccoz,2,3
R.
Bergero,1
A.
Lazzari,1
G.
Defago,2
P.
Bonfante,1 and
A.
M.
Luppi1
Dipartimento di Biologia Vegetale and
CSMT-CNR, 10125 Torino, Italy1;
Phytopathology Group, Institute of Plant Sciences, Swiss
Federal Institute of Technology (ETH), CH-8092 Zürich,
Switzerland2; and UMR CNRS Ecologie
Microbienne, Université Claude Bernard (Lyon 1), F-69622
Villeurbanne cedex, France3
Received 24 July 2000/Accepted 29 December 2000
 |
ABSTRACT |
Little is known about the effects of Pseudomonas
biocontrol inoculants on nontarget rhizosphere fungi. This issue was
addressed using the biocontrol agent Pseudomonas
fluorescens CHA0-Rif, which produces the antimicrobial
polyketides 2,4-diacetylphloroglucinol (Phl) and pyoluteorin (Plt) and
protects cucumber from several fungal pathogens, including
Pythium spp., as well as the genetically modified
derivative CHA0-Rif(pME3424). Strain CHA0-Rif(pME3424) overproduces Phl
and Plt and displays improved biocontrol efficacy compared with
CHA0-Rif. Cucumber was grown repeatedly in the same soil, which was
left uninoculated, was inoculated with CHA0-Rif or CHA0-Rif(pME3424),
or was treated with the fungicide metalaxyl (Ridomil). Treatments were
applied to soil at the start of each 32-day-long cucumber growth cycle,
and their effects on the diversity of the rhizosphere populations of
culturable fungi were assessed at the end of the first and fifth
cycles. Over 11,000 colonies were studied and assigned to 105 fungal
species (plus several sterile morphotypes). The most frequently
isolated fungal species (mainly belonging to the genera
Paecilomyces, Phialocephala, Fusarium, Gliocladium, Penicillium,
Mortierella, Verticillium, Trichoderma, Staphylotrichum, Coniothyrium,
Cylindrocarpon, Myrothecium, and Monocillium) were
common in the four treatments, and no fungal species was totally
suppressed or found exclusively following one particular treatment.
However, in each of the two growth cycles studied, significant
differences were found between treatments (e.g., between the control
and the other treatments and/or between the two inoculation treatments)
using discriminant analysis. Despite these differences in the
composition and/or relative abundance of species in the fungal
community, treatments had no effect on species diversity indices, and
species abundance distributions fit the truncated lognormal function in
most cases. In addition, the impact of treatments at the 32-day mark of
either growth cycle was smaller than the effect of growing cucumber
repeatedly in the same soil.
| |
INTRODUCTION |
Introduction of beneficial
microorganisms into soil or the rhizosphere has been proposed for
biological control of soilborne crop diseases (6, 20, 34).
In certain cases, genetically modified (GM) strains with increased
expression of biocontrol traits have been developed to improve
biocontrol efficacy (8, 27, 53). However, the release of
large populations of biocontrol agents into the environment raises
important biosafety issues related to the possible ecological
consequences of such introductions on resident populations and
ecosystem functioning (14). This concern is of primary
relevance in the case of GM inoculants (8, 19, 27, 61),
and in many countries the existing regulatory framework makes their
deliberate field release strictly dependent on detailed assessment of
their potential environmental impact and associated risks
(49).
Soilborne fungal pathogens cause considerable damage to crop plants
(1), and they have been often targeted in biocontrol (20, 34, 60). However, the majority of soil fungi are not pathogenic, and a large number of them may even be beneficial to plants
and/or contribute positively to ecosystem functioning. Indeed,
nonpathogenic saprotrophic microfungi perform key ecological roles in
the soil ecosystem through decomposition of organic matter, nutrient
cycling, natural control of plant pathogens, and a myriad of other
functions (9, 11, 17). Common rhizosphere fungi are well
documented as decomposers of celluloses and hemicelluloses (Trichoderma, Penicillium, and Fusarium), as well
as chitin (Mortierella) (18). The ability of
certain saprotrophic Fusarium and Trichoderma strains to protect plants against pathogenic fungi through competition, parasitism, antagonism, and/or induced resistance is also well known
(2, 5, 23). In this context, it is surprising that saprotrophic rhizosphere fungi have been largely neglected as nontarget, beneficial resident microorganisms potentially affected by
bacterial biocontrol inoculants, especially when the latter produce or
overproduce antifungal metabolites with a relatively broad range of
action. Indeed, investigations of the ecological impact of biocontrol
bacteria have focused mainly on the effects on crops, on nontarget
resident bacteria, and on ecosystem functioning (15, 41, 45,
48).
The few studies dealing with nontarget fungi have mostly monitored the
impact of GM inoculants with antifungal biocontrol traits on total
fungal counts (reviewed in reference 65). These studies
may allow the assessment of catastrophic effects on the resident fungi,
but they do not address the possibility of specific changes in
microfungal community organization, e.g., in terms of the relative
abundance of fungal species. Such alterations in the composition and
structure of fungal communities might have immediate or lasting effects
on ecosystem functioning (35), as there is now
experimental evidence of a link between microbial biodiversity and the
maintenance and regulation of ecosystem functions (46).
Mathematical methods to analyze fungal diversity data are still the
subject of considerable debate in mycological literature, especially in
the case of soil microfungal communities and/or when ecological
interpretation of community response to perturbation is attempted
(16, 24, 71, 72). Species abundance distribution analysis
may provide both a complete mathematical description of the data and
information on resource-partitioning patterns among component species
in a given assemblage (71, 72). For large, species-rich
equilibrium communities, the species abundance distribution is usually
lognormal, while for species-poor nonequilibrium communities under a
harsh environmental regime a geometric series often pertains
(40), thus making modeling a useful tool to examine the
effects of disturbance. Species richness and dominance indices provide
simpler information but may be useful when comparing treatments (40). Multivariate analysis techniques (especially
ordination methods) have also been used to analyze soil fungal
communities and generate hypotheses on the factors involved in
community changes (see, e.g., references 66 to 68).
In this study, the ecological impacts of the biocontrol agent
Pseudomonas fluorescens CHA0-Rif and its GM derivative
CHA0-Rif(pME3424) on the diversity of the culturable microfungal
assemblages in the rhizosphere of cucumber (Cucumis sativus
L.) were examined. P. fluorescens CHA0-Rif produces several
bioactive compounds, including the antimicrobial polyketides
2,4-diacetylphloroglucinol (Phl) and pyoluteorin (Plt), and can protect
cucumber against Pythium ultimum Trow (32, 34,
63). P. ultimum rapidly infects seeds and causes both
pre- and postemergence damping-off of cucumber seedlings, but it can
produce root rots even at later plant growth stages (1).
The GM strain P. fluorescens CHA0-Rif(pME3424) overproduces
the antimicrobial compounds Phl and Plt and displays enhanced
biocontrol activity against P. ultimum (53).
Phl and Plt inhibit the growth of a broad spectrum of bacteria and
fungi (21, 25, 32, 55, 60).
In the present work, a soil with low disease pressure was chosen, so
that the potential negative impacts of inoculation on nontarget fungi
could not be compensated for by the biocontrol effects of the
inoculants. The inoculation treatments were compared with a control (no
inoculation) and a chemical treatment, in which soil was treated with
metalaxyl (Ridomil), a phenylamide fungicide with selective action
almost exclusively against Peronosporales (Oomycetes) (12, 54). The chemical treatment
served as positive control since (i) metalaxyl is one of the main
chemical fungicides currently used against Pythium spp. and
(ii) CHA0 and its derivatives are being studied as potential biocontrol
agents against these fungal pathogens. Chemical fungicides may be
applied several times within a given growing season and/or in
successive growing seasons, and this is also relevant for biocontrol
products. Therefore, several cucumber growth cycles were carried out in
the same soil, and treatments (bacterial inoculum or metalaxyl) were
applied to soil at the start of each cycle. Since the objective of this work was to assess whether treatments could have an impact on the
composition and structure of rhizosphere microfungal assemblages, different approaches (species abundance distributions, diversity indices, and multivariate analysis) were followed for the description and characterization of the fungal community.
| |
MATERIALS AND METHODS |
Bacterial strains.
P. fluorescens CHA0-Rif
(47) is a spontaneous rifampin-resistant mutant of the
wild-type strain CHA0 (57). Strains CHA0 and CHA0-Rif
display the same growth rate and produce the same amounts of Phl and
Plt in laboratory media. The ecology of strain CHA0-Rif has been
assessed in the field on several occasions (62). Strain
CHA0-Rif(pME3424) carries a recombinant plasmid (i.e., pME3424
[53]) constructed by inserting a copy of the
rpoD gene (coding for sigma factor 70) of CHA0 into the IncP
vector pVK100 (36). Introduction of the plasmid into the
pseudomonad caused a severalfold increase in the amounts of Phl and Plt
produced in vitro and resulted in enhanced suppression of P. ultimum-mediated damping-off of cucumber in soil microcosms
(53).
Experimental setup.
Soil was collected from the surface
horizon of a fallow located at Eschikon (Zürich, Switzerland).
The soil at the site corresponds to a cambisol and was described by
Natsch et al. (47) and Guntli et al. (29).
Disease pressure is usually low in this soil, unless it is
experimentally inoculated with phytopathogenic fungi. In that case,
Eschikon soil becomes disease conducive, but plant protection can be
achieved using CHA0 as inoculant, both in the field and in soil
microcosms (13, 69). This has been observed with several
fungal pathogens, including P. ultimum. The soil was
air-dried until friable and passed through a 5-mm-mesh screen.
The four treatments studied were as follows: (i) inoculation with
CHA0-Rif, (ii) inoculation with CHA0-Rif(pME3424), (iii) addition of
the chemical fungicide metalaxyl, and (iv) no treatment (control). At
the start of the experiment, the soil was sprayed (17 ml kg of
soil
1) with either a cell suspension of CHA0-Rif, a cell
suspension of CHA0-Rif(pME3424), a metalaxyl solution (at 1 mg
ml1), or sterile distilled water. Inoculations resulted
in 9 × 106 CFU of introduced pseudomonads per g of
dry soil. The soil was mixed thoroughly to ensure an even distribution
of inoculants and chemical fungicide and placed in pots. Each
3.8-dm3 pot contained the equivalent of 3.5 kg of dry soil.
Cucumbers seeds (cultivar Chinesische Schlange) were obtained from R. Geissler (Zürich, Switzerland). They were sown at a
depth of 1 cm
(four seeds per pot), and the pots were placed in
a growth chamber set
at 70% relative humidity with 16 h of light
(160 mE m
2
s
1) at 22°C and 8 h of dark at 18°C. The soil
was sprayed with 50
ml of sterile distilled water per pot per day on
six consecutive
days. The number of seedlings was adjusted to three per
pot at
10 days after sowing. Each pot received 100 ml of sterile
distilled
water, poured in the plate under the pot, at 2, 3, and 4 weeks
after
sowing.
The pots were emptied at 32 days after sowing. The root systems were
shaken to dislodge weakly adhering soil, which was mixed
with bulk soil
and put back into the pots. On the same day, each
treatment was applied
a second time on the same pots, at the same
rates, except that volumes
were sprayed onto the soil surface
without mixing the soil (which
mimics certain commercial protocols
for inoculation). Cucumber was sown
and a second 32-day-long cycle
of plant growth was carried out. The
same procedure was repeated
three times for a total of five cycles of
cucumber
growth.
Sampling of rhizosphere, isolation of fungi, and identification
at species level.
The effects of treatments on the culturable
rhizosphere microfungi were investigated in the first and fifth cycles
of cucumber growth. Strain CHA0-Rif was present at 7.6 and 5.6 log CFU
g of root1 and CHA0-Rif(pME3424) was present at 6.9 and
4.9 log CFU g of root1 at the end of the first and fifth
cycles, respectively. At the end of each of the two cycles, one plant
was randomly chosen from each pot to study rhizosphere microfungi. The
roots were excised and weighed after the excess soil had been removed.
The roots were then washed under mechanical agitation (by vortexing) in sterile distilled water, blotted dry, and weighed again. Following determination of the amounts of suspended soil, the rhizosphere suspension was adjusted to reach a final dilution of 1:10,000. Aliquots
(2 ml) of the final suspension were then plated on 2% malt agar (20 g
of malt extract, 18 g of agar, 1 liter of distilled water)
acidified to pH 5.5 (with HCl) and containing chloramphenicol (150 ppm). Plates were maintained at room temperature (average temperature,
22°C). Most isolates were obtained after a few days of incubation,
but plates were checked over several weeks to allow isolation of
slow-growing fungi.
Identification of fungal isolates at the species level was carried out
on the basis of classical macro- and microscopic criteria
(taking into
account cultural features and morphology of vegetative
and asexual or
sexual reproductive structures), with the help
of pertinent keys and
literature (
30,
64). Morphotypes were
recognized for
sterile isolates. More than 11,000 isolates were
thus
characterized.
Statistical design, data reduction, and analyses.
Each
treatment was studied with 10 replicates. The experiment followed a
randomized block design, with ten blocks and one pot (i.e., one
replicate) per treatment per block. Each rhizosphere suspension was
plated onto five plates, and data obtained from each plate were
analyzed separately.
The soil dilution plate method is effective for detecting changes due
to pollution, management practices, or environmental
disturbances
(
22) but tends to favor the most-sporulating species
(
24,
72). Therefore, data were expressed with regard to
fungal
frequency (i.e., the number of plates in which the species
occurred)
rather than numbers of CFU in all data analyses, except when
comparing
fungal population levels or considering species abundance
distribution
(because this would have resulted in too few abundance
classes).
Differences in CFU and species numbers (including sterile morphotypes)
among treatments within each cucumber growth cycle,
as well as between
the two cycles within each treatment, were
tested for significance by
means of two-way analysis of variance
with post hoc simple contrasts.
SYSTAT (version 5.2; SYSTAT, Evanston,
Ill.) was used to perform the
test (
P < 0.05).
Species abundance distributions.
Species abundance
distribution analysis was performed by testing the fit to the four main
species abundance distribution models (geometric series, logarithmic
series, lognormal, and broken stick). The truncated form of the
lognormal function was used. The procedure for fitting the models
consisted of calculating the number of species expected in each
abundance class and comparing it with the number of species actually
observed. Expected species abundances for each model were derived as
described by Magurran (40). Expected and observed values
were compared using a chi-square goodness-of-fit test. The fit was
considered statistically significant when P was >0.05.
Diversity indices.
Diversity indices represent a useful mean
to quantify community diversity and have been instrumental in revealing
the impact of biocontrol inoculants on resident populations
(48). Here, several diversity indices were used to compare
treatment effects. Both species richness indices (weighing towards
uncommon species) and indices based on the proportional abundances of
species (weighing towards abundant species) were computed.
Among species richness indices, Margalef's diversity measure
(
DMg), the log series
index, Shannon's
index (
H'), and Brillouin's
index (
HB) were
chosen. Margalef's index was calculated from the
formula
DMg = (
S 
1)/ln
N (here and throughout,
S is the number
of taxa
and
N is the total number of individuals); Shannon's index
was calculated from the formula
H' =
Spi(ln
pi), where
pi is
the proportion of individuals found in the
ith species (in a sample,
the true value of
pi is unknown but is estimated as
ni/
N, [here
and throughout,
ni is the number of individuals in the
ith species]);
and Brillouin's index was calculated from
the formula
HB = [ln
N!
S(ln
ni!)]/
N. The log series
index is
derived in the course
of the calculations for fitting the log series
model of species
abundance distribution and was obtained from the
equation
=
[
N (1
x)]/
x, in which
x is estimated from the iterative solution
of
S/
N = (1
x)/
x[
ln (1
x)].
Among indices based on the proportional abundance of species,
Simpson's index (
D) and Berger-Parker's index
(
d) were computed.
Simpson's index was calculated as
D =
S {[
ni
(
ni 1)]/[
N (N 1)]}.
Berger-Parker's index expresses the proportional importance
of the
most abundant species and was obtained using the formula
d =
Nmax/
N, where
Nmax is the number of individuals in the most
abundant
species.
Treatments were compared within each cucumber growth cycle studied,
using two-way analysis of variance with post hoc simple
contrasts
(
P < 0.05). Similarly, each treatment was compared
across
the two cycles. SYSTAT version 5.2 was
used.
Discriminant analysis.
Discriminant analysis (DA [also
known as canonical variate analysis]) is an ordination technique based
on an a priori partition of objects into groups: the new axes, called
the canonical variates, are located so as to reveal the best
discrimination among groups. In this study, the underlying assumption
tested was that different treatments would result in different fungal assemblages.
Since the method is considerably robust against the violation of linear
data structures, it can be used without knowing any
property of the
data set (
50), and untransformed data (frequency
values)
were used in the analysis. In DA, the number of variables
should never
exceed the number of objects, because otherwise the
analysis will not
run due to singularity problems (
50); therefore,
the
rarest fungal species (occurring with only one plant in one
or several
treatments in either cycle) were excluded from the
analysis. With this
lower bound, 34 fungal taxa (including sterile
morphotypes) were
retained and subjected to the analysis. The
analysis was performed
using the SYN-TAX 5.0 package subroutine
"Canonical Variates" with
the "Spherized scores of objects" (normalization
of eigenvectors)
option. Correlations with the original variables
were also analyzed. DA
biplots were taken into account to evaluate
the relationships between
the original variables and the new axes
graphically. These biplots
showed lines connecting the origin
to the variable positions as well as
isodensity circles representing
each group (circles drawn around group
centroids and expected
to contain 95% of the observations within each
group).
| |
RESULTS |
Effect on fungal propagule counts and species numbers.
Treatments had no effect on the number of fungal CFU within the first
cycle (i.e., 3.0 × 105 to 3.3 × 105
CFU per g of rhizosphere soil) or the fifth cycle (i.e., 1.2 × 105 to 5.6 × 105 CFU per g of rhizosphere
soil) of cucumber growth (Table 1). However, when treatments were compared
over the two samplings, the numbers of CFU were significantly lower for
the control in the fifth cycle than in the first one.
In total, over 11,000 colonies were identified and assigned to 105 fungal species (plus 50 sterile morphotypes) (Table
1).
Many fungal
species, which were the most abundant, were found
in all treatments in
both cycles. The main exception was
Mycelium sterile
moniliaceum 1, which was absent from the first cycle.
No
statistically significant difference in species numbers (sterile
morphotypes included) was found among treatments at either of
the two
cycles studied or when the treatments were compared across
both
cycles.
Effect on species abundance distributions.
Distributions of
fungal species in the rhizosphere following treatments with the
control, CHA0-Rif(pME3424), and metalaxyl could be described both by
the log series and the truncated lognormal models in each of the two
growth cycles studied. However, results from the chi-square test
indicated that the truncated lognormal function yielded a better fit in
all cases. Instead, neither the log series nor the truncated lognormal
model was appropriate for the CHA0-Rif treatment in the first cycle,
mainly because the log series model predicted fewer rare species and
the truncated lognormal model predicted fewer species of
intermediate-high abundance than were actually recorded. In the fifth
cycle, the fungal assemblage from the same treatment did not fit the
log series, because of several discrepancies between actual and
expected data for rare species and species with high abundances, but
fit the truncated lognormal model (Fig.
1).
View larger version (33K):
[in this window]
[in a new window]
|
FIG. 1.
Species abundance distribution for rhizosphere fungal
assemblages in the control, the CHA0-Rif, the CHA0-Rif(pME3424), and
the metalaxyl treatments at the end of the first and the fifth cycles
of cucumber growth. The abundance of each species is given based on CFU
data. The x axis corresponds to the octave abundance classes
(upper limit of each abundance class) where the different species
identified have been ranked; the y axis corresponds to the
actual or expected number of fungal isolates falling within each
abundance class. Bars, observed values; solid line, expected values
according to the truncated lognormal function; dotted line, expected
values according to the log series function. S, number of
taxa, N, number of CFU. The chi-square and associated
probability values for the models are also indicated. Details of
calculations are as described by Magurran (40). The fit is
statistically significant when P is >0.05. For the first
cycle, both the truncated lognormal model and the log series model were
statistically significant for the control, the CHA0-Rif(pME3424), and
the metalaxyl treatments. For the fifth cycle, the truncated lognormal
model was statistically significant for each of the four treatments,
and the log series model was statistically significant for the control,
the CHA0-Rif(pME3424), and the metalaxyl treatments.
|
|
None of the species abundance distributions obtained from the different
treatments in the first or the fifth cycle fit the
broken stick model
or the geometric series (
P 
0.05). For the
broken stick
model, this resulted from the fact that this function
predicted fewer
rare species than were recorded, except for the
control in the fifth
cycle (the number of species of intermediate
abundance was too large to
fit the model). Instead, the main reason
that data did not conform to
the geometric series was that the
species with the highest rank (i.e.,
abundance) were more abundant
than was predicted by the
model.
Effect on diversity indices.
When diversity indices were
computed (Table 2), no significant
difference was detected between the treatments in either cycle. This
was observed with indices based on the proportional abundance of
species (Shannon's and Brillouin's indices), dominance measures (Simpson's and Berger-Parker's indices), and species richness indices
(log series and Margalef's indices). In a comparison between
cycles, a significant difference was found for the control with log
series index, for the CHA0-Rif and the metalaxyl treatments with
Berger-Parker's index.
Discriminant analysis of treatment effects.
DA was carried out
to compare the four treatments in the first cycle of cucumber growth
(Fig. 2). The first axis (50% of the total variation) discriminated between the control and the other three
treatments, but there was no difference between the latter along this
axis. The second axis (27% of the total variation) distinguished the
CHA0-Rif(pME3424) treatment from the CHA0-Rif treatment. In the fifth
cycle of cucumber growth, the first DA axis (67% of the total
variation) mainly discriminated between the CHA0-Rif and the
CHA0-Rif(pME3424) treatments (Fig. 3).
Axis 2 (20% of the total variation) failed to discriminate between treatments.
View larger version (24K):
[in this window]
[in a new window]
|
FIG. 2.
DA biplot showing group centroids, isodensity circles,
and correlations of fungal species with canonical variates for fungal
assemblages at the end of the first cycle of cucumber growth. Group A,
control; group B, CHA0-Rif; group C, CHA0-Rif(pME3424); group D,
metalaxyl. Fungal species names (values in parentheses are correlation
with axis 1 and correlation with axis 2, respectively): 1, Acremonium roseo-griseum (0.26, 0.04); 2, Cladosporium cladosporioides (0.32, 0.004); 3, Coniothyrium cerealis (0.06, 0.19); 4, C. fuckelii (0.01, 0.21); 5, C. destructans (0.07, 0.04); 6, V. nigrescens (0.13, 0.27); 7, Fusarium equiseti (0.30, 0.20); 8, F. oxysporum (0.08, 0.32); 9, F. solani (0.10,
0.23); 10, F. tabacinum (0.04, 0.07); 11, G. roseum (0.19, 0.21); 12, Monocillium mucidum (0.08, 0.18); 13, Mortierella alpina (0.26, 0.02); 14, Mucor hiemalis f. hiemalis (0.27, 0.14); 15, Mucor racemosus f. racemosus (0.23, 0.35);
16, Myrothecium roridum (0.11, 0.07); 17, Myrothecium verrucaria (0.20, 0.24); 18, P. marquandii (0.24, 0.06); 19, P. rugulosum (0.31,
0.01); 20, P. humicola (0.38, 0.30); 21, Staphylotrichum coccosporum (0.05, 0.14); 22, Trichoderma hamatum (0.29, 0.01); 23, Trichoderma
harzianum (0.15, 0.09); 24, Trichosporon beigelii
(0.12, 0.14); 25, Verticillium chlamydosporium (0.18,
0.22); 26, Exophiala sp. (0.003, 0.14);27,
Humicola fuscoatra (0.10, 0.31); 28, P. canescens (0.08, 0.12); 29, S. chartarum (0.11,
0.15); 30, S. cylindrospora (0.24, 0.36); 31, T. angustata (0.07, 0.11); 32, Mycelium sterile
moniliaceum 1 (0.03, 0.14); 33, Mycelium sterile
dematiaceum 4 (0.05, 0.08). The three canonical variates
accounted for 63.3, 14.7, and 7.3% of the total variation,
respectively.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 3.
DA biplot showing group centroids, isodensity circles,
and correlations of fungal species with canonical variates for fungal
assemblages at the end of the fifth cycle of cucumber growth. Group A,
control; group B, CHA0-Rif; group C, CHA0-Rif(pME3424); group D,
metalaxyl. Fungal species names (values in parentheses are correlation
with axis 1 and correlation with axis 2, respectively): 1, A. roseo-griseum (0.12, 0.34); 2, C. cladosporioides
(0.27, 0.23); 3, C. cerealis (0.16, 0.05); 4, C. fuckelii (0.01, 0.13); 5, C. destructans (0.07, 0.18); 6, V. nigrescens (0.02, 0.09); 7, F. equiseti (0.28, 0.02); 8, F. oxysporum (0.15, 0.14); 9, F. solani (0.36, 0.10); 10, F. tabacinum (0.41, 0.02); 11, G. roseum (0.20, 0.20); 12, M. mucidum (0.07, 0.16); 13, M. alpina (0.09, 0.15); 14, M. hiemalis f.
hiemalis (0.25, 0.10); 15, M. roridum (0.06,
0.08); 16, M. verrucaria (0.16, 0.14); 17, P. marquandii (0.02, 0.11); 18, P. rugulosum (0.00, 0.02); 19, P. humicola (0.05, 0.21); 20, S. coccosporum (0.07, 0.22); 21, T. hamatum ( 0.31, 0.28); 22, T. harzianum (0.05, 0.08); 23, T. beigelii (0.05, 0.04); 24, V. chlamydosporium
(0.08, 0.30); 25, Exophiala sp. (0.22, 0.22); 26, Mycelium sterile moniliaceum 1 (0.10, 0.06); 27, H. fuscoatra (0.03, 0.08); 28, P. canescens (0.18,
0.01); 29, S. chartarum (0.12, 0.08); 30, S. cylindrospora (0.21, 0.02); 31, T. angustata
(0.31, 0.22); 32, Mycelium sterile moniliaceum 2 (0.08, 0.03); 33, Mycelium sterile dematiaceum 4 (0.21, 0.05).
The three canonical variates accounted for 66.9, 20.3, and 12.8% of
the total variation, respectively.
|
|
When the control and the CHA0-Rif treatments were compared in both
cucumber growth cycles (Fig.
4), the
first axis (74% of
the total variation) distinguished the control in
the first cycle
from the other three situations. Whereas this axis
clearly discriminated
the CHA0-Rif treatment from the control in the
first cycle, the
two treatments could not be distinguished in the fifth
cycle.
The distance between the two controls was greater than the
distance
between the CHA0-Rif treatment and the respective control in
either
cycle. Axis 2 (20% of the total variation) discriminated the
bacterial
treatment in the first cycle from the other three situations.
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 4.
DA biplot showing group centroids, isodensity circles,
and correlations of fungal species for fungal assemblages in the
control and the CHA0-Rif treatments, at the end of the first and the
fifth cycles of cucumber growth. Group A, control in the first cycle;
group B, CHA0-Rif in the first cycle; group C, control in the fifth
cycle; group D, CHA0-Rif in the fifth cycle. Fungal species names
(values in parentheses are correlation with axis 1 and correlation with
axis 2, respectively): 1, A. roseo-griseum (0.30, 0.02); 2, C. cladosporioides (0.14, 0.32); 3, C. cerealis
(0.10, 0.09); 4, C. fuckelii (0.10, 0.28); 5, C. destructans (0.02, 0.11); 6, V. nigrescens (0.26, 0.04); 7, F. equiseti (0.13, 0.52); 8, F. oxysporum (0.42, 0.52); 9, F. solani (0.12, 0.41);
10, F. tabacinum (0.11, 0.12); 11, G. roseum
(0.14, 0.23); 12, M. mucidum (0.14, 0.11); 13, M. alpina (0.08, 0.40); 14, M. hiemalis f.
hiemalis (0.49, 0.12); 15, M. racemosus f.
racemosus (0.11, 0.66); 16, M. roridum (0.02, 0.13); 17, M. verrucaria (0.04, 0.40); 18, P. marquandii (0.09, 0.31); 19, P. rugulosum (0.03,
0.36); 20, P. humicola (0.13, 0.51); 21, S. coccosporum (0.47, 0.36); 22, T. hamatum (0.06,
0.49); 23, T. harzianum (0.26, 0.44); 24, T. beigelii (0.09, 0.28); 25, V. chlamydosporium
(0.48, 0.70); 26, Exophiala sp. (0.22, 0.18); 27, Mycelium sterile moniliaceum 1 (0.41, 0.14); 28, H. fuscoatra (0.01, 0.37); 29, P. canescens (0.11, 0.02); 30, S. chartarum (0.20, 0.22); 31, S. cylindrospora (0.09, 0.50); 32, T. angustata (0.07,
0.31); 33, Mycelium sterile moniliaceum 2 (0.06, 0.42);
34, Mycelium sterile dematiaceum 4 (0.19, 0.22). The three
canonical variates accounted for 74.0, 19.6, and 6.4% of the total
variation, respectively.
|
|
When the control and the CHA0-Rif(pME3424) treatments of both cucumber
growth cycles were compared (Fig.
5),
each of the four
situations was clearly different from the others along
the first
axis (82% of the total variation). The distance between the
bacterial
treatment and the respective control was greater in the fifth
cycle than in the first cycle but was always smaller than the
distance
between the two controls. Axis 2 (only 11% of the total
variation)
mainly distinguished the control from the bacterial
treatment in the
fifth cycle.
View larger version (19K):
[in this window]
[in a new window]
|
FIG. 5.
DA biplot showing group centroids, isodensity circles,
and correlations of fungal species for fungal assemblages in the
control and the CHA0-Rif(pME3424) treatments, at the end of the first
and the fifth cycles of cucumber growth. Group A, control in the first
cycle; group B, CHA0-Rif(pME3424) in the first cycle; group C, control
in the fifth cycle; group D, CHA0-Rif(pME3424) in the fifth cycle.
Fungal species names (values in parentheses are correlation with axis 1 and correlation with axis 2, respectively): 1, A. roseo-griseum (0.26, 0.03); 2, C. cladosporioides
(0.20, 0.28); 3, C. cerealis (0.01, 0.17); 4, C. fuckelii (0.15, 0.01); 5, C. destructans (0.05, 0.11); 6, V. nigrescens (0.20, 0.42); 7, F. equiseti (0.14, 0.23); 8, F. oxysporum (0.59, 0.21);
9, F. solani (0.40, 0.02); 10, F. tabacinum
(0.23, 0.14); 11, G. roseum (0.40, 0.21); 12, M. mucidum (0.11, 0.08); 13, M. alpina (0.18, 0.29); 14, M. hiemalis f. hiemalis (0.50, 0.13); 15, M. roridum (0.03, 0.06); 16, M. verrucaria
(0.04, 0.09); 17, P. marquandii (0.02, 0.14); 18, P. rugulosum (0.17, 0.25); 19, P. humicola (0.03, 0.07); 20, S. coccosporum (0.68, 0.23); 21, T. hamatum (0.04, 0.36); 22, T. harzianum (0.39, 0.22);
23, T. beigelii (0.16, 0.20); 24, V. chlamydosporium (0.66, 0.25); 25, Exophiala sp.
(0.28, 0.31); 26, Mycelium sterile moniliaceum 1 (0.61, 0.05); 27, H. fuscoatra (0.08, 0.20); 28, P. canescens (0.04, 0.13); 29, S. chartarum
(0.39, 0.25); 30, S. cylindrospora (0.21, 0.16); 31, T. angustata (0.07, 0.03); 32, Mycelium sterile
moniliaceum 2 (0.10, 0.06); 33, Mycelium sterile
dematiaceum 4 (0.38, 0.11). The three canonical variates
accounted for 81.6, 10.9, and 7.5% of the total variation,
respectively.
|
|
When the control and the metalaxyl treatments of both cucumber growth
cycles were compared (Fig.
6), the first
axis (54% of
the total variation) distinguished the control from the
metalaxyl
treatment for the first cycle but not for the fifth cycle.
Along
this axis the distance between the two controls was smaller. The
second axis (33% of the total variation) distinguished the metalaxyl
treatment of the first cycle from the two treatments of the fifth
cycle.
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 6.
DA biplot showing group centroids, isodensity circles,
and correlations of fungal species for fungal assemblages in the
control and the metalaxyl treatments, at the end of the first and the
fifth cycles of cucumber growth. Group A, control in the first cycle;
group B, metalaxyl in the first cycle; group C, control in the fifth
cycle; group D, metalaxyl in the fifth cycle. Fungal species names
(values in parentheses are correlation with axis 1 and correlation with
axis 2, respectively): 1, A. roseo-griseum (0.34,
0.16); 2, C. cladosporioides (0.32, 0.18); 3, C. cerealis (0.03, 0.19); 4, C. fuckelii (0.11, 0.07);
5, C. destructans (0.01, 0.14); 6, V. nigrescens (0.19, 0.26); 7, F. equiseti (0.28,
0.22); 8, F. oxysporum (0.21, 0.70); 9, F. solani
(0.02, 0.37); 10, F. tabacinum (0.05, 0.04); 11, G. roseum (0.07, 0.52); 12, M. mucidum (0.11, 0.16); 13, M. alpina (0.17, 0.31); 14, M. hiemalis f.
hiemalis (0.44, 0.36); 15, M. racemosus f.
racemosus (0.21, 0.21); 16, M. roridum (0.13,
0.19); 17, M. verrucaria (0.33, 0.28); 18, P. marquandii (0.11, 0.00); 19, P. rugulosum (0.12,
0.32); 20, P. humicola (0.33, 0.12); 21, S. coccosporum (0.09, 0.78); 22, T. hamatum (0.24,
0.16); 23, T. harzianum (0.20, 0.63); 24, T. beigelii (0.16, 0.16); 25, V. chlamydosporium (0.02,
0.74); 26, Exophiala sp. (0.07, 0.42); 27, Mycelium
sterile moniliaceum 1 ( 0.10, 0.61), 28, H. fuscoatra (0.12, 0.12), 29, P. canescens (0.19, 0.03), 30, S. chartarum (0.17, 0.25), 31, T. angustata (0.16, 0.13), 32, Mycelium sterile
moniliaceum 2 (0.04, 0.12), 33, Mycelium sterile
dematiaceum 4 (0.15, 0.27). The three canonical variates
accounted for 54.3, 33.3, and 12.4% of the total variation,
respectively.
|
|
Correlations between fungal species and canonical variates are shown in
Fig.
2 to
6. Some species were found to display low
correlations with
canonical variates, i.e., to discriminate little
among the treatments
studied in either growth cycle (Fig.
2 and
3). These species were often
among the commonest based on their
overall occurrence, e.g.,
Fusarium tabacinum and
Cylindrocarpon destructans
in the first cycle and
Penicillium rugulosum, Coniothyrium fuckelii, Verticillium nigrescens, Paecilomyces marquandii in
the
fifth cycle (Table
1). Some species consistently discriminated
between
the CHA0-Rif and the CHA0-Rif(pME3424) treatments in both
cycles; e.g.,
Fusarium solani and
Fusarium oxysporum were
associated
with the CHA0-Rif treatment, and
Exophiala sp.,
Penicillium canescens,
and
Stachybotrys chartarum
were associated with the CHA0-Rif(pME3424)
treatment (Fig.
2 and
3).
Other species instead appeared to be
associated with one bacterial
treatment in one cycle and with
the other bacterial treatment in the
other. For instance,
Gliocladium roseum was associated with
the CHA0-Rif(pME3424) treatment in
the first cycle and the CHA0-Rif
treatment in the fifth cycle,
whereas
Stachybotrys
cylindrospora and
Truncatella angustata were
associated
with the CHA0-Rif treatment in the first cycle and
the
CHA0-Rif(pME3424) treatment in the fifth cycle (Fig.
2 and
3).
| |
DISCUSSION |
Many biocontrol strains of fluorescent Pseudomonas spp.
produce the antimicrobial polyketides Phl and/or Plt active against phytopathogenic fungi (33, 56). The ability of P. fluorescens CHA0-Rif to produce Phl and Plt was increased
following the introduction of pME3424 in the strain, and the resulting
derivative protected cucumber better against P. ultimum-mediated damping-off (53). In the current
work, colony counts performed at the end of each cycle showed that the
ability of the inoculants to colonize the rhizosphere of cucumber
declined steadily with time, so that population levels of culturable
cells of the inoculant strains present at the end of the fifth cucumber
growth cycle were smaller than those found at the end of the first
cycle. This suggests that growing cucumber repeatedly in the same soil
favored resident microorganisms that were better adapted to the
environmental conditions prevailing in the rhizosphere than CHA0-Rif
or, especially, CHA0-Rif(pME3424) was. How this translated in terms of
ecological impact on resident fungi is difficult to assess since (i)
the biocontrol effect of CHA0 starts shortly after inoculation (i.e.,
at a time where population levels of the inoculants were still close to
inoculation level) and (ii) the exact time at which interactions
between indigenous fungi and introduced bacteria may take place is
unknown. Population levels of the inoculants at the end of each growth
cycle were still in the order of 105 CFU per g of root or
higher, which is generally considered enough for disease suppression by
biocontrol pseudomonads (51). However, it is possible that
the impact on resident culturable fungi in each cycle could have been
bigger at an earlier sampling, and that mostly resilience was being
recorded at 32 days (48).
Isolation of microfungi from the rhizosphere of cucumber yielded a
broad fungal spectrum (Table 1) dominated by genera and species rather
widespread and frequently found in agricultural soils, rhizospheres,
and roots of crop plants (the main exception being Phialocephala
humicola, which is mostly a forest soil fungus) (3, 4, 18,
26). This fungal spectrum overlaps the one obtained by Hong
(31), who found that rhizosphere fungi protect cucumber
seedlings against damping-off caused by Fusarium oxysporum f. sp. cucumerinum. No Oomycetes were isolated in
this study, despite the fact that they can grow on the laboratory
medium used. This is in accordance with the fact that disease pressure
is usually low in Eschikon soil. The choice of experimental conditions
not conducive to plant disease in such a study enabled us (i) to assess potential negative ecological impacts of a biocontrol inoculant without
the interference of the positive biocontrol effect on target pathogens
(45) and (ii) to avoid large-scale plant mortality in the
unprotected control treatment (56). Perhaps
Oomycetes were present at population levels too low to be
detected, or perhaps they were not competitive enough on the plates.
Soil and rhizosphere fungal communities have been shown to fit the
lognormal species abundance distribution model (39), but
Zak (71) found a logarithmic function, not a lognormal
function, to be typical for root surface fungal assemblages, and Thomas and Shattock (59) observed that phylloplane fungal
populations were best described by both the geometric and the log
series model. In this study, most fungal assemblages fit both the log
series function and the truncated lognormal model, but the latter
provided a better fit (Fig. 1). Polluted or stressed communities are
often characterized by a switch from the lognormal model (arising in response to many independent factors controlling the abundance of a
heterogeneous collection of organisms) to the geometric series (associated with species-poor habitats, where dominant species preempt
a significant portion of a limiting resource of the habitat and reach
population levels proportional to the amount of resource utilized)
(40, 71). Against this background, it appears that neither
those treatments nor the fact of growing cucumber repeatedly in the
same soil represented a significant perturbation for the fungal
community of the rhizosphere. No conclusion could be drawn for the
CHA0-Rif treatment in the first cucumber growth cycle, as none of the
models tested were appropriate in this case.
These findings were strengthened by comparing fungal diversity levels
(Table 2). Indeed, the majority of diversity indices did not highlight
any difference between the fungal assemblages studied. Statistically
significant differences were obtained only within treatments over the
two cycles, fungal diversity being higher in the fifth cycle than in
the first cycle. Instead, no significant effect of the different
treatments was revealed in either cucumber growth cycle, indicating
that the species richness and evenness of the fungal assemblages were
unaffected by such treatments.
However, significant differences between treatments were found when
fungal diversity was analyzed in more detail, using a DA approach (Fig.
2 to 6). Depending on the growth cycle, such differences were found
between the control and each of the three other treatments and/or
between the two inoculation treatments. Correlations were found between
treatments and the occurrence of certain fungal species in both
cucumber growth cycles, as illustrated by DA biplots. Seed inoculation
with P. fluorescens E6 resulted in enhanced colonization of
zinnia roots by Fusarium spp. and reduced colonization by
Penicillium spp. at 3 weeks (70), which did not
take place in this work with either inoculant (data not shown). The
effects of the CHA0-Rif(pME3424) treatment were not greater in
intensity than those of the CHA0-Rif treatment when compared with the
effects of the control (Fig. 2 and 3). Some of the differences between
the CHA0-Rif and CHA0-Rif(pME3424) treatments could already be
anticipated by comparing the results for fungal species listed in Table
1, as, e.g., in the fifth cycle Fusarium solani was
recovered from 7 of 10 plants in the CHA0-Rif treatment but from only 2 plants in the CHA0-Rif(pME3424) treatment. An effect of the metalaxyl
treatment was also detected (Fig. 4), although this fungicide is
thought to act almost exclusively against Peronosporales
(12, 54). Peronosporales were not found in this
work, but metalaxyl may display a wider specificity towards fungi
considering that it inhibits rRNA synthesis. Indeed, there is evidence
that arbuscular mycorrhizal fungi are sensitive to metalaxyl
(58). Interestingly, treatment effects were greater in the
fifth cycle than in the first cycle for the CHA0-Rif(pME3424) treatment
(Fig. 5), while the effects of the CHA0-Rif treatment (Fig. 4) and the
metalaxyl treatment (Fig. 6) had decreased by the fifth cycle. However,
the effects of the bacterial treatments were less than those linked to
cucumber monoculture (Fig. 4 to 6).
Whether the effects of the bacterial treatments resulted from the
ability of the inoculants to produce (or overproduce) the antimicrobial
compounds Phl and Plt remains to be ascertained. Under gnotobiotic
conditions, it can be shown using a translational phlA'-'lacZ fusion that the biosynthetic gene
phlA involved in Phl synthesis is expressed when CHA0
colonizes the cucumber rhizosphere (R. Notz, M. Maurhofer, U. Schnider-Keel, D. Haas, and G. Défago, unpublished).
CHA0-Rif(pME3424) displayed enhanced protection of cucumber against
P. ultimum compared with CHA0 (44), confirming previous disease suppression data obtained using the prototype plasmid
pME3090 (42), from which pME3424 was constructed.
Interestingly, high-pressure liquid chromatography analysis indicated
that CHA0 can produce Phl and Plt in the rhizosphere and that the
concentrations reached by these polyketides in the rhizosphere were
higher when CHA0 contained pME3090 (44). It is considered
that Phl-positive biocontrol pseudomonads should reach population
levels of 105 CFU per g of root (or more) for effective
production of Phl (52), which was the case here. Further
work will assess whether repeated inoculations of CHA0-Rif or
CHA0-Rif(pME3424) resulted in the enrichment of fungi resistant to Phl
and/or Plt.
Interactions between CHA0-Rif or CHA0-Rif(pME3424) and the resident
culturable fluorescent pseudomonads in the cucumber rhizosphere were
not mediated by the production of antimicrobial polyketides (48). Likewise, some of the effects of the bacterial
treatments were perhaps not mediated by Phl or Plt. For instance,
positive interactions can take place between introduced biocontrol
pseudomonads and resident antagonistic fungi, which may contribute to
suppression of Fusarium wilt of radish (38).
Indeed, interactions among component species have been recognized to
play an important role in shaping fungal communities (66),
and bacteria can have an indirect effect on saprotrophic microfungi by
influencing interfungal interactions. This may generate great
complexity and lead to the occurrence of unexpected phenomena in
natural communities (10, 37). In addition, one of the
mechanisms by which CHA0 can protect against disease corresponds to
induced systemic resistance (43). Plant physiology is
different when systemic resistance has been induced (23),
which in turn may influence the way fungi can colonize the rhizosphere.
This issue deserves further attention.
Comparable studies on the effects of introduced biocontrol bacteria on
indigenous rhizosphere bacterial populations have shown that the impact
of the inocula can vary depending on the experimental conditions
(28). Therefore, it will be necessary to extend impact assessment of biocontrol bacterial inoculants on the indigenous fungal
community to take into account the various environmental conditions
prevailing in the field (e.g., from one year to the next), as well as
crop rotation conditions. The effect of fungal disease pressure is
another important issue that will have to be addressed in further work,
as fungal pathogens can influence the growth and physiology of roots
and the release of root exudates (1), which in turn will
likely affect saprotrophic fungi.
In conclusion, DA revealed that inoculation with P. fluorescens CHA0-Rif or its GM derivative CHA0-Rif(pME3424) had a
detectable influence on the fungal community of the cucumber
rhizosphere at 32 days in the two cycles studied. The effects of the
two inoculants were different, but in both cases the magnitude of the
impact was small. The most frequently isolated fungal species were
common to the different treatments, and no fungal species was totally suppressed or found exclusively in one particular treatment. No dramatic change in species abundance distribution or diversity level
was observed following the various treatments, indicating that
community organization was not profoundly altered, and treatment effects were equal to or smaller than the effects of growing cucumber repeatedly in the same soil.
| |
ACKNOWLEDGMENTS |
We thank Felipe Wettstein, Carsten Hase, Zensi Hopf, and Fabio
Mascher (ETH) for technical help.
This work was supported by the EU IMPACT2 project (BIO4-CT96-0027) and
by grant MURST (40%).
| |
FOOTNOTES |
*
Corresponding author. Mailing address: Dipartimento di
Biologia Vegetale and CSMT-CNR, Viale P.A. Mattioli 25, 10125 Torino, Italy. Phone: 39 011 6707446 47. Fax: 39 011 6707459. E-mail: girlanda{at}bioveg.unito.it.
| |
REFERENCES |
| 1.
|
Agrios, G. N.
1997.
Plant pathology, 4th ed.
Academic Press, San Diego, Calif.
|
| 2.
|
Alabouvette, C., and C. Steinberg.
1995.
Suppressiveness of soils to invading micro-organisms, p. 3-12.
In
H. M. T. Hokkanen, and J. M. Lynch (ed.), Plant and microbial biotechnology research series, vol. 4. Biological control: benefits and risks. Cambridge University Press, Cambridge, United Kingdom.
|
| 3.
|
Barron, G. L.
1968.
The genera of Hyphomycetes from soil.
The Williams and Wilkins Company, Baltimore, Md.
|
| 4.
|
Booth, C.
1971.
The genus Fusarium.
Commonwealth Mycological Institute, Kew, United Kingdom
|
| 5.
|
Chet, I.,
J. Inbar, and Y. Hadar.
1997.
Fungal antagonists and mycoparasites, p. 165-184.
In
D. T. Wicklow, and B. Söderström (ed.), The mycota, vol. 4. Environmental and microbial relationships. Springer-Verlag, Berlin, Germany.
|
| 6.
|
Cook, R. J.
1993.
Making greater use of introduced microorganisms for biological control of plant pathogens.
Annu. Rev. Phytopathol.
31:53-80[CrossRef][Medline].
|
| 7.
|
Cook, R. J.
1996.
Assuring the safe use of microbial biocontrol agents a need for policy based on real rather than perceived risks.
Can. J. Plant Pathol.
18:439-445.
|
| 8.
|
Cook, R. J.,
W. L. Bruckart,
J. R. Coulson,
M. S. Goettel,
R. A. Humber,
R. D. Lumsden,
J. V. Maddox,
M. L. Memanus,
L. Moore,
S. F. Meyer,
P. C. Quimby,
J. P. Stack, and J. L. Vaughn.
1996.
Safety of microorganisms intended for pest and plant disease controla framework for scientific evaluation.
Biol. Control
7:333-351[CrossRef][Medline].
|
| 9.
|
Cooke, R. C., and A. D. M. Rayner.
1984.
Ecology of saprotrophic fungi.
Longman, London, United Kingdom.
|
| 10.
|
Culver, D. C.
1992.
Introduction to the theory of species interactions, p. 229-241.
In
G. C. Carroll, and D. T. Wicklow (ed.), The fungal community: its organization and role in the ecosystem, 2nd ed. Marcel Dekker, Inc, New York, N.Y.
|
| 11.
|
Curl, E. A., and B. Truelove.
1986.
The rhizosphere.
Springer-Verlag, Berlin, Germany.
|
| 12.
|
Davidse, L. C.
1995.
Phenylamide fungicidesbiochemical action and resistance, p. 347-354.
In
H. Lyr (ed.), Modern selective fungicides: properties, applications, modalities of action. Gustav Fischer Verlag, Iena, Germany.
|
| 13.
|
Défago, G.,
C.-H. Berling,
S. Henggeler,
W. Hungerbühler,
H. Kern,
P. Schleppi,
E. W. Stutz, and M. Zürrer.
1987.
Survie d'un Pseudomonas fluorescens dans le sol et protection du blé contre les maladies d'origine fongique.
Schweiz. Landw. Fo.
26:155-160.
|
| 14.
|
Défago, G.,
C. Keel, and Y. Moënne-Loccoz.
1997.
Fate of released Pseudomonas bacteria in the soil profile: Implications for the use of genetically-modified microbial inoculants, p. 403-418.
In
J. T. Zelikoff, J. M. Lynch, and J. Shepers (ed.), Ecotoxicology: responses, biomarkers and risk assessment. SOS Publications, Fair Haven, N.J.
|
| 15.
|
de Leij, F. A. A. M.,
E. J. Sutton,
J. M. Whipps, and J. M. Lynch.
1994.
Effect of a genetically modified Pseudomonas aureofaciens on indigenous microbial populations of wheat.
FEMS Microbiol. Ecol.
13:249-257.
|
| 16.
|
Dighton, J.
1994.
Analysis of micromycete communities in soil: a critique of methods.
Mycol. Res.
98:796-798.
|
| 17.
|
Dix, N. J., and J. Webster.
1995.
Fungal ecology.
Chapman and Hall, London, United Kingdom.
|
| 18.
|
Domsch, K. H.,
W. Gams, and T. H. Anderson.
1980.
Compendium of soil fungi.
IHW-Verlag, Eching, Germany.
|
| 19.
|
Doyle, J. D., and G. Stotzky.
1993.
Methods for the detection of changes in the microbial ecology of soil caused by the introduction of microorganisms.
Microb. Releases
2:63-72.
|
| 20.
|
Dunne, C.,
I. Delany,
A. Fenton,
S. Lohrke,
Y. Moënne-Loccoz, and F. O'Gara.
1996.
The biotechnology and application of Pseudomonas inoculants for the biocontrol of phytopathogens, p. 441-448.
In
G. Stacey, B. Mullin, and P. M. Gresshoff (ed.), Biology of plant-microbe interactions. IS-MPMI, St. Paul, Minn.
|
| 21.
|
Dutrecq, A.,
P. Debras,
J. Stevaux, and M. Marlier.
1991.
Activity of 2,4-diacetylphloroglucinol isolated from a strain of Pseudomonas fluorescens to Gaeumannomyces graminis var. tritici, p. 252-257.
In
A. B. R. Beemster, G. J. Bollen, M. Gerlagh, M. A. Ruissen, B. Schippers, and A. Tempel (ed.), Biotic interactions and soil-borne diseases. Elsevier Science Publishers, Amsterdam, The Netherlands.
|
| 22.
|
Fritze, H., and E. Bååth.
1993.
Microfungal species composition and fungal biomass in a coniferous forest soil polluted by alkaline deposition.
Microb. Ecol.
25:83-92.
|
| 23.
|
Fuchs, J.-G.,
Y. Moënne-Loccoz, and G. Défago.
1997.
Nonpathogenic Fusarium oxysporum strain Fo47 induces resistance to Fusarium wilt in tomato.
Plant Dis.
81:492-496.
|
| 24.
|
Gams, W.
1992.
The analysis of communities of saprophytic microfungi with special reference to soil fungi, p. 183-223.
In
W. Winterhoff (ed.), Fungi in vegetation science. Kluwer Academic Publishers, Dordrecht, The Netherlands.
|
| 25.
|
Garagulya, A. D.,
E. A. Kirprianova, and O. I. Boiko.
1974.
Antibiotic effect of bacteria from the genus Pseudomonas on phytopathogenic fungi.
Mikrobiol. Zh.
36:197-202.
|
| 26.
|
Gerlach, W., and H. Nirenberg.
1982.
The genus Fusariuma pictorial atlas.
Biologischen Bundesanstalt für Land und Forstwirtschaft, Berlin, Germany.
|
| 27.
|
Giddings, G.
1998.
Tansley review no. 99. The release of genetically engineered micro-organisms and viruses into the environment.
New Phytol.
140:173-184[CrossRef].
|
| 28.
|
Gilbert, G. S.,
J. L. Parke,
M. K. Clayton, and J. Handelsman.
1993.
Effects of an introduced bacterium on bacterial communities on roots.
Ecology
74:840-854[CrossRef].
|
| 29.
|
Guntli, D.,
S. Burgos,
Y. Moënne-Loccoz, and G. Défago.
1999.
Calystegine degradation capacities of microbial rhizosphere communities of Zea mays (calystegine-negative) and Calystegia sepium (calystegine-positive).
FEMS Microbiol. Ecol.
28:75-84.
|
| 30.
|
Hawksworth, D. L.,
P. M. Kirk,
B. C. Sutton, and D. N. Pegler.
1995.
Ainsworth and Bisby's dictionary of the fungi, 8th ed.
CAB International, Wallingford, United Kingdom
|
| 31.
|
Hong, C.-Y.
1969.
The relation between rhizosphere fungi and the occurrence of damping-off of cucumber seedlings.
Ann. Phytopathol. Soc. Jpn.
25:308-314.
|
| 32.
|
Keel, C.,
U. Schnider,
M. Maurhofer,
C. Voisard,
J. Laville,
U. Burger,
P. Wirthner,
D. Haas, and G. Défago.
1992.
Suppression of root diseases by Pseudomonas fluorescens CHA0: Importance of the bacterial secondary metabolite 2,4-diacetylphloroglucinol.
Mol. Plant-Microbe Interact.
5:4-13.
|
| 33.
|
Keel, C.,
D. M. Weller,
A. Natsch,
G. Défago,
R. J. Cook, and L. S. Thomashow.
1996.
Conservation of the 2,4-diacetylphloroglucinol biosynthesis locus among fluorescent Pseudomonas strains from diverse geographic locations.
Appl. Environ. Microbiol.
62:552-563[Abstract].
|
| 34.
|
Keel, C., and G. Défago.
1997.
Interactions between beneficial soil bacteria and root pathogens: mechanisms and ecological impact, p. 27-46.
In
A. C. Gange, and V. K. Brown (ed.), Multitrophic interactions in terrestrial systems. Blackwell Scientific Publishers, London, United Kingdom.
|
| 35.
|
Kennedy, A. C., and K. L. Smith.
1995.
Soil microbial diversity and the sustainability of agricultural soils.
Plant Soil
170:75-86[CrossRef].
|
| 36.
|
Knauf, V. C., and E. W. Nester.
1982.
Wide host range cloning vectors: a cosmid clone bank of an Agrobacterium Ti plasmid.
Plasmid
8:45-54[CrossRef][Medline].
|
| 37.
|
Lavelle, P.
1987.
Interactions, hiérarchies et régulations dans le sol: a la recherche d'une nouvelle approche conceptuelle.
Rev. Ecol. Biol. Sol
24:219-229.
|
| 38.
|
Leeman, M.,
F. M. Den Ouden,
J. A. Van Pelt,
C. Cornelissen,
A. Matamala Garros,
P. A. H. M. Bakker, and B. Schippers.
1996.
Suppression of fusarium wilt of radish by co-inoculation of fluorescent Pseudomonas spp and root-colonizing fungi.
Eur. J. Plant Pathol.
102:21-31.
|
| 39.
|
Lussenhop, J.
1981.
Analysis of microfungal component communities, p. 37-45.
In
D. T. Wicklow, and G. C. Carroll (ed.), The fungal community. Its organization and role in the ecosystem. Marcel Dekker, Inc, New York, N.Y.
|
| 40.
|
Magurran, A. E.
1988.
Ecological diversity and its measurement.
Chapman and Hall, London, United Kingdom.
|
| 41.
|
Mahaffee, W. F., and J. W. Kloepper.
1997.
Bacterial communities of the rhizosphere and endorhiza associated with field-grown cucumber plants inoculated with a plant growth-promoting Rhizobacterium or its genetically modified derivative.
Can. J. Microbiol.
43:344-353[Medline].
|
| 42.
|
Maurhofer, M.,
C. Keel,
U. Schnider,
C. Voisard,
D. Haas, and G. Défago.
1992.
Influence of enhanced antibiotic production in Pseudomonas fluorescens strain CHA0 on its disease suppressive capacity.
Phytopathology
82:190-195[CrossRef].
|
| 43.
|
Maurhofer, M.,
C. Hase,
P. Meuwly,
J.-P. Métraux, and G. Défago.
1994.
Induction of systemic resistance of tobacco to tobacco necrosis virus by the root-colonizing Pseudomonas fluorescens strain CHA0: Influence of the gacA gene and of pyoverdine production.
Phytopathology
84:139-146[CrossRef].
|
| 44.
|
Maurhofer, M.,
C. Keel,
D. Haas, and Défago.
1995.
Influence of plant species on disease suppression by Pseudomonas fluorescens strain CHA0 with enhanced antibiotic production.
Plant Pathol.
44:40-50[CrossRef].
|
| 45.
|
Moënne-Loccoz, Y.,
J. Powell,
P. Higgins,
J. Britton, and F. O'Gara.
1998.
Effect of the biocontrol agent Pseudomonas fluorescens F113 released as sugarbeet inoculant on the nutrient contents of soil and foliage of a red clover rotation crop.
Biol. Fertil. Soils
27:380-385[CrossRef].
|
| 46.
|
Naeem, S.,
L. J. Thompson,
S. P. Lawler,
J. H. Lawton, and R. M. Woodfin.
1994.
Declining biodiversity can alter the performance of ecosystems.
Nature
368:734-737[CrossRef].
|
| 47.
|
Natsch, A.,
C. Keel,
H. A. Pfister,
D. Haas, and G. Défago.
1994.
Contribution of the global regulator gene gacA to persistence and dissemination of Pseudomonas fluorescens biocontrol strain CHA0 introduced into soil microcosms.
Appl. Environ. Microbiol.
60:2553-2560[Abstract/Free Full Text].
|
| 48.
|
Natsch, A.,
C. Keel,
N. Hebecker,
E. Laasik, and G. Défago.
1997.
Influence of biocontrol strain Pseudomonas fluorescens CHA0 and its antibiotic overproducing derivative on the diversity of resident root colonizing pseudomonads.
FEMS Microbiol. Ecol.
23:341-352[CrossRef].
|
| 49.
|
Nuti, M. P.,
A. Squartini, and A. Giacomini.
1994.
European Community regulation for the use and release of genetically modified organisms (GMOs) in the environment, p. 165-173.
In
F. O'Gara, D. Dowling, and B. Boesten (ed.), Molecular ecology of rhizosphere microorganisms. VCH Publications, Weinheim, Germany.
|
| 50.
|
Podani, J.
1994.
Multivariate data analysis in ecology and systematics. Ecological computations series, vol. 6.
SPB Academic Publishing, The Hague, The Netherlands.
|
| 51.
|
Raaijmakers, J. M.,
M. Leeman,
M. M. P. van Oorschot,
I. van der Sluis,
B. Schippers, and P. A. H. M. Bakker.
1995.
Dose-response relationships in biological control of Fusarium wilt of radish by Pseudomonas spp.
Phytopathology
85:1075-1081.
|
| 52.
|
Raaijmakers, J. M.,
R. F. Bonsall, and D. M. Weller.
1999.
Effect of population density of Pseudomonas fluorescens on production of 2,4-diacetylphloroglucinol in the rhizosphere of wheat.
Phytopathology
89:470-475[Medline].
|
| 53.
|
Schnider, U.,
C. Keel,
C. Blumer,
J. Troxler,
G. Défago, and D. Haas.
1995.
Amplification of the housekeeping sigma factor in Pseudomonas fluorescens CHA0 enhances antibiotic production and improves biocontrol abilities.
J. Bacteriol.
177:5387-5392[Abstract/Free Full Text].
|
| 54.
|
Schwinn, F., and T. Staub.
1995.
Phenylamides and other fungicides against Oomycetes, p. 323-346.
In
H. Lyr (ed.), Modern selective fungicides. Properties, applications, modalities of action. Gustav Fischer Verlag, Iena, Germany.
|
| 55.
|
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[Abstract/Free Full Text].
|
| 56.
|
Sharifi-Tehrani, A.,
M. Zala,
A. Natsch,
Y. Moënne-Loccoz, and G. Défago.
1998.
Biocontrol of soil-borne fungal plant diseases by 2,4-diacetylphloroglucinol-producing fluorescent pseudomonads with different restriction profiles of amplified 16S rDNA.
Eur. J. Plant Pathol.
104:631-643[CrossRef].
|
| 57.
|
Stutz, E.,
G. Défago, and H. Kern.
1986.
Naturally occurring fluorescent pseudomonads involved in suppression of black root rot of tobacco.
Phytopathology
76:181-185.
|
| 58.
|
Sukarno, N.,
F. A. Smith,
S. E. Smith, and E. S. Scott.
1996.
The effects of fungicides on vesicular-arbuscular mycorrhizal symbiosis. 2. The effects on area of interface and efficiency of P uptake and transfer to plant.
New Phytol.
132:583-592[CrossRef].
|
| 59.
|
Thomas, M. R., and R. C. Shattock.
1986.
Filamentous fungal associations in the phylloplane of Lolium perenne.
Trans. Br. Mycol. Soc.
87:255-268.
|
| 60.
|
Thomashow, L. S., and D. M. Weller.
1996.
Current concepts in the use of introduced bacteria for biological disease control: mechanisms and antifungal metabolites, p. 187-235.
In
G. Stacey, and N. T. Keen (ed.), Plant-microbe interactions, vol. 1. Chapman and Hall, New York, N.Y.
|
| 61.
|
Tiedje, J. M.,
R. K. Colwell,
Y. L. Grossman,
R. E. Hodson,
R. E. Lenski,
R. N. Mack, and P. J. Regal.
1989.
The planned introduction of genetically engineered organisms: Ecological considerations and recommendations.
Ecology
70:298-315[CrossRef].
|
| 62.
|
Troxler, J.,
M. Zala,
Y. Moënne-Loccoz,
C. Keel, and G. Défago.
1997.
Predominance of nonculturable cells of the biocontrol strain Pseudomonas fluorescens CHA0 in the surface horizon of large outdoor lysimeters.
Appl. Environ. Microbiol.
63:3776-3782[Abstract].
|
| 63.
|
Voisard, C.,
C. Bull,
C. Keel,
J. Laville,
M. Maurhofer,
U. Schnider,
G. Défago, and D. Haas.
1994.
Biocontrol of root diseases by Pseudomonas fluorescens CHA0: current concepts and experimental approaches, p. 67-89.
In
F. O'Gara, D. Dowling, and B. Boesten (ed.), Molecular ecology of rhizosphere microorganisms. VHC Publishers, Weinheim, Germany.
|
| 64.
|
von Arx, J. A.
1980.
The genera of fungi sporulating in pure culture, 3rd ed.
J. Cramer, Vaduz, Germany.
|
| 65.
|
Whipps, J. M.,
F. A. A. M. de Leij,
J. M. Lynch, and M. J. Bailey.
1996.
Impact of genetically modified microorganisms on the terrestrial microbiota including fungi, p. 299-316.
In
J. C. Frankland, N. Magan, and G. M. Gadd (ed.), Fungi and environmental change. Cambridge University Press, Cambridge, United Kingdom.
|
| 66.
|
Widden, P.
1986.
Microfungal community structure from forest soils in southern Quebec, using discriminant function and factor analysis.
Can. J. Bot.
64:1402-1412[CrossRef].
|
| 67.
|
Widden, P.
1986.
Seasonality of forest soil microfungi in southern Quebec.
Can. J. Bot.
64:1413-1423.
|
| 68.
|
Widden, P.
1987.
Fungal communities in soils along an elevation gradient in northern England.
Mycologia
79:298-309[CrossRef].
|
| 69.
|
Wüthrich, B., and G. Défago.
1990.
Suppression of wheat take-all and black root rot of tobacco by Pseudomonas fluorescens strain CHA0: results of field and pot experiments, p. 17-22.
In
C. Keel, B. Koller, and G. Défago (ed.), Plant growth-promoting rhizobacteria. Interlaken International Workshop, Interlaken, Switzerland.
|
| 70.
|
Yuen, G. Y., and M. N. Schroth.
1986.
Interactions of Pseudomonas fluorescens strain E6 with ornamental plants and its effect on the composition of root-colonizing microflora.
Phytopathology
76:176-180.
|
| 71.
|
Zak, J. C.
1992.
Response of soil fungal communities to disturbance, p. 403-425.
In
G. C. Carroll, and D. T. Wicklow (ed.), The fungal community: its organization and role in the ecosystem, 2nd ed. Marcel Dekker, Inc, New York, N.Y.
|
| 72.
|
Zak, J. C., and S. C. Rabatin.
1997.
Organization and description of fungal communities, p. 33-46.
In
D. T. Wicklow, and B. Söderström (ed.), The mycota, vol. 4. Environmental and microbial relationships. Springer-Verlag, Berlin, Germany.
|
Applied and Environmental Microbiology, April 2001, p. 1851-1864, Vol. 67, No. 4
0099-2240/01/$04.00+0 DOI: 10.1128/AEM.67.4.1851-1864.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Jousset, A., Lara, E., Wall, L. G., Valverde, C.
(2006). Secondary Metabolites Help Biocontrol Strain Pseudomonas fluorescens CHA0 To Escape Protozoan Grazing.. Appl. Environ. Microbiol.
72: 7083-7090
[Abstract]
[Full Text]
Top
Abstract
Introduction
